Vous êtes sur la page 1sur 20

Language & Communication 22 (2002) 437456

www.elsevier.com/locate/langcom

The problem of context in social and cultural


anthropology
R.M. Dilley*
Department of Social Anthropology, University of St Andrews, St Andrews,
Fife KY19 9SA, Scotland, UK

Abstract
This article addresses the problem of the construction of context as a key analytical concept
in the methodology of social and cultural anthropology. It takes a developmental view,
showing how the problem has been re-dened over time. It also adopts a transdisciplinary
approach plotting the development of the problem under dierent disciplinary conditions. It
argues that context is linked to interpretation in terms of connection and disconnection of
phenomena construed as relevant or not. It also argues that context is a social construct: it has
a social life, and this life is susceptible to anthropological analysis.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Context; Interpretation; Social and cultural anthropology

1. Introduction

The problem of context has raised its head in a number of academic disciplines
over recent years. Although the problem may appear in dierent guises in each of
these separate disciplines, its various manifestations can be shown to be interlinked
in numerous and complex ways. The concept of context has a very broad currency
(in social anthropology, linguistics, philosophy, history and so forth), and its forms
of exchange are varied across the boundaries of various disciplines. This article
attempts to plot something of the signicance of the concept of context within the
eld of social and cultural anthropology. Its aim is to illuminate aspects of the
problem of context that have been apprehended recently by anthropologists; it also

* Fax: +44-1334-462985.
E-mail address: rmd@st-andrews.ac.uk (R.M. Dilley).

0271-5309/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0271-5309(02)00019-8
438 R.M. Dilley / Language & Communication 22 (2002) 437456

aims to point to some of the cross-border exchanges of the concept of context across
the boundaries of proximate disciplines. In social anthropology, contexts can be
economic, political, religious and so on (see Morris, 1999), or they can be rural
or urban, or even be found at the micro-level of the village, or the wider level of the
whole society, the nation-state or the world system. Contexts indicate levels of
analysis as well as the substantive focus of analyses. The concept of context has
moreover been passed fruitfully back and forth between social anthropology and
linguistics, for example. The context of the use of words and the social situation of
interlocutors has been increasingly rened. Rates of exchange and the direction of
exchanges across these boundaries have uctuated over the recent history of the
concept, and it is the transdisciplinary perspectives on context that yield some of the
most telling insights into the problem. Indeed, this present collection of articles in
this journal is a ne example of comparative transdisciplinary analyses that are
required to make sense of the problem in all its subtlety.
Context is one of the central concepts of social and cultural anthropology; indeed,
stress on context in interpretation is one of its distinguishing features; and it is relied
upon as an indispensable part of anthropological method. One of its earliest uses in
social anthropology from the 1920s onwards was by Malinowski who used it in a
counter-argument to Sir James Frazers evolutionist method which compared and
contrasted isolated social institutions from across the globe in order to illustrate
levels of civilisation. Malinowski and other functionalists objected strongly on the
grounds that social institutions should not be abstracted from the matrix of social
relations of which they formed a part and in which they performed a specic func-
tion. Despite the importance of context to the disciplineand the fact that it has
been so central for so longit is surprising to nd how little attention has been
given to the topic over the history of the discipline. Of late, a number of key pub-
lications have emerged which have begun to address this central issue of methodo-
logical and theoretical importance.1
Ethnographic interpretation by social and cultural anthropologists relies on a
sense of context. They should appeal, conventional wisdom has it, to features and
characteristics surrounding a phenomenon in order to illuminate it and to under-
stand or give sense to it. The idea is that anthropologists who interpret social and
cultural phenomena do so with reference, therefore, to something called context.
This process of contextualisation has had an all too self-evident ring to it; it is a
process that has been seen as unproblematic. For Radclie-Brown, for example, a
structural-functionalist anthropologist, the relevant context for a religious ritual was
the set of local social relationships connecting the ritual to other social institutions.
In this context, it was the function the ritual played in maintaining local social
cohesion and order that became the central explanatory method. While shifts in
theory led to criticisms of earlier ideassuch as Radclie-Browns, these shifts
involved the redenition of context by shifting the process of selection of relevant
surroundings. New connections and disconnections were made. A religious ritual
might now become the locus of human experience that dealt with existential

1
See in particular Hobart (1982, 1985, 1986a,b, 1999), Strathern (1987, 1995), and Dilley (1999).
R.M. Dilley / Language & Communication 22 (2002) 437456 439

questions rather than social structural functions. Phenomena are illuminated,


therefore, by appeal to their surroundings; but the problem is that these surround-
ings themselves are selected and interpreted in dierent ways. The apparently simple
notion that it is contextualisation and the invocation of context that give form to
our interpretations raises important questions about context denition: What is a
context? How it is dened and selected, and by whom? These dierences are relevant
to what is seen as problematic and to what counts as explanation.
Another key term of social and cultural anthropological methodology is, there-
fore, mutually implicated in this investigation of contextnamely, interpretation.
One way in which I have chosen to frame the problem of context and its related term
interpretation is to think of them both in terms of connection. To interpret is to
make a connection. Context too involves making connections and, by implication,
disconnections. Frazer might have disconnected an institution from its local sur-
roundings, as Malinowski complained of, but he also connected it in a comparative
analysis with other like institutions from elsewhere in the world. One mans con-
nection is another mans disconnection.
An object is set in context, connected by relations to its relevant surroundings.
Such a view, as I attempt to show below, can be contrasted with those conventional
perspectives in social and cultural anthropology that frequently treat context as a
given or a self-evident constructthat is, it is not questioned at the beginning of
the analysis but simply taken o-the-peg, as it were. To start from the recognition
that context is itself problematic, and is the result of prior interpretation, is to begin
to recognise the social nature of the problem. Akman (2000) has recently recognised
the need in the eld of Articial Intelligence to study contexts as social constructs,
a conclusion reached after exhausting the limits of what can be said about for-
malized contexts. A similar shift has occurred in social and cultural anthropology,
as I will outline below, from a position that regarded context as a real and positive
phenomenon to be isolated and formally described (what Fabian, 1999, calls the
positivity of context) to a view that context is generated and negotiated in the
course of social interaction and exchange. This would suggest another angle on the
issue, namely that there is a social life to the problem of context.2
There are at least two aspects to this social life of context. The rst is to reect
upon the way in which dierent disciplines involved in the interpretation of phe-
nomena construe contexts in dierent ways. Through comparative inter-disciplinary
or transdisciplinary perspectives, the procedures of context denition can be
thrown into relief. Comparison in an anthropological perspective can be achieved,
furthermore, not only by considering other disciplinary practices, but also by
attending to the interpretative practices of people situated in dierent places and in
dierent cultures. We as analysts interpret phenomena by placing them in context;
local actors too conduct a similar process. The use of context can be compared
across disciplinary boundaries and across cultural boundaries. The rst aspect of the
social life of context is comparative; it is concerned to trace some of the movements

