http://ant.sagepub.com
Anthropological Theory
DOI: 10.1177/1463499608096646
2008; 8; 399Anthropological Theory
Matt Hodges
Rethinking time's arrow: Bergson, Deleuze and the anthropology of time
http://ant.sagepub.com/cgi/content/abstract/8/4/399
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:Anthropological TheoryAdditional services and information for
http://ant.sagepub.com/cgi/alertsEmail Alerts:
http://ant.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.co.uk/journalsPermissions.navPermissions:
http://ant.sagepub.com/cgi/content/refs/8/4/399Citations
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
Anthropological Theory
Copyright © 2008 SAGE Publications
(Los Angeles, London, New Delhi and Singapore)
http://ant.sagepub.com
Vol 8(4): 399–429
10.1177/1463499608096646
399
Rethinking time’s arrow
Bergson, Deleuze and the anthropology
of time
Matt Hodges
University of Exeter, UK
Abstract
Since the early 1970s, time has come to the fore as a constitutive element of social
analysis in the guise of what I term here ‘ﬂuid time’. Anthropologists of multiple
theoretical persuasions now take for granted that social life exists in ‘time’, ‘ﬂow’, or
‘ﬂux’, and this temporal ontology is commonly accepted as a universal, if habitually
unquestioned, attribute of human experience. Similarly, it underpins today’s dominant
paradigm of ‘processual’ analysis, in its many forms. Yet this concept is notably
under-theorized, in keeping with a history of uneven study by social scientists of time.
In this article I draw on anthropological approaches by Gell and Munn, and
philosophical work by Bergson and Deleuze, to put forward a critical theorization. I
then discuss its ramiﬁcations. Ultimately, I argue that this model points to a
rapprochement between the anthropological study of time and history, sociality and
temporality, and an enhanced role for temporal analysis in anthropological theory.
Key Words
Bergson • Deleuze • emergence • ﬂuid time • history • la durée • processual analysis •
temporal ontology • time
To restore to practice its practical truth, we must therefore reintroduce time into the
theoretical representation of a practice which, being temporally structured, is intrinsically
deﬁned by its tempo.
Pierre Bourdieu, Outline of a Theory of Practice (1977 [1972]: 8)
‘Action’ or Agency, as I use it, thus does not refer to a series of discrete acts combined
together, but to a continuous ﬂow of conduct.
Anthony Giddens, Central Problems in Social Theory (1979: 55)
The central assertion of this book is that the world of humankind constitutes a
manifold, a totality of interconnected processes . . .
Eric Wolf, Europe and the People without History (1982: 3)
399-429 096646 Hodges (D) 3/11/08 09:06 Page 399
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
As soon as culture is no longer primarily conceived as a set of rules to be enacted by
individual members of distinct groups, but as the speciﬁc way in which actors create
and produce beliefs, values, and other means of social life, it has to be recognized that
Time is a constitutive dimension of social reality.
Johannes Fabian, Time and the Other (1983: 24)
TIME’S FLOW: ATHEORETICAL LACUNA?
Social scientists and philosophers, even physicists, often quote St Augustine’s perplexed
remarks on time when beginning their discussions, to underline the complexity of the
task ahead of them, and warn readers up front that things are about to take a demanding
turn.1 Yet despite its complexity, time is not quite as slippery a ﬁsh as St Augustine
suggests, and some recent writers have done well to demonstrate that, while shedding
important light on how the human experience of time might be grasped (e.g. Adam,
1990, 1998; Gell, 1992; James and Mills, 2005a; Munn, 1992). The reason for such
introductory disclaimers about time, for anthropologists at least, perhaps lies elsewhere.
Nancy Munn has made some critical comments that paint an unﬂattering portrait of
the history of the anthropology of time.2 She notes (1992: 93) an ‘insufﬁcient theoretical
attention to the nature of time as a unitary, focal problem’ among anthropologists:
When time is a focus, it may be subject to oversimpliﬁed, single-stranded descriptions
or typiﬁcations, rather than to a theoretical examination of basic sociocultural
processes through which temporality is constructed . . . [T]he problem of time has
often been handmaiden to other anthropological frames and issues . . . with which it
is inextricably bound up . . . [and] frequently fragments into all the other dimensions
and topics anthropologists deal with in the social world.
Time has certainly not been lacking from anthropological theorizing, and inﬂuential
ﬁgures have contributed important works, including Durkheim (1915), Malinowski
(1927), Evans-Pritchard (1939, 1940), Leach (1961), Lévi-Strauss (1963), and Geertz
(1973). However, through a concise overview of such works, Munn illustrates the relatively
cursory attention paid to the subject until the 1970s, particularly with respect to
its role in core theoretical models. It is only in the 1970s and 1980s, Munn argues, that
writers began to pay more attention to time as a focus, and even then, the ﬁeld of time
studies has remained relatively undeveloped compared to other areas of anthropological
study. Despite some valuable work, therefore, it is true to say that the anthropological
understanding of time is comparatively immature.