2
This idea of a social life of things and concepts derives from Appadurai (1986), who calls the
approach a form of methodological fetishism.
440 R.M. Dilley / Language & Communication 22 (2002) 437456

and transformations that context has undergone in the process of a biography of


displacement from one discipline to another. It views analysts as social actors caught
in the act of contextualising.
The second aspect of the social life of context involves a life of connections that
is made in the interpretative processes of any specic instance of context denition.
The act of interpreting has been described as the act of creating connections; that to
interpret is to make a connection (Ormiston and Schrift, 1990). Context too involves
making connections and, by implication, disconnections. A phenomenon is con-
nected to its surroundings: contexts are sets of connections construed as relevant to
someone, to something or to a particular problem, and this process yields an expla-
nation, a sense, an interpretation for the object so connected. The context or frame
also creates a disjunction between the object of interest and its surroundings on the
one hand, and those features which are excluded and deemed as irrelevant on the
other. Contexts connect and they disconnect. This second aspect of the life of con-
text might be dubbed contexts by connection, whereas the rst might be dubbed
contexts in comparison. The emphasis here is upon the way in which analysts as
well as ordinary people are caught or framed in the process of making or construing
contexts, interpretations and meanings for themselves. Their contextualising
moves, or the ways in which their subject matter is ordered, connected and dis-
connected, is part of the life of context. Contextualising is a form of social action,
and a very particular form too. It is rst and foremost a discursive strategy, a mode
of articulation that can be detailed within specic analyses and particular disciplines.
If dierent disciplines dene context in dierent ways, this subsequently leads to
intellectual debate, the life of academe. If, however, social anthropologists as out-
sider analysts dene contexts in ways dierent from local social actors and inter-
preters, then should we privilege one over the other? Or should we look for a more
general context that can be used to help interpret both sets of contextualising
moves? Contextualisation is, then, a social practice which has a performative char-
acter to it. If this is true, it begs another question: If the relevant context for a par-
ticular interpretation can appear to be given or self-evident, then what analytical
and even political strategies are adopted in order to suggest this misrepresentation
or act of bad faith?3 That is, if contexts are actually constructed yet are made to
appear as if they are natural objects or taken-for-granted objects, then how is this
process of misrepresentation achieved? This is the politics of context denition.
An examination of the problem of context with respect to social and cultural
anthropology demands attention to two related sets of issues. The rst aspect takes
context as part of a set of methodological and epistemological problems in anthro-
pology itself. Analysts claim to place phenomena in context; sometimes they are
accused of taking things out of context. What is entailed in these analytical pro-
cesses? How do we make the distinction between being in and out of context? In
short, how have social and cultural anthropologists construed as relevant the

3
Misrepresentation and bad faith are terms taken from Bourdieu (although the latter is borrowed
from Sartre) to indicate the way in which symbolic elaborations come to mask or misrepresent the actual
nature of things (Bourdieu, 1977). It is an act of bad faith to go along with such cultural misrepresentations.
R.M. Dilley / Language & Communication 22 (2002) 437456 441

contexts they deploy in their analyses? These are huge questions to address in an
article of limited length, and I will focus primarily on interpretative approaches
within social and cultural anthropology.
The second set of issues around the problem of context involves not only prob-
lematising anthropological methodology in itself but is related to the focus of
anthropological endeavour: local, indigenous social and cultural practice of peoples
throughout the world. That is, we must attend to and examine context construction
in terms of an appreciation of the acts of interpretation and contextualisation that
those whom we study claim for themselves. The goal within interpretative anthro-
pology has been to seek denitive native meanings for social practices, symbols,
concepts and so on. In the process, however, it has perhaps for too long super-
imposed our own conceptions of what interpretation is about and what the appro-
priate contexts of indigenous practice might be. It has thereby overlooked how
contexts might be construed by local social agents with respect to the denitions,
negotiations and contestations of meaning within situated contextualising practices.
One must enter a caveat here: there is probably no short cut to denitive native
meanings by means of some new magic which the anthropological contextualiser
might weave with a superior wand of context. Instead, it suggests that they, like us,
are involved at various times in the active pursuit of meanings and the drawing of
relevant contexts for their explanations and interpretations. These two sets of pro-
cesses demand comparison and contrast.
Both analyst and local subject are, therefore, potentially caught in the act of con-
textualising: the rst in relation to the interpretative practices of an academic discipline;
the second in relation to the practical and explanatory concerns of a form of social
life. These two related aspects to the problem of context can be subsumed under the
gloss of the knowledge of context. Implied here is a query about how we come to
know about the contexts we use in a variety of forms of social and cultural practice.
There is, however, another dimension to this query. Our own anthropological
knowledge is obviously implicated in the construction of those contexts we regard as
relevant to particular problemsthese problems are themselves a function of our
knowledge systems. The construction of contexts is therefore intimately connected
with how we conceive of knowledge, and this fact must therefore have consequences
for how we conceive of what contexts might or might not be appropriate for the
analysis of other peoples practices. If the idea holds that contextualising is a very
specic form of social practiceas a discursive, expressive and performative type
then the kinds of knowledge claims behind this type of practice need inspecting. This
aspect of the problem can be glossed as the issue of contexts of knowledge. Indeed,
it raises the question of whether there can be, as some contemporary theorists sug-
gest, an a-contextual basis for anthropological knowledge about other people.

2. A brief history of a word: context

The word context is derived from the Latin verb texere, to weave, and the related
Latin verb contexere carries the meaning of to weave together, to interweave,
442 R.M. Dilley / Language & Communication 22 (2002) 437456

to join together or to compose (The Shorter Oxford English Dictionary).4


Obsolete meanings of the word range from the weaving together of words and
sentences, and literary composition; to the connection or coherence between parts
of a discourse. More contemporary meanings suggest the parts which immediately
precede or follow a particular passage or text and determine its meaning. Also,
gurative uses refer to a type of connection or relationship of a general kind as in
the circumstances relevant to something under consideration or in this connec-
tion.
The history of usage of context suggests a shift in reference of the term from the
act of composing meaningful stretches of language either as speech or writing (as in
to contex [sic] a history, O.E.D.) to the conditions for understanding a stretch of
language and of the possibility of determining its meaning. It initially denoted the
act of composition, of bringing together parts of language into meaningful utter-
ances or written texts. It later took on the sense of the conditions under which
meaning is attributed to a stretch of language, and indeed how those conditions give
sense to it. From describing the act of conjoining, the term then comes to designate
the conditions shaping that which has been conjoined. A similar sense of involution
in the concept of context in social and cultural anthropology will be illustrated
below. The idea of culture has become the primary context for many social and
cultural anthropologists. The conception of culture, however, has been transformed
from an idea that comprises what people say, do and think to the context for the
interpretation of what people say, do and think. This shift in meaning also indicates
aspects of the relationship between knowledge and context: in the rst instance,
context is an item of knowledge; in the second it is a condition which shapes
knowledge.