Nevertheless, during recent decades, time has come to the fore as a constitutive
element of theoretical analysis, chieﬂy, I argue here, in the guise of the somewhat hazy
notion of time as ‘ﬂow’, or ‘ﬂux’. While earlier anthropological movements such as functionalism
and structuralism largely overlooked time as a feature of theoretical models, a
necessary oversight for any static, a-temporal analytical perspective, from the 1970s
onwards several inﬂuential theorists introduced a more explicit temporality back into
social analysis with their visions of social life as constituted ‘in time’ (cf. Giesen, 1992).3
Bourdieu (1977: 6–9), for example, posited that human interaction as a form of practice
is ‘inscribed in the current of time’, ‘with its rhythm, its orientation, its irreversibility’
(1997: 9, my emphasis). It is thus intrinsically deﬁned by its tempo, which must form a
ANTHROPOLOGICALTHEORY 8(4)
400
399-429 096646 Hodges (D) 3/11/08 09:06 Page 400
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
core component of any analysis, particularly of the gift. Indeed, his concept of practice
is intrinsically dependent on the spatialization of time through such seemingly riverine
imagery. Similarly, he inﬂuentially criticized structuralist and objectivist models for their
‘detemporalized’ character.4 In this respect, Bourdieu is said to have been inﬂuenced by
his early readings of phenomenological philosophy, but it is hard to locate explicit clariﬁcation
of his own, inﬂuential temporal ontology, which one must effectively infer from
his writings, or tacitly adopt through utilizing his theoretical models. Anthony Giddens
(1979: 53–65, 198–233) proposed a related approach, whereby the ﬂow of time underwrites
the key process of historical reproduction, or ‘structuration’. He is more explicit
on his sources than Bourdieu, and openly grounds his understanding of ‘lived-through
experience’ in the work of phenomenological philosophers (Giddens, 1979: 54–5).
However, he does so only cursorily, and thus still relies on a problematically unclariﬁed
root vocabulary of ﬂow and process – for example: ‘“Action” or agency, as I use it, thus
does not refer to a series of discrete acts combined together, but to a continuous ﬂow of
conduct’ (Giddens, 1979: 55, emphasis retained). These innovations were accompanied
by a development in political economic perspectives pioneered by writers such as Wolf
(1982), who saw ‘history’ in the canonical Marxian sense as providing the foundation
for wide-ranging social analyses of interlinked modes of production, or ‘historical
processes’, as opposed to the relatively a-temporal, static analyses associated with structural
Marxists in the 1970s.5 While Wolf’s approach to time is more explicitly ‘linear’
and Anglo-Saxon in outlook, and closely tied to the totalizing Marxian model of ‘historical
time’ (cf. Agamben, 1993: 91–105), it nevertheless takes the notion of time’s ﬂuidity
and accompanying processual vocabulary that underpins this model as an unproblematic
given. Despite the signiﬁcant differences in the approaches of these theorists, therefore,
their foundational understanding of time as a ﬂuid underlying process, ﬂow, or ﬂux
is comparable, if difﬁcult to pin down with any precision. They neglect to clarify the
temporal ontology of their models.6 With respect to an explicit anthropology of time, in
such models, time is chieﬂy a constitutive, not focal aspect of the theoretical apparatus
(cf. Thomas, 1996: 5–6).7
Since the late 1980s, this intrinsic temporal ontology has become a deﬁning feature
of social analysis. Many political economists have adopted a reﬂexive approach to
‘historical process’, ditching the lack of coevalness (cf. Fabian, 1983: 156–65) characteristic
of earlier Marxist frameworks, and signalling an approximation to the processual,
symbolic, or practice-theory approaches that have developed in direct descent, or by the
distaff from Bourdieu, Giddens and others.8 And in their turn, many practice theorists,
who initially tended to retain a focus on discrete and questionably bounded cultures,
have integrated the wider frameworks of political economists into their models. During
the 1990s and 2000s this trend intensiﬁed and broadened its inﬂuence, as acknowledgement
of the impact of a globalized economy and accompanying rapid historical change
on even the most isolated locales seemed to underline the pertinence of these approaches
(cf. Gupta and Ferguson, 1997; Ong and Collier, 2004). Indeed, as globalization has
become indubitably implicated in anthropological analysis, and the notion of bounded
cultures increasingly problematic, the sense that social experience is ‘ﬂuid’, and exists in
relation to a global capitalist system that is constituted of capital, labour and other
‘ﬂows’, has clearly become the dominant paradigm – propositions which are in turn
predicated on implicit notions of temporal ﬂow or historical ﬂux. In this respect, ﬂuidity
HODGES Rethinking time’s arrow
401
399-429 096646 Hodges (D) 3/11/08 09:06 Page 401
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
has become an even more appropriate root metaphor for the temporality of analysis,
facilitating both the empirical analysis that rapid social change is endemic to modern
social life and the political economic thesis that social change was also intrinsic to ‘premodern’
societies. We can thus propose that entangled notions of temporal ﬂow and
socio-historical change lie at the heart of the temporal modalities of contemporary
anthropological theory.9 Despite differences, therefore, regarding the most effective way
to analyse human sociality, and differing conceptions of how to implement a temporally-inﬂected
social analysis, along with an increased coeval sensitivity as to how anthropological
subjects experience time, change and history, all such approaches are
underpinned by a tacit unspeciﬁed temporal ontology that is evoked through a common
root vocabulary of process, ﬂow or ﬂux – itself implying, and facilitating in an
unspeciﬁed way the notion that time involves ‘change’.10
These approaches are characteristic of a broader trend in cultural and philosophical
theory in the last 30 years, sometimes associated with the phenomenon of postmodernism.