3. Transdisciplinary perspectives on the problem of contextviews from philosophy,


linguistics and literary studies

I attempt in this section to review a series of views on context from a selection of


disciplines. This review is neither complete nor exhaustive, but illustrates the sig-
nicance of a transdisciplinary exercise. I attempt to achieve a movement back and
forth across disciplinary boundaries in order to highlight the cross-fertilisation that
has produced the array of versions of the concept of context. What follows is an
exercise in plotting the life of context in its many guises, and how this life is com-
posed of dierent forms of connection made within dierent disciplines. Having
completed this task I will then move on to consider the problem with respect to
social and cultural anthropology, in which similar kinds of pattern will be seen to
emerge.
Work on philosophical perspectives on the problem of context are most obviously
dealt with in two works in particular: B.-A. Scharfsteins The Dilemma of Context

4
All these denitions are taken from The Shorter Oxford English Dictionary; see also Holbart (1985,
1986a) for a parallel treatment.
R.M. Dilley / Language & Communication 22 (2002) 437456 443

(Scharfstein, 1989) and Gayle Ormiston and Alan Schrifts collection Transforming
the Hermeneutic Context (Ormiston and Schrift, 1990). The rst highlights the
inherent problems in deploying the concept of context in analyses; the second plots
the transformations that have occurred in the treatment of context in the work of a
range of hermeneutic philosophers.
Scharfstein states that the problem of context is too dicult for philosophers or
anyone else to solve (Scharfstein, 1989, p. 4). The problem as he sees it is that it
lays an intellectual burden on us that we cannot evade but that can become so
heavy that it destroys the understanding it is meant to further (Scharfstein, 1989,
p. xi). He highlights the connection between contextualism and other forms of rela-
tivism: Dependence on context is a kind of limited relativism, and that relativism,
looked at philosophically, is hard to limit (Scharfstein, 1989, p. 59). However, con-
textualism/relativism is only one form of philosophical analytical strategy, and
Scharfstein makes the point that as a form of philosophising, contextualism as a
type of relativism is never far apart from universalism, another form. Where some-
one is invoking context, it would seem, someone else is proposing universals, essen-
ces, formal principles of general applicability and the like. He argues that relativism
and absolutism are equally essential and mutually necessary, in that concepts
that appear to be absolute depend on relative notions for conceptual contrast in
terms of their meaning as absolutes, and likewise, relative concepts depend on
absolute ones for their sense (Scharfstein, 1989, p. 131). A moral absolutism, for
example, that proposes a universal standard of good and bad depends for its
meaning on a relation to a form of moral relativism that suggests that this standard
may vary from place to place. One mutually entails the other. Relatedly, where the
moral absolute thou shalt not kill is invoked, then not far away are those who argue
for the consideration of mitigating circumstances or for the relevance of context, such
as that of war, self-defence and so on that modies the impact of the absolute.
The dilemma of context for Scharfstein is that persistent contextualism leads to
extreme relativity, which in his view is insupportable as a philosophical position. He
states: the attempt to be thorough in understanding context leads to a total con-
textualisation, in which everything becomes the context of everything else. Such a
contextualisation is equivalent to total relativity (Scharfstein, 1989, pp. xiixiii).
This is what he refers to as the diculty in limiting context once it is admitted into
an analysis, and is similar to what Culler, a literary critic, has called the unbound-
edness of context (Culler, 1983). For Scharfstein, if there is total relativity, every-
thing is individual and unique and no comparison is possible; for Culler, the
unboundedness of context refers to the diculty of limiting a context in relation to a
particular text, and so meaning becomes elusive.
The dilemma of context, which Scharfstein points to, can also be read as the
dilemma of social and cultural anthropology caught between the Scylla of con-
textual relativism and the Charybdis of extreme sameness and objectivity. The
intellectual dilemma at the heart of the discipline involves one in which anthro-
pologists have conventionally claimed priority for one of two mutually opposed
analytical possibilities: rst, that it should explore the essential unity of human
beings (the psychic unity of mankind, or the embodiment of universal mind or
444 R.M. Dilley / Language & Communication 22 (2002) 437456

meaning-making or rational calculation); second, it should explore the uniqueness of


each culture and society in its own terms, leading to a kind of relativism. The mutual
necessity and implication of relativism and universalism creates a theoretical dialec-
tic that is the motor of disciplinary debate. While pointing out the dilemmas asso-
ciated with context, Scharfstein does nonetheless argue for the impossibility of
absolute truth; indeed, he suggests that since no external source of adjudication can
be brought in to give one vision priority over the other, he personally nds a greater
sense of adventure and exhilaration in the pursuit of contextualism (Scharfstein,
1989, p. 192). This is reminiscent of Hobarts dictum that anthropology is our one
chance of escaping the sheer tedium of our own thought (Hobart, 1982, p. 58).
What needs to be emphasised is that, from our own philosophical background,
contextualism and its contraries are linked in a discourse of mutual implication,
such that to propose one presupposes the other. It is an empirical question open to
investigation whether they are or are not linked in other philosophical traditions.
The dilemma of context for Culler is seen in terms of the indeterminacy of mean-
ing. Context in the form of the concept of the hermeneutic circlea relation of part
to wholehas remained central to the project of hermeneutics. Palmer gives the
following denition of it: We understand the meaning of an individual word by
seeing it in reference to the whole sentence; and reciprocally, the sentences meaning
as a whole is dependent on the meaning of the individual words (Palmer, 1969,
p. 87). He then broadens its scope: By extension, an individual concept derives its
meaning from the context or horizon within which it stands; yet the horizon is made
up of the very elements to which it gives meaning (Palmer, 1969, p. 87).
The hermeneutic circle points to the circularity of all human understanding and to
the problems that arise in the process of interpretation. Indeed, this theme is echoed
in the earlier statement made above that interpretation in context requires the pre-
interpretation of the relevant context, that in turn informs the subsequent inter-
pretation. There is an inherent circularity of interpretative procedures that rely on
context. In art history, for example, a painting might be viewed on the one hand as
the product of a history of the development of styles, methods and practical tech-
niques, or on the other in a relation of comparison and contrast with other artists
within a contemporary period. In the rst, the relevant context is dened historically
or diachronically, in the second it is dened synchronically. The pre-interpretation
of which context is to be the focus of analysis then predetermines what kind of
information about a painting is relevant. The pre-analytical choice of context con-
rms that context and the type of connection it entrains. Another example comes
from social anthropology, which saw in the 1970s and 1980s a shift toward a
semantic anthropology. According to this approach humans were conceived as
meaning-makers rather than say the maximisers of value, and connections made
during the analysis duly showed that humans were indeed meaning-makers. Ana-
lyses were again conducted on the basis of a pre-interpreted context of relevance,
and the success of an analysis was determined by the extent to which it revealed that
pre-interpreted context.
Ormiston and Schrift (1990) move the debate on further by reviewing the work of
numerous hermeneuts. Context is no longer something that simply illuminates the
R.M. Dilley / Language & Communication 22 (2002) 437456 445

text or vice versa. They talk of a traditional conception of the hermeneutic context
that has now been replaced by a post-modern sense of context. The traditional
conception was engaged in the pursuit of the primordial meaning of hermeneutics,
that is to say, as a way to present the meaning of the question of being (Ormiston
and Schrift, 1990, p. 33); whereas a post-modern sense can be seen in the work of
Derrida, who argues that texts refer to nothing outside of themselves but only to
other texts. The transformations that have occurred within the hermeneutic context
result in a set of consequences for social and cultural anthropology, to be reviewed
below.
The reliance on context in linguistics mirrors its usage in philosophy, namely that
it sets up a counter movement to those theories and perspectives which view lan-
guage as a formal system, or which develop context-free grammars or structures of
meaning. Contextualist linguistics can be contrasted to theoretical linguistics whose
highly abstracted and idealised view of language divorces it from the conditions in
which it is actually used. In this vein, Dell Hymess work (e.g. Hymes, 1977) chal-
lenges Chomskyan linguistics, which attempts to dene an autonomous realm of
language as an abstract cognitive system to be studied in isolation from the contexts
of language use. Context also bears the burden of multiple interpretations: while it
can refer to parts of an utterance surrounding a linguistic unit that may aect both
its meaning and its grammatical contribution, context can also embrace the wider
situation of either a speaker or of the accompanying activities and social situation.
Language in this view is a form of social action to be viewed alongside other social
activities.
As an aspect of the speakers situation or as an aspect of the surrounding social
situation, context in linguistics covers a broad range of social phenomena as well as
actor-oriented characteristics. Duranti and Goodwin (1992) spell out a number of
positions taken by linguists over the question of context of linguistic performance.
They highlight four basic parameters of context: (i) the settinga social and spatial
framework within which encounters are situated; (ii) the behavioural environment
the use of bodies or behaviour for framing talk; (iii) language as context the way
talk itself invokes context and provides context for other talk; and (iv) the extra-
situational context or background knowledge and frames of relevance (Duranti and
Goodwin, 1992, pp. 68). These parameters mirror to some extent those that can be
detected in anthropological use, although some commentators would wish to add
historical context as a major consideration.
Duranti and Goodwin themselves argue for an interactionist view of context
which places the social person on centre stage. This reects not only the view of
language as context, but more generally of interaction itself as context. Their theme
is the capacity of human beings to dynamically reshape the context that provides
organisation for their actions within the interaction itself (Duranti and Goodwin,
1992, p. 5). Context is both constitutive of social action and itself the outcome of
social action; it is both a generative principle and a resulting outcome. The denition
and deployment of context is also part of actors strategies wherein individual
participants can actively attempt to shape context in ways that further their own
interests (Duranti and Goodwin, 1992, p. 6). Context is thus analysed as an
446 R.M. Dilley / Language & Communication 22 (2002) 437456

interactively constituted mode of praxis, and their focus on interaction links their
approach to ethnomethodologys concern with face-to-face exchange as a primary
exemplar of [a type of] context that is dynamic, and time-bound (Duranti and
Goodwin, 1992, p. 22).
Duranti and Goodwin note two areas neglected in their own work: one is social
and cultural anthropological interpretativism, and the second is the question of lit-
eracy and written texts. The salience of the idea of interactive context, as they deal
with it, is diminished with respect to literacy and writing in particular. These are two
of the main procedures, they argue, through which the ground within which lan-
guage emerges is systematically erased (Duranti and Goodwin, 1992, p. 32 fn 8).
Indeed, it is to literary theory that we must turn if we are to pursue these interests.
Structuralism and Saussurean linguistics throw up examples of what might be
called internal, closed-system contexts. Saussure saw meaning linked not only to
the relation between a signier and a signied, but also to the value of a sign or its
connection to other signs that might be substituted for it, or against which it stands
in contrast (Hervey, 1982). Structuralist developments based on Saussures insight
tended to discard the relation between signier and signied, and sought meaning
instead in terms of a closed system of signs. Hobart states a similar view for semio-
tics in general: meaning is no longer to be dened by either an external reality or
an external context (Hervey, 1982, p. 48), but shifts the problem to one of an
internal context.
Like structuralism and semiotics, branches of literary theory and criticism posit
internal contexts of a particular sort. Kristevas idea of intertextuality is a concept
that refers to how the meaning of a text is dependent upon a prior knowledge of
other texts, and not necessarily upon the texts reference to or representation of an
external reality. Culler points out (Culler, 1988) that literary criticism as a discipline
has been a net exporter of theoretical discourse in the 1980s, leading to develop-
ments in the concept of context. This has been achieved through the investigation of
the processes of signication and the production and play of meaning. The shift in
what the meaning of a text is all about refers to the problem of whether it is possible
to put a limit on context.
Derrida throws down a challenge to the proponents of speech act theory in terms
of a critique centred on the problem of context (Derrida, 1977). He attacks the idea
of the re-introduction of speakers intention as a means of controlling or arresting
the innite regress in the process of contextualisation, the contextualising of context
and so on. This is what Culler calls the unboundedness of context. Derrida focu-
ses on how this re-introduction of the presence of a signifying intention in the
consciousness of the speaker occurs. He declares: This is my starting point: no
meaning can be determined out of context, but no context permits saturation
(quoted in Culler, 1983, p. 123). Culler concludes that total context is unmasterable,
both in principle and practice. Meaning is context-bound but context is boundless
(Culler, 1983, p. 123). Context is boundless, he argues, in two senses: rst, there is no
limit to its contents, it is unsaturable and is always open to further description; second,
any denition of context can itself be contextualised by means of a new context, and
the process is open to innite regression. Reintroducing speakers intention is not a
R.M. Dilley / Language & Communication 22 (2002) 437456 447

solution, but a distraction from the problem of the unboundedness of context once
the concept has been admitted into the analysis.
If univocal meaning grounded in presence is not a viable option for Derrida,
then the armation of the free-play of meaning is. The deconstructive method is
identied with the twin principles of the contextual determination of meaning and
the innite extendibility of context (Culler, 1983, p. 215). If there is no limit to
context, there is no limit to the meanings that can be interpreted, and so interpreta-
tion becomes a game. Deconstruction is the turning of the text against itself in
order to investigate how a particular perspective is produced through the connec-
tions within texts. For Derrida il ny rien de hors texte, there is no outside-the-
text.5 The internal context of structuralism here shrinks to a minute singularity
which oers nothing outside of itself. Hobart captures this sense in his arresting
description of Derridas glum world where we shunt around forever in the prison
of our own metaphysic (Hobart, 1985, p. 44). These views have implications for
what is coined the literary turn in social and cultural anthropology, a development
dating from the late 1980s onwards. If there is no external world we can reliably
draw upon in interpreting and assessing anthropological texts, then the internal
contextualising of deconstruction leaves us only with text. Such a position has huge
implications for the utility and even possibility of anthropological eldwork.