As Harris (1996: 7) clariﬁes: ‘the “postmodern moment” deﬁned as a
distinctive sense of temporality in anthropology does not identify ﬂuidity as a symptom
of modernity, but rather claims it to be a universal property of human societies and
culture’. Harris’s article alights on this most notable, yet under-theorized feature of
contemporary anthropological theory, which was arguably fundamental to the incipience
of contemporary paradigms. It is a notion that in different ways has been present
in philosophical systems for many thousands of years – as the ancient Greek philosophers
and early Buddhist scriptures testify (cf. Kahn, 1981; Kirk et al., 1983; Thomas and
Hjort, 1996).11 It also features signiﬁcantly in the work of philosophers who have been
inﬂuential in shaping the temporal ontology of modern social analysis (e.g. Heidegger,
1993 [1927]; Whitehead, 1979 [1927–8]). It has ﬁgured prominently in other ﬁelds of
western intellectual production, such as the work of canonical ‘high modernist’ novelists
such as Woolf (2000 [1925]) and Faulkner (1993 [1937]) – although inﬂuential
‘postmodernist’ ﬁction writers such as Borges (1970 [1941]) or Pynchon (1973), interestingly
enough, have tended towards more ‘event-centred’ theories of temporal perception
(Heise, 1997: 47–68).12 And it has a broad, but undoubtedly contingent, presence
in the cultural practices of contemporary western and other societies, which cannot be
unrelated to the radical and accelerated pace of global social change experienced in the
last 200 years. Yet, remarkably, deﬁnitions of this assumption are rare in the social
scientiﬁc literature, as are challenges to its hegemony in contemporary social analysis.
Fluid time, as we can term it, is inherent in current theoretical models as the (chieﬂy
metaphorical) motor facilitating the ongoing reproduction and modiﬁcation of social
life, and is a constituent component of many varied forms of social analysis broadly
treating of ‘historical’, ‘processual’, ‘political economic’, or ‘practice-based’ approaches –
despite the fact that, as James and Mills point out (2005b: 13), ‘the realm of time is not
agreed even by the specialist physical scientists and philosophers to be one thing, one
ﬁeld’. Indeed, even though there is highly nuanced ethnographic attention paid in recent
anthropological studies of time to the divergent and at times contradictory ways in
which time is experienced (e.g. Orlove, 2002; Hirsch and Stewart, 2005b; James and
Mills, 2005a), this underlying vocabulary remains necessarily present. In this sense, it is
arguably a totalizing category of modern anthropological theory (if a relatively ﬂexible
and productive one), operating, in Osborne’s (1995: 28) deﬁnition, ‘insofar as all such
ANTHROPOLOGICALTHEORY 8(4)
402
399-429 096646 Hodges (D) 3/11/08 09:06 Page 402
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
totalizations abstract from the concrete multiplicity of differential times co-existing in
the global “now” a single differential (however internally complex) through which to
mark the time of the present’. In current usage, it is also but one, contingent evocation
of how we might conjure the complex workings of ‘spacetime’, with the potential for
further conceptual elaboration. Clearly overlooked, like time more generally, the
assumption of time’s ‘ﬂuidity’ has received ‘insufﬁcient attention’.13
An extended anthropological study of the links between metaphors of ﬂuidity,
change, related historical contexts and contemporary cultural theory would undoubtedly
be fascinating, and might possibly provide a basis for reconsidering our reliance
on ﬂuid time in theoretical models – but that is beyond the scope of this article.14
And I would stress that I do, of course, accept that this complex of foundational
tropes has offered invaluable ways of analysing human sociality – notably through the
approximation it affords between the disciplines of anthropology and history. It is
clear, however, that the ‘concept’ suffers from a lack of clarity and a potential analytical
tendency towards homogenizing temporal difference. Rather than offer a summary
deﬁnition at this point, however, my intention – for reasons that will become apparent
– is to ﬂesh out a complex working deﬁnition of the most metaphorically neutral of
these terms, ‘ﬂux’, and tease out some of its implications – including addressing the
charge that it necessarily operates as a totalizing concept. For orientation, I begin with
a critical discussion of two key approaches to the anthropology of time, by Gell
(1992) and Munn (1992), paying special attention to the question of ﬂux. I move on
to critically examine philosophical works by Bergson and Deleuze that constitute an
increasingly inﬂuential source for a contemporary model of ﬂux. Then, in the last
section of the article, I synthesize and develop these findings. This will enable
precise theorization, and development of this key component of social analysis. In
turn, it will illuminate how problematizing fluidity has ramifications that extend
beyond the study of time into the configuration of our models of social practice
more generally – and in this respect, I will comment in passing on the implications
for topics such as social change, cross-cultural analysis and historical anthropology.
In conclusion, it will also permit the intimation of a paradigm for the anthropological
analysis of time and historical practice with an explicit notion of ﬂux at its core,
which places the temporal squarely and explicitly at the heart of anthropological
analysis.
THE ANTHROPOLOGY OFTIME:TWO KEY ORIENTATIONS
Let us now turn to examine the temporal ontologies put forward by key anthropological
theorists of time, and the implications for our interest in ﬂux. The ﬁrst of our
studies, Gell’s The Anthropology of Time (1992), is one of the few major works on time
by an anthropologist and, largely theoretical in nature, is an admirable work of interdisciplinary
synthesis. Gell is clear on the theoretical stance he presents in his study, and
the debt he owes to the analytic philosopher D.H. Mellor, from whose book Real Time
(1981) he draws many of his philosophical foundations. He describes his position as ‘the
moderate version of the B-series position’ (Gell, 1992: 156), and it comprises a synthesis
of what are termed by philosophers ‘A-series’ and ‘B-series’ time. These form the basis
for the 20th-century analytic tradition in the philosophy of time, of which Mellor is the
most inﬂuential contemporary exponent.
HODGES Rethinking time’s arrow
403
399-429 096646 Hodges (D) 3/11/08 09:06 Page 403
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
B-series time, Gell suggests, is objective, ‘real’ time: ‘it reﬂects the temporal
relationships between events as they really are, out there . . . All events, including
future events, have their dates, which are unqualiﬁed temporal attributes of events’
(Gell, 1992: 165, 156–7).15 B-series time, however, is unlike time as perceived by
human subjects, as it has no past, present and future dimensions, that is, it is fundamentally
untensed. It instead refers to what Gell, following Mellor, describes as the
unchanging nature of events outside the realm of human agency: namely, that all
events have ‘dates’ in relation to each other, and these dates are permanent, unchanging
and situate events in deﬁnitive temporal relation to each other. Notions of past,
present and future are simply ﬂeeting, insigniﬁcant attributes of events that are gained
and lost from day to day. Change, of course, takes place, and events themselves, Gell
suggests, ‘are the changes that happens to things, bringing about new states of affairs’
(1992: 161). And Gell also acknowledges the apparently ethnocentric nature of the
notion of dates, and stresses that by being ‘dated’ he merely means to say that events
have deﬁnite and unchanging temporal relationships to each other, rather than
implying that events exist in some form of ethnocentric calendar (1992: 159). In sum,
what Gell’s B-series amounts to is a metaphysical statement about the objective,
autonomous nature of real time: events exist, have deﬁnite relationships to each other,
and effectively provide an objective ground for, and structure to, the world and its
‘history’. But importantly, their ‘true’ nature is objectively inaccessible to the tools of
human perception. Gell’s B-series model, therefore, is the explicit foundation for his
theory’s temporal ontology.
Now, human beings evidently experience and perceive events and change, and the
way they do so is in keeping with A-series time, that is to say the subjective, tensed existence
involving past, present and future relations that comprises everyday human time
perception. B-series time provides the basis for A-series perception, which Gell models
on Husserl’s phenomenological theory of internal time consciousness; but the ‘real’
world does not exist according to A-series laws of perception. As Gell writes, ‘[w]e have
no direct access to the temporal territory [of the B-series] because all our mental life, all
our experiences, beliefs, expectations, etc. are themselves datable events, conﬁned to their
localized time-frames, like all other datable events’ (Gell, 1992: 238). Instead, we know
B-series time through temporal models, which reﬂect the structure of B-series time
without accessing it directly (1992: 240):
Our access to time is conﬁned to the A-series ﬂux, through which we interact with
‘real’ time, via the mediation of temporal maps which provide us with a surrogate for
real time. These reconstructions of B-series time are not the real thing . . . but we are
obliged to rely on them.
Gell’s grounding of the human perception of time, or A-series time, in the ‘real’
world of B-series time provides a valuable directive for grounding human subjects in
the physical world, hence suggesting how the perception of time might be conceived
as physically shaped, as well as culturally constructed and ultimately individually
perceived, and I sympathize with the desire to situate the human perception of time
‘within’ a wider theory of ‘real’, or ‘non-human’ time. This is in contrast to other recent
studies of the anthropology of time, such as that of Greenhouse (1996), who suggests
ANTHROPOLOGICALTHEORY 8(4)
404
399-429 096646 Hodges (D) 3/11/08 09:06 Page 404
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
that the human experience of time is exclusively the product of A-series perception.