4. Context in social and cultural anthropology

One of the fathers of the discipline, B. Malinowski, pioneered work on the prob-
lem of context in social anthropology in two publications in particular: The Problem
of Meaning in Primitive Languages in 1923 [1938], and Coral Gardens and Their
Magic in 1935. He laid the foundations not only for the way the problem was con-
ceived for generations of social and cultural anthropologists, but also inuenced
developments in other disciplines, particularly linguistics. These elaborations subse-
quently fed back into anthropology. Malinowskis work threw down a challenge to
an earlier generation of evolutionist anthropologists who proposed a series of spec-
ulative histories for forms of social organisation and for a range of social institu-
tions. In this vein was the work of Sir James Frazer, whose use of selected examples
of isolated and abstracted social institutions in the process of constructing universal
theories relevant to all of humankind, was seen as a form of anthropological enter-
prise that took things and facts out of context.
The rst of Malinowskis publications on the topic of context appeared in a
volume written by two philosophers, Ogden and Richards, on the meaning of
meaning. The authors confronted the claims of causal theories of reference which
posited meaning as the result of direct knowledge between minds and things.
Meaning, they argued instead, is to be linked to context. Malinowski coined the

5
Culler adds a critical dimension to the contextualising moves Derrida employs, by pointing out his
paradoxical use of privileging rst history to challenge philosophy, and then using philosophy against
history (Culler, 1983, p. 129).
448 R.M. Dilley / Language & Communication 22 (2002) 437456

phrase context of situation (a parallel to Ogden and Richards concept of sign-


situation, or the context of original, immediate and direct experience (Malinowski,
1938, p. 53) in order to address the pragmatic circumstances in which languageas
a mode of action and not a countersign to thoughtwas used and articulated. He
argued against the idea that a word contains a meaning as one of its essential fea-
tures: . . . the meaning of a word must always be gathered, not from passive con-
templation of this word, but from an analysis of its functions, with reference to the
given culture (Malinowski, 1935, p. 309). Only by referring a word to its given
culture, therefore, could its meaning be properly assigned.
The conception of culture had to be broadened, he argued, to include the
immediate and direct situation in which the word under investigation was used. In
his later publication he developed much more fully the stages through which con-
textualisation proceeds in the course of the translation of native terms and mean-
ings. The process starts with the context of words, a linguistic context referring to
the properties of language, its vocabulary as well as the structure of the utterance
and so on. The context of situation is now rephrased as the context of culture
such that translation takes place with reference to it and against a cultural back-
ground of a society (Malinowski, 1935, pp. 1718). He goes on to argue that . . . it
is very protable . . . to widen the concept of context so that it embraces not only
spoken words but facial expression, gesture, bodily activities, the whole group of
people present during an exchange of utterances and the part of the environment on
which these people are engaged. He refers to these features later as the context of
cultural reality, by which he means the natural equipment, the activities and
interests and aesthetic values with which the words are correlated (Malinowski,
1935, p. 22). It is well to bear in mind that this commentary was stimulated by his
reections on how to translate specic Kiriwinian words set within networks of
associations that might be simplistically glossed as, for example, garden or fam-
ine, etc. The twists and turns through which the concept of context has been
taken by all manner of hermeneuts who have reied the concept and frozen it as
text contrasts starkly with Malinowskis original context of use of the term. As
indicated in the discussion above on the etymology of the word context, Malinowskis
use of the phrase context of culture resembles the earlier sense of the term context as
elements that are brought together to aid the understanding of a stretch of language.
The hermeneutic twists given to the idea of culture as a primary interpretative context
moves away from Malinowskis sense and more towards the later denition of it as the
condition which gives rise to cultural meaning. I return to this issue below.
The linguist Firth develops Malinowskis ideas about context (see Hervey, 1999).
Another inheritor of this ne-grained examination of language in use is Hymes
work on the ethnography of speaking in which he develops Malinowskis concept of
context of situation. While Malinowski was one of the rst social anthropologists
to take seriously the problem of native language, its use and its translation, Hymes
was one of the rst linguists to elaborate non- and extra-linguistic frames of refer-
ence for the interpretation of speech events. An ethnographic project is now seen to
be part of linguistic analysis, such that a wide range of social and cultural features
are considered relevant to analysis (see Duranti and Goodwin, 1992).
R.M. Dilley / Language & Communication 22 (2002) 437456 449

The concept of culture became the dening feature of social anthropology after
Malinowski, and it came to represent that which is local, particular and distinctive,
compared to the global, general and common. For Malinowski, Trobriand society
as a set of local social relations was the context in which Trobrianders institutions
and social actions should be interpreted. The word context became part of a stock
anthropological vocabulary used to denote a bewildering variety of characteristics,
domains and environments. Contexts could be cultural, social, political, ritual and
religious, economic or ecological; they could be interactional, systemic or historical.
The term, it seems, is suciently elastic to be stretched in numerous directions for
diverse purposes.
One of the rst to express disquiet over the concept of context was Roger Keesing,
particularly in his 1972 publication on kinship. Here he pondered the need for a
formal theory of context denition, and for a grammar for creating contexts
(Keesing, 1972, p. 28). He recognised the positivity of context (Fabian, 1999) by
stating that they are in our heads, not out there (ibid.). They are not self-evident
aspects of reality that are pre-given or to be taken-for-granted, in the sense of being
understood as existing prior to analysis. They are part of the analysis and inter-
pretation itself. Holy later repeated these calls for anthropological attention to be
given to the subject (Holy, 1989).6
C. Geertz (1973) developed an interpretive approach within anthropology, and
he drew particularly heavily on literary and textual models. Geertzian developments
exposed the frailty of the anthropological grasp on the concept as conceived by
textual hermeneuts, whose increasing focus on text paradoxically left context unex-
amined.7 Social life was identied with a text, which could be read by the observing
anthropologist. There remains, however, an ambiguity in this interpretive approach
about what exactly a context is. If social life is the text, what is the context? In
Malinowskis hands, context was the multiple aspects of social life, against which
any particular piece of social action was set. Context comprised of the rest of the
pieces of the local jigsaw puzzle, against which one particular piece gains its place
and signicance. By contrast, texts appear everywhere in the Geertzian tradition
without obvious accompanying contexts in reference to which the analysts inter-
preted meanings take shape or gain prominence. Geertz describes various processes
of ethnographic interpretation in textual terms: (a) what the ethnographer does is
produce text, turning passing events into accounts, thus inscribing social discourse
(Geertz, 1973, p. 19); (b) the culture of a people is an ensemble of texts which the
anthropologist strains to read over the shoulders of others (Geertz, 1973, p. 452),
and this ensemble presumably comprises the individual texts of social activities and
the actions of individual agents; (c) culture is not just an ensemble of texts but an