We can note, therefore, that Gell’s B-series provides one possible foundation for the
notion of ﬂuid time found in theorists such as Bourdieu, Giddens and Wolf, even if
it doesn’t constitute a model of temporal ‘ﬂux’ per se. Indeed, it is in relation to such
an ‘objective’, B-series time, I suggest, and its grounding of human sociality, that any
attempt to deﬁne the notion of time as ﬂux would have to lie. While I therefore accept
the framework of Gell’s model for an approach to time, I nevertheless take issue with
his particular conception of its workings. First, his reliance on Husserl’s model of timeconsciousness
is problematic given Husserl’s concern with a transcendent consciousness
(Kearney, 1994: 22–3), and the subsequent developments in phenomenological
philosophy that have critiqued and developed Husserl’s approach (e.g. Heidegger,
1993; Merleau-Ponty, 1989). And secondly, Gell’s characterization of B-series time as
related to dated events, despite his disclaimers, remains a dominant tenet of his presentation
of Mellor’s work, and is signiﬁcantly questionable, resembling as it does a spatialized
conception of linear time that is brazenly cultural in character.16 When, for
example, does an ‘event’ begin? How does one deﬁne the positioning of events in
relation to one another? Surely the analogy of the calendar which he puts forward for
this purpose derives much too explicitly from human, A-series laws of perception. In
what sense, therefore, does the language of chronological quantity and identity drawn
upon by Gell accurately reﬂect the ‘unrepresentable’ nature of ‘real time’? How does
such a ‘dated’ model of time relate to anthropological notions of historical time, for
which it must surely provide a framework, or indeed the elusive notion of ﬂuid time,
which we are pursuing in terms of ﬂux? And, more importantly, might one draw on
a less culturally-inﬂected approach to underpin an explicitly anthropological model of
‘real’ B-series time? For it is certainly arguable that Gell’s conception of the B-series
incorporates a spatialized conception of linear time that is too western in character of
be of value to the cross-cultural anthropological project. Now, I evidently do not wish
to try and outmanoeuvre Mellor’s distinguished analytical approach – on which Gell
closely relies – in this article. My approach is more pragmatic. At this stage, I shall
simply note that in our quest for a temporal ontology that ﬂeshes out a notion of ﬂux,
it would be productive to examine an alternative theory of the relation between
‘human’ (A-series) and ‘objective’ (B-series) time, that might present less culturallycontingent
features.17
Turning now to the work of Nancy Munn, her article ‘The Cultural Anthropology of
Time: A Critical Essay’ (1992), while not as expansive as Gell’s work, nevertheless
presents the most practically applicable anthropological theory of time currently articulated.
Her approach, a variant of contemporary ‘practice theory’, is most valuable for its
utility in interpreting both the lived experience, and conceptual perception of time by
human subjects; a dimension of her theory illustrated in detail in her ethnographic work
(Munn, 1983, 1986). In this respect she moves signiﬁcantly beyond Gell’s discussion
and development of Husserl, which remains mainly philosophical in nature. Grounding
her position in a phenomenological theory of the human perception of time as
temporality, Munn suggests that the conscious and tacit, embodied experience of time is
the product of concrete, temporalizing practices whereby the inherent temporal character
of social life is brought out. The summary of her position, although abstract in nature,
is worth quoting in full:
HODGES Rethinking time’s arrow
405
399-429 096646 Hodges (D) 3/11/08 09:06 Page 405
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
[Human temporality is] a symbolic process continually being produced in everyday
practices. People are ‘in’ a sociocultural time of multiple dimensions (sequencing,
timing, past-present-future relations, etc.) that they are forming in their ‘projects’. In
any given instance, particular temporal dimensions may be foci of attention or only
tacitly known. Either way, these dimensions are lived or apprehended concretely via
the various meaningful connectivities among persons, objects, and space continually
being made in and through the everyday world. (Munn, 1992: 116)
Human temporality, or temporalities if one considers its multiple dimensions, a
symbolic process, is thus grounded in everyday social practices, and is the product of
these practices, or what Munn also calls, in a phenomenological vein, ‘intersubjectivity’.
It is simultaneously an inescapable dimension of these practices, and that includes
anthropological writing about time: ‘We cannot analyse or talk about time without using
media already encoded with temporal meanings nor, in the course of doing so, can we
avoid creating something that takes the form of time’ (1992: 94). This has signiﬁcant
consequences for the relationship between time and space, of course. For Munn, ‘[i]n a
lived world, spatial and temporal dimensions cannot be disentangled, and the two
commingle in various ways’ (1992: 94). This leads her to characterize social life as
comprising a ‘lived spacetime’, a position she summarizes concisely in an earlier article
on Gawan kula: ‘[S]ociocultural action systems . . . do not simply go on in or through
time and space, but they form (structure) and constitute (create) the spacetime manifold
in which they “go on”’ (Munn, 1983: 280).18
With reference to the workings of how time is experienced, it will sufﬁce to examine
Munn’s comments on the past. Perception and experience of the past involves actualizing
it in the present, or, in Munn’s terminology, temporalizing the past, and ‘foregrounds
the implications of the meaningful forms and concrete media of practices for apprehension
of the past’ (Munn, 1992: 113). At the same time, it can also involve future-orientations,
as ‘the past-present-future relation . . . is intrinsic to all temporalizations
irrespective of focus, inasmuch as people operate in a present that is always infused, and
which they are further infusing, with pasts and futures’ (1992: 115). This viewpoint sees
perception and experience of the past (present and future) as implicated in the dynamic
process of temporalization that comprises the lived present. Finally, temporalizing practices
are also viewed as a dimension of the exercise of power, as temporality is a hinge
that connects subjects to wider social horizons, and control over pasts and futures that
are temporalized also inﬂuence action in the present. Thus ‘[c]ontrol over time is not
just a strategy of interaction; it is also a medium of hierarchic power and governance’
(1992: 109); and in addition to control over pasts and futures, this may take place
through all the other various temporal dimensions of social life, from clock time and
calendars to the organization of working routines and the biological rhythms of the body.
It is of the essence of human time viewed as temporality, therefore, that such temporal
media are not only known through reﬂective perception, but also embodied unconsciously
in and through the intersubjective practices of daily existence. Munn’s theory
of temporality thus provides a counterpart to the practice paradigm mentioned earlier.
I endorse much of what Munn has to say about temporalizing practices, although I
will now present two observations of relevance to the theme of this article. First, in the
restricted space of her essay, the philosophical foundations for Munn’s approach are
ANTHROPOLOGICALTHEORY 8(4)
406
399-429 096646 Hodges (D) 3/11/08 09:06 Page 406
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
necessarily abbreviated, although she makes clear here and elsewhere in her work the
debt she owes to the phenomenological tradition (cf. Munn, 1986: 20). Her efforts to
see temporality as the product of sociocultural processes, however, and her reliance on
the subject-centred account of time provided by the phenomenological project, are problematic.
For although she locates temporality in the context of a wider temporal universe,
referring to ‘time’s pervasiveness as an inescapable dimension of all aspects of social
experience and practice’ (Munn, 1992: 93), she is not explicit on the ontological
foundations which might locate time within material, as well as symbolic processes. In
other words, Munn transposes A-series concerns into an anthropological framework
with a good deal of success, but neglects to explicitly articulate any relationship to the
B-series. This suggestion of a more materialist dimension is side-stepped, jeopardizing
as it does her emphasis on social experience and knowledge as culturally constituted,
although one could immediately point out, of course, that her conception of social life
as processual is itself universalist in nature (cf. Harris, 1996).19 Similarly, this ensures
that her discussion of time remains discrete, a dimension of social practice, in other
words, severed from a model of historical time or globalized processes (cf. Thomas, 1996:
5). As a result, we also have little sense of how non-human processes can be theorized
to have agency in human affairs (cf. Latour, 2007). Likewise, we have no sense of how
these apparently discrete temporalizing practices relate to social practice more generally.
For the phenomenological philosophers, temporality was a core dimension of human
being (Heidegger, 1993; Merleau-Ponty, 1989). For Munn, it has become a conceptual
adjunct of sociality. In sum, we have no ontological framework through which to correlate
the multiple social and material trajectories of historical time.
Secondly, there is the question of terminology, namely the vocabulary of ﬂuid time
itself. Like so many other writers on practice, Munn provides no clear deﬁnition of this
regime of tropes, but still invokes it freely. That would not be so controversial were she
not writing explicitly about the anthropology of time. It is an issue directly connected
to Munn’s side-stepping of a deﬁnition of the B-series. In sum, Munn’s impressively
condensed work is incomplete, as we are left with an approach that appears to deﬁne
the cultural component of human temporality with reference to a general ontology of
ﬂuid time, but neither deﬁnes its B-series (‘objective’) character, nor its relationship to
wider social practice, with any transparency. (An observation, one might add, that I have
already made in related terms of processual theory more generally.)
Both Munn and Gell present similarities in the phenomenological approach they
adopt to human temporal experience (although for anthropologists, Munn presents a
more advanced analytical apparatus). But as we have seen, neither Munn nor Gell
provide an effective model of B-series time – which is vital to any precise theorization
of ﬂux. Munn is content to use a practice model for time experience that does not seek
to examine its own temporal foundations. But her implication that anthropologists are
preoccupied with the cultural experience of time (cf. Greenhouse, 1996), hence obviating
the need for a coherent B-series theory, does not elide the need for reﬂexive knowledge
of the temporal ontology of our theoretical models. Gell, meanwhile, presents a
model of time that relies on clearly ‘ethnocentric’ metaphors for time perception, and
his phenomenological model is problematic. Finally, neither addresses the role that their
respective temporal metaphors and ontologies play in the practice paradigm itself, so
neither proﬁt from the potential beneﬁts to be had in doing so, as we shall see – in
HODGES Rethinking time’s arrow
407
399-429 096646 Hodges (D) 3/11/08 09:06 Page 407
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
particular with regard to the explicit integration of temporal analysis with praxis. In this
regard, both are content to approach the anthropology of time as a discrete area of study,
despite arguing for time’s pervasiveness in social life. To which one might reply: if time
is pervasive, then we must develop a way of analysing it pervasively.