6
Holy and Stuchlik raised versions of the problem of context in various writings, especially their 1981
and 1983 publications, in questions relating to the dierences between the analysts and native knowledge.
They attempted to pose a particular solution to the problem of native cultural context by dening the
concept of folk modela broad, common-sense background knowledge, intersubjectively shared and
often beyond the means of conscious articulation and reection by local actors.
7
Hobart over many years has sustained a series of searching critiques against the excesses of inter-
pretivism (see especially Hobart, 1982, 1985, 1986b, 1990, 1999).
450 R.M. Dilley / Language & Communication 22 (2002) 437456

integral document, a kind of script in itself (culture, this acted document; Geertz,
1973, p. 10); (d) culture also stands as a context that frames the meanings actors
themselves attach to their lives and experiences (culture . . . [is] the structure of
meaning through which men give shape to their experience; Geertz, 1973, p. 312).
This last denition reies the concept of culture to the status of a thing and then
identies it with an interpretative context, now referred to as the condition for all
Balinese social action, meaning and experience. Not only might Geertz be seen to
lose sight of the contextual wood for the textual trees to borrow Hobarts phrase
(Hobart, 1985, p. 46), but he neglects to examine the relationship between types of
text, how they may articulate one with another, and how they relate to context. If
the concept of culture as a reied object came to stand for many as the modernist
anthropological conception of context par excellence, then the certainty of both
master concepts has been shaken. In Geertzs hands what the Balinese said and did
were not simply aspects of Balinese culture, but they gained meaning only in relation
to a more reied concept of Balinese culture, the master context within which all
local activity was to be referred. Balinese cockghts and cocks took on meaning
because of the deep play of specic symbols within that culture.
Cliord highlights the idea that the culture concept has served its time, and he
argues recently for a shift in perspective from a xed, static and holistic notion of
culture to one focusing on travel as the locus of human experience (Cliord, 1997).
The modern world is characterised by movements of people, and by exchanges of
ideas and things across previously discreet cultural boundaries. He states: The old
conception of context as a xed framework within which social activities take place
becomes stretched and even shattered by means of movements of people between
intersecting contexts (Cliord, 1997, p. 81).8 He still nds it necessary to retain
some sets of relations which preserves the concepts dierential and relativistic
functions (quoted in Fardon, 1990, p. 11), that is precisely those functions ear-
marked as the task of context in challenging universalism, formalism and so on.
While the processes of globalisation might led to cultural homogeneity, most people
still by and large live in relatively small, situated social worlds from where they gain
their view of the world. However, the crisis about the certainty of modernist social
anthropological context has led to a post-modern mood, the attitude of which is to
make deliberate play with context in the sense of Derridas deconstruction (Stra-
thern, 1987, p. 265). The xed, stable cultural contexts of tribal groups (if they were
ever such) have given way to a more unbounded sense of what the relevant contexts
of social action might be. One consequence of this play is that the conventional
distinction between us and them, the distance between the object of interpretation
and the interpreter is discredited.9 The anthropological observer who interprets

8
What constitutes, in his view, the contexts that now intersect with each other requires further expla-
nation; he states nothing denite, leaving only scattered hints that they might, for example, be construed
in terms of world systems (Cliord, 1997, p. 276) or via Bourdieus concept of habitus (Cliord, 1997, p.
44).
9
Ormiston and Schrift (1990) make a similar point in their discussion of two conceptions of inter-
pretationthe Stoic and the Platonic. The former eliminates the gap between interpretation and object,
the latter generates and maintains it as part of its method.
R.M. Dilley / Language & Communication 22 (2002) 437456 451

exotic behaviour is him- or herself observed and interpreted by exotic others in the
act of observing and interpreting. Moreover, in this view, ethnography should
become a dialogue such that no one position or context of any particular agent is
privileged, and the reader is allowed a degree of free play, in a Derridean sense, in
reading meanings into the text. The anthropologist no longer has the privilege of
dening contexts and dening meanings for others. This should be done as part of a
dialogue between both parties. Strathern suggests, furthermore, that a new rela-
tionship between reader, writer and subject matter should be contemplated (Stra-
thern, 1987, p. 265), such that the reader is invited to interact with the text and
exotica itself. The shift from modernist context to a post-modern play with context
appears to be part of a movement from external context to an internal one in
which there is a deferral of immediate signication. That is, where the reader is
invited to interact with text, it is with a text and not with its sense of reference to or
correspondence with an external world. Harvey (1999) in her analysis of Expo 92
points to something similar. Visitors to the Exposition found meaning in the indivi-
dual national exhibitions not by referring them to the nation-states they purported
to represent, but to other exhibitions and theme-park type amusements they had
visited in the past.10
Critics of the conventional usage of context in social anthropology appear, how-
ever, to be equally reliant upon contextualisation, in the sense of giving order to
material, as do those they criticise. The nature of the context invoked, nonetheless, is
radically dierent. The new critics of ethnography have brought about a paradigm
shift in anthropology associated with a re-denition of relevant context. But the fear
is that the new knowledge generated by this shift is reduced to self-knowledge
(Strathern, 1987, p. 268). Hobart voices a similar concern when he accuses these
critics of smuggling in a number of essentialisms in their re-contextualisation of
anthropological enquiry, not least in viewing the nature of the evoking self [of the
ethnographer] as curiously unproblematic (Hobart, 1990, p. 311).
Taussig heralds the collapse, even the irrelevance, of conventional context in the
following passage:

One could, I suppose, do the usual thing and analyze them [Cuna gurines] as
anthropologists do, as things to be layered with context, then stripped, but that
would seem to be an evasion and miss the pointthat something crucial about
what made oneself was implicated and imperilled in the object of study, in its
power to change reality, no less (Taussig, 1993, pp. 252253).

10
This is a form of play with context, as described by Harvey (1999). The exhibitors played with the
idea of context in their national presentations, and visitors framed their experiences not simply by refer-
ence to the features of the exhibition in Spain, but sought contrasts and comparisons within their own
experience from similar types of exhibition or leisure-world setting in other places. Their interpretations
referred to previous interpretations of parallel experiences elsewhere. This suggests the idea of a series of
self-referential signiers and the idea of a type of inter-textuality; or the idea expressed by Derrida that
everything begins by referring back . . . to other traces and the traces of others (quoted in Ormiston and
Schrift, 1990, p. 25).
452 R.M. Dilley / Language & Communication 22 (2002) 437456

It is the relationship between the self and the object of study that provides the
frame for analysis, particularly the way in which it gives on to self-knowledge.
Taussig states again, more specically:

I know next to nothing of the context of ritual, belief, or of social practice in


which an older anthropology, eager for the natives point of view, would
enmesh this African [Igbo representation of a] white man, explain him (away),
Africanize him (as opposed to whitenize him). All I have is the image and its
brief caption [from a 1967 publication by Cole], and I am my own gaping sub-
ject of analysis. . . (Taussig, 1993, p. 238).