B-series time, then, and by implication the nature of temporal ﬂux, would beneﬁt
from the exploration of an alternative deﬁnition, both regarding its explicit nature and
its relationship to human sociality. And given that no anthropological alternative is
forthcoming, we turn to the philosophy of time. Let us set down some brief pointers for
orientation. For an anthropologist, any such enquiry must ultimately address the pragmatic
requirement of how material time might most usefully be evoked for the purposes
of cross-cultural enquiry. It should also query what validity this deﬁnition might have
as a universalist claim. It must also acknowledge the limitations of such an enquiry, given
the decidedly complex, multiform and ultimately socio-cultural nature of our understanding
and conceptualization of time. But, of course, it should respond with precision
to the pressing requirement for a deﬁnition of ﬂux. The goal is therefore an elucidation
of the concept of ﬂux, in the context of a pragmatic, working, cross-cultural model of
historical time experience, and with these demands before us I now turn to examine the
philosophical work of Bergson and Deleuze. The potential beneﬁts for social analysis
will be seen to include reﬂexive awareness of the temporal ontology of anthropological
paradigms; the elision of spatialized metaphors of time from theoretical paradigms; and
a more dynamic model of socio-temporal experience that facilitates fuller integration of
temporal analysis with models of socio-historical practice.
TOWARDS ATHEORY OF FLUX: BERGSON, DELEUZE AND LA DURÉE
Time has, of course, been an important focus of study throughout the history of philosophy.
Turetzky’s study (1998) provides a valuable overview and distinguishes three
principal traditions in 20th-century philosophy: the analytic tradition, the phenomenological
tradition and a distaff tradition. The analytic tradition, taking as its starting point
McTaggart’s (1908) thesis concerning time’s ‘unreality’, focuses on the relationship
between an objective, universal notion of time as static and quantitative (‘B-series time’),
and the human experience of time and temporal becoming (‘A-series time’). It has found
explicit recent voice within anthropology in Gell’s (1992) work, as I have shown,
although as Gell points out, many ‘common-sense’ notions that have informed previous
anthropological conceptions of time can be viewed as incorporating A-series and B-series
insights, to varying extents.
The phenomenological tradition has had a profound inﬂuence on anthropological
thought, and on anthropological treatment of time more generally, as it tends to focus
explicitly on A-series time – that is to say, the world of human experience that preoccupies
anthropologists. Its most inﬂuential theorists have been Husserl (1966 [1887]),
Heidegger (1993 [1927]), and Merleau-Ponty (1989 [1945]). Phenomenologists insist
on placing subjectivity and intentionality at the centre of any investigation of human
temporality; they have thereby been open to criticism for their subject-centred, A-series
approach. They have perhaps their most comprehensive social scientiﬁc exponent in
Schutz (1962); and, more recently, phenomenological theories of temporality have been
inﬂuential in the work of theorists such as Bourdieu and Giddens, and in symbolic
anthropology, in particular of course in Munn’s work on time. Finally, there is a
ANTHROPOLOGICALTHEORY 8(4)
408
399-429 096646 Hodges (D) 3/11/08 09:06 Page 408
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
signiﬁcant ‘distaff’ tradition, which Turetzky grounds in the work of Bergson (1944,
1960, 1988) and Deleuze (1990, 1991, 1994), and which, as its name suggests,
constitutes a less institutionalized, though no less signiﬁcant line of development.
Although the work of Bergson inﬂuenced to some extent the phenomenology of
Heidegger and Merleau-Ponty, distaff writers have had little explicit inﬂuence on the
way anthropologists think about time, despite the fact that related thinkers such as
Foucault have acquired a considerable anthropological following. It is of interest,
however, that from Turetzky’s cogent perspective (1998: 118), the distaff tradition (as it
appears in Deleuze’s interpretation and adaptation of Bergson’s work) can both account
for and incorporate the most signiﬁcant aspects of both analytic and phenomenological
approaches to time, while remedying their principal drawbacks. This suggests a relevance
to our concerns, given queries raised concerning the drawbacks of Gell’s reliance on
analytical philosophy, and Munn’s under-theorizing of the relationship between human
temporality and B-series time. Indeed, Bergson and Deleuze’s primary concern derives
from a preoccupation with the continuous, multi-stranded emergence of novelty within
the universe and the ramiﬁcations of this proposition for the understanding of time and
consciousness – that is, the very question of ﬂux itself. Let us consider, therefore, the
nature of what, we should note beforehand, is a complex but potentially valuable
theory.20
Bergson presented his inﬂuential theory of time as la durée, or ‘duration’, and its
signiﬁcance for the understanding of consciousness in a series of works published during
the late 19th and early 20th centuries, but most fully in Matière et Mémoire (ﬁrst
published in 1896), and L’Evolution créatrice (appeared 1907).21 Deleuze, writing from
the 1950s onwards, elucidated and extended Bergson’s position in his own philosophical
publications, ﬁrst in a number of shorter works on Bergson himself, ‘Bergson
(1859–1941)’ (1956a), ‘La Conception de la différence chez Bergson’ (1956b) and
Bergsonisme (1966); and then principally in two wider-ranging philosophical works,
Différence et Répétition (1968) and Logique du sens (1969).22 It is not my intention here