He does not explain why self-knowledgethat he is his own gaping subject of


analysisis any less problematic than the natives point of view, which might, one
could well imagine, be a good deal more interesting and engaging. A view similar to
Taussigs is echoed in Marcus and Cushmans assertion that through the writers
[ethnographers] self-reection as a narrative vehicle, a reader will be able to iden-
tify more readily with anthropological eldwork, an identication also claimed to be
facilitated by a shift from the classic us-them didactic form to the me-them form of
contrast (Marcus and Cushman, 1982). Again, Marcus and Fischer argue that:
Anthropology is not the mindless collection of the exotic, but the use of cultural
richness for self-reection and self-growth (Marcus and Fischer, 1986, pp. ixx).
The literary turn in anthropology, stimulated in large part by the Derridean
notions of the free-play of meaning and the boundlessness of context, ends up pro-
posing a kind of project that Derrida would not himself want to identify with. That
is, the project revolves around a set of contextualising moves that invoke an essential
self that becomes the centre piece of anthropological investigation. (Recall Derridas
criticism of speech act theory that ultimately relied upon speakers intentions to limit
the notion of unbounded context.) It is one thing to argue that our selves should
become research objects alongside other selves rendered as objects in our enquiry
(Crick, 1992); it is another to dene anthropology in terms of self-growth.

5. Conclusions

If the preceding section helps us gain something of a knowledge of context within


the realm of social and cultural anthropology, then it will also be apparent by now
that there is another aspect to the relationship between context and knowledge. That
is, dierent constructions of the concept of context rest upon dierent bodies or
forms of knowledge. The epistemological foundations of twentieth-century anthro-
pology has been the contextualisation of knowledge (Strathern, 1995, p. 3); and the
way in which, through this process of contextualisation, context itself was made one
of its objects of knowledge (Strathern, 1995, p. 160; 1987, p. 276). Native knowledge
was conventionally contextualised in terms of integrated local cultures, and anthro-
pological knowledge was set against the background of western ethnocentrism and
the Malinowskian-inspired vision of the detection of civilisation under savagery
R.M. Dilley / Language & Communication 22 (2002) 437456 453

(Strathern, 1987, p. 256). Strathern makes the point (Strathern, 1995) that when
people shift contexts they thus make knowledge for themselves. Context is then
intimately connected to knowledge. The semantic turn in British social anthro-
pology in the 1980s was a crucially reexive moment for the consideration of the
parallels between the interpretative, meaning-making processes of anthropologists
and those of whom anthropologists study (see, for example, Parkin, 1982). A greater
transparency was given to how our interpretative frameworks were constructed and
how we had the power to describe and control contexts (Parkin, 1982, pp. xlvi
xlvii). This emphasis opened up for examination questions about forms of knowl-
edge, and alerted us to the distinction between our own contextualising moves as
analysts and the types of moves native interpreters might make. Holy and Stuchlik
(Holy and Stuchlik, 1981, 1983) suggested that there was a problem of context in the
relative treatment of bodies of native knowledge and of anthropological analytical
knowledge respectively. The problem of context is therefore duplicated: how has
anthropology as an interpretative discipline composed its own analytical contexts
that give shape to its explanations and interpretations? How and in what ways do
those people whom we study invoke context as part of their own local practices of
attaching meaning to the processes and events in social life? These questions give on
to the problem of commensurability between the two domains of interpretation and
contextualisation. Hobart (1999) talks of the double account of knowledge
involved in anthropological interpretation, and this involves depicting native
knowledge as distinct from anthropological knowledge, and then giving an account
of the former in terms of the latter.
The investigation of the problem of context necessitates an appreciation of the
orders of knowledge that provide the conditions for its emergence. Paradigm shifts
in anthropology involve shifts in orders of knowledge and shifts in the correspond-
ing concept of what a relevant context is. But the anthropological enterprise also
involves a double account of knowledge, an articulation of anthropological knowl-
edge with local indigenous forms of knowledge. This double account of knowledge
has to be kept in balance, such that anthropological exegeses do not skew indigen-
ous interpretations, or vice versa.
There is, therefore, the issue of power that underlies the ability to represent one
form of knowledge and set of contextualising practices in terms of another form of
knowledge and contextualising practices. The issue of power also underlies the pro-
cesses of context denition in the dynamics of social life. The ability to dene a
context in a particular way or to initiate a series of contextualising moves in a par-
ticular direction can be construed as a political act in the light of the other possible
denitions or moves that could have been made. Are the individuals captured in
Afghanistan and now held at a US base in Cuba to be identied as prisoners in the
context of war or as criminals in the context of criminal law? And the well-worn
issue of whether someone is a terrorist or a freedom ghter involves a subtle but
signicant shift in the denition of relevant context. To contextualise is to frame in
Cullers vocabulary. And this sense of frame highlights the act of drawing a line, of
excluding as much as including thingsof connecting and disconnectingwithin a
set of limits. This process of inclusion and exclusion is a process of power. It is our
454 R.M. Dilley / Language & Communication 22 (2002) 437456

sense of relevance, driven by our theoretical outlooks and practical dispositions


towards the world, that denes where these frames are to be placed. Moreover, to
sharpen our own sense of the way we fabricate contexts in the processes of our own
analyses might help us to become aware, in turn, of the interpretative practices and
contextualising moves used by others situated elsewhere and outside the academy.
The task of anthropology is to investigate the conditions of possibility of context
within not only our own bodies of knowledge but also the bodies of knowledge of
native interpreters and commentators. The best that anthropological accounts can
hope to do is to bring these two sets of conditions together into a relationship that is
dialogic. Context has been shown to be an emergent as well as a generative property
of knowledge. Indeed, contexts are sets of relations and not self-evident things in
themselves. We must therefore be alive to the possibility that there are two parallel
processes of construing context: for us from within our own bodies of knowledge;
and for them within theirs. The conjunction of these parallel processes in the course
of eldwork or in our writing about the eld and its subsequent dissemination to
other readers may generate further contexts of knowledge through a dialogical
relationship.
Anthropologists cannot kid themselves into believing that there is a position they
can readily adopt that does not rely on context or a frame of reference. They cannot
escape from the connes of context. But context is expandable, innitely so; and we
must never lose sight of the fact that a claim about context is precisely thatan
articulation concerning a set of connections and disconnections thought to be rele-
vant to a specic agent that is socially and historically situated, and to a particular
purpose. If the meanings and interpretations we give to phenomena as analysts and
as situated social actors have a sense of potential indeterminacy about them, then
the best we can do is to look to the relations between knowledge, context and power
that seem to x some meanings and interpretations rather than others as pre-
dominant or even hegemonic forms.

References

Akman, V., 2000. Rethinking context as a social construct. Journal of Pragmatics 32, 743759.
Appadurai, A. (Ed.), 1986. The Social Life of Things: Commodities in Cultural Perspective. Cambridge
University Press, Cambridge.
Bourdieu, P., 1977. The Outline of a Theory of Practice. Cambridge University Press, Cambridge.
Cliord, J., 1997. Routes: Travel and Translation in the Late Twentieth Century. Harvard University
Press, Cambridge, MA.
Crick, M., 1992. Ali and me: an essay in street corner anthropology. In: Okely, J., Callaway, H. (Eds.),
Anthropology and Autobiography. Routledge, London, pp. 175192.
Culler, J., 1983. On Deconstruction: Theory and Criticism after Structuralism. Routledge & Kegan Paul,
London.
Culler, J., 1988. Framing the Sign: Criticism and its Institutions. Blackwell, Oxford.
Derrida, J., 1977. Signature event context. Glyph: John Hopkins Textual Studies 1, 172197.
Dilley, R., 1999. Introduction. In: Dilley, R. (Ed.), The Problem of Context. Berghahn Books, New York
and Oxford, pp. 143.
Duranti, A., Goodwin, C. (Eds.), 1992. Rethinking Context: Language as an Interactive Phenomenon.
Cambridge University Press, Cambridge and New York.
R.M. Dilley / Language & Communication 22 (2002) 437456 455

Fabian, J., 1999. Ethnographic misunderstanding and the perils of context. In: Dilley, R. (Ed.), The
Problem of Context. Berghahn Books, New York and Oxford, pp. 85104.
Fardon, R., 1990. General introduction. In: Fardon, R. (Ed.), Localising Strategies: Regional Traditions
of Ethnographic Writing. Scottish Academic Press, Edinburgh; Smithsonian Institution Press,
Washington, DC, pp. 135.
Geertz, C., 1973. The Interpretation of Cultures, Selected Essays. Basic Books, New York.
Harvey, P., 1999. Culture and context: the eects of visibility. In: Dilley, R. (Ed.), The Problem of Con-
text. Berghahn Books, New York and Oxford, pp. 213236.
Hervey, S., 1982. Semiotic Perspectives. Allen & Unwin, London.
Hervey, S., 1999. Context, the ghost in the machine. In: Dilley, R. (Ed.), The Problem of Context. Ber-
ghahn Books, New York and Oxford, pp. 6172.
Hobart, M., 1982. Meaning or moaning? An ethnographic note on a little understood tribe. In: Parkin, D.
(Ed.), Semantic Anthropology. Academic Press, London, pp. 3964.
Holbart, M., 1985. Texte est un con. In: Barnes, R.H., de Coppet, D., Parkin, R.J. (Eds.), Contexts and
Levels: Anthropological Essays on Hierarchy. Journal of the Anthropological Society of Oxford,
Oxford, pp. 3350.
Holbart, M., 1986a. Introduction: context, meaning and power. In: Hobart, M., Taylor, R.H. (Eds.),
Context, Meaning and Power in Southeast Asia. Cornell Southeast Asia Program, Ithaca, NY, pp. 7
19.
Holbart, M., 1986b. Thinker, thespian, soldier, slave? Assumptions about human nature in the study of
Balinese society. In: Hobart, M., Taylor, R.H. (Eds.), Context, Meaning and Power in Southeast Asia.
Cornell Southeast Asia Program, Ithaca, NY, pp. 131156.
Holbart, M., 1990. Who do they think they are? the authorized Balinese. In: Fardon, R. (Ed.), Localising
Strategies: Regional Traditions of Ethnographic Writing. Scottish Academic Press, Edinburgh; Smith-
sonian Institution Press, Washington DC, pp. 303338.
Holbart, M., 1999. As they like it: overinterpretation and hyporeality in Bali. In: Dilley, R. (Ed.), The
Problem of Context. Berghahn Books, New York and Oxford, pp. 105144.
Holy, L., 1989. Kinship, Honour and Solidarity: Cousin Marriage on the Middle East. Manchester Uni-
versity Press, Manchester.
Holy, L., Stuchlik, M. (Eds.), 1981. The Structure of Folk Models. Academic Press, London.
Holy, L., Stuchlik, M., 1983. Actions, Norms and Representations: Foundations of Anthropological
Inquiry. Cambridge University Press, Cambridge.
Hymes, D., 1977. Foundations in Sociolinguistics: An Ethnographic Approach. Tavistock, London.
Keesing, R., 1972. Simple models of complexity: the lure of kinship. In: Reining, P. (Ed.), Kinship Studies
in the Morgan Centennial Year. The Anthropological Society of Washington, Washington, DC, pp.
1731.
Malinowski, B., 1935. Coral Gardens and their Magic: A Study of the Methods of Tiling the Soil and of
Agricultural Rites in the Trobriand Islands. Allen & Unwin, London.
Malinowski, B., 1923 [1938]. The problem of meaning in primitive languages, a supplementary essay. In:
Ogden, C. K., Richards, I. A. (Eds.), The Meaning of Meaning: A Study of the Inuence of Language
upon Thought and of the Science of Symbolism. Kegan Paul, Trench, Trubner, London, pp. 296336.
Marcus, G.E., Cushman, D., 1982. Ethnographies as texts. Annual Review of Anthropology 11, 2569.
Marcus, G., Fischer, M.M.J., 1986. Anthropology as Cultural Critique: An Experimental Moment in the
Human Sciences. Chicago University Press, Chicago.
Morris, B., 1999. Context and interpretation: reections on Nyau rituals in Malawi. In: Dilley, R. (Ed.),
The Problem of Context. Berghahn Books, New York and Oxford, pp. 145166.
Ormiston, G.L., Schrift, A.D. (Eds.), 1990. Transforming the Hermeneutic Context: From Nietzsche to
Nancy. State University of New York Press, Bualo, NY.
Palmer, R.E., 1969. Hermeneutics: Interpretation Theory in Schleiermacher, Dilthey, Heidegger, and
Gadamer. Northwestern University Press, Evanston.
Parkin, D., 1982. Introduction. In: Parkin, D. (Ed.), Semantic Anthropology. Academic Press, London.
Scharfstein, B.-A., 1989. The Dilemma of Context. New York University Press, New York and London.
Anon., The Shorter Oxford English Dictionary. Oxford.
456 R.M. Dilley / Language & Communication 22 (2002) 437456

Strathern, M., 1987. Out of context: the persuasive ctions of Anthropology. Current Anthropology 28
(1), 251281.
Strathern, M. (Ed.), 1995. Shifting Contexts: Transformations in Anthropological Knowledge. Routledge,
London and New York.
Taussig, M., 1987. Mimesis and Alterity: A Particular History of the Sesnses. Routledge, London and
New York.

Dr. Roy Dilley is Senior Lecturer in the Department of Social Anthropology at the University of St
Andrews. He has interests in social theory and recently edited the volume entitled The Problem of Context
(Oxford: Berghahn, 1999), which draws together contributions from a group of scholars in the elds of
social anthropology and linguistics. He has also written on the concept of the market [Dilley, R. (ed.),
Contesting Markets: Analyses of Ideology, Discourse and Practice, Edinburgh University Press, 1992], and
has published widely on ethnographic issues relating to Senegal, West Africa.