to explore and illuminate the shifts and developments in the position of each writer
within their different publications, or to examine in detail the critiques of Bergson’s
earlier theories and Deleuze’s accommodation and development of them. The synthesis
offered of their respective work draws largely on Deleuze’s extension of, and amendments
to, Bergson’s position; and while for the sake of clarity I retain elements of Bergson’s
technical vocabulary, the position outlined is principally inﬂuenced by my reading of
Deleuze and his interpreters, and tailored for an anthropological agenda.23
Both Bergson and Deleuze can be said to ground their theories of time in the notion
of la durée, although the ways in which they present and explicate it differ. Central for
both, however, is the distinction, adapted by Bergson from the theories of the physicist
and mathematician G.B.R. Riemann, between quantitative, or discrete, and qualitative,
or continuous, multiplicities.24 Quantitative multiplicities are numerical in nature, and
take the form of the one and the many: their differences are homogeneous differences
of degree, and such multiplicities can therefore be divided without occasioning a difference
in kind. Qualitative multiplicities, by contrast, on division create heterogeneous
differences. Simpliﬁed for the purposes of this discussion, they comprise an interrelated
(i.e. relational) inﬁnite whole, where any multiple is fused with all other multiples, and
any one cannot either be isolated or change without all others changing – as Deleuze
HODGES Rethinking time’s arrow
409
399-429 096646 Hodges (D) 3/11/08 09:06 Page 409
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
speciﬁes, they are multiplicities ‘of succession, of fusion, of organization, of heterogeneity,
of qualitative discrimination, or of difference in kind . . . that cannot be reduced
to numbers’ (1991: 38). Drawing on this distinction, la durée is abstractly and analogously
deﬁned as consisting of concrete, qualitative multiplicities, which divide continuously.
These multiplicities in reality comprise the life and matter of the universe, which
one can therefore describe as existing in a state of incessant, relational division, ‘ﬂux’, or
‘individuation’.25 La durée is therefore a non-chronological conception in its essential
nature, and its tendency to differentiate may be viewed as the origin of the phenomenon
we subsequently call ‘time’.
The way in which the reproductive logic of qualitative multiplicities can be thought
of or represented is thus problematic, and differs from the logic of representation
associated with the division of quantitative multiplicities. For both Bergson and Deleuze,
la durée, as the differentiation of qualitative multiplicities, occurs through the ‘actualization
of the virtual’, rather than the ‘realization of the possible’. Following the logic of la
durée, the possible can only exist in retrospect: while the new may be ‘possible’ before it
exists in the sense that there is nothing to prevent its occurring, this does not mean that
its actual occurrence is concretely foreseeable, for as a ‘possibility’ it only exists once it has
occurred (cf. Turetzky, 1998: 197–8). As Grosz (1999: 26) elaborates: ‘To reduce the
possible to a preexistent phantomlike version of the real is to curtail the possibility of
thinking about emergence, an open future not bound directly or strictly to the present.’26
For la durée, with respect to the act of division, one must instead see the virtual as
productive of the actual, without, of course, the virtual ever being actualized (cf.
Deleuze, 1994: 258–65). The virtual is therefore deﬁned by Deleuze in Proust’s (1999:
264) words as ‘real without being actual, ideal without being abstract’. It is the tendency
that produces the actual, a de jure, rather than de facto principal of differentiation that
is never actualized. This explanation is necessary because, as Boundas (1996: 91)
explicates:
Left to its own resources, the process/production of entities will permit only the
discernment of nuances or of differences of degree, in which case the notion of
difference will be left subordinated to the concept of identity. Different/ciation [the
actualization of the virtual] expresses simultaneously the compossibility of the
‘elements’ inside the virtual and the divergence of the series in which the virtual is
actualized.
Both Bergson and Deleuze present their own versions of this theory, Deleuze going
farthest in developing it via his method of ‘transcendental empiricism’, whereby la durée
divides and differentiates according to the process of what he terms ‘different/ciation’.27
I shall come to the force that propels the process of actualization in due course.28
With respect to the relationship between time and consciousness, Deleuze offers the
most concise, detailed, and relevant explanation through his notion of the ‘three syntheses
of time’, and I turn exclusively to his exposition at this point.29 The ﬁrst characteristic
to be noted about la durée is what Deleuze terms the ‘ﬁrst synthesis’ of time. La
durée, as the continuous different/ciation of multiplicities, produces a ‘living present’,
by linking what might commonly be thought of as successive instants into a fusion of
what is conventionally termed past and future. Dividing time into instants would of
ANTHROPOLOGICALTHEORY 8(4)
410
399-429 096646 Hodges (D) 3/11/08 09:06 Page 410
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
course imply a quantitative perspective on time, similar to notions of B-series time prevalent
in the analytic tradition. This ﬁrst synthesis of time is therefore both connective and
contractile, in its joining and fusing of different multiplicities; it is also passive, consisting
of the different levels of organic and inorganic matter as they exist in the universe.
In this sense, it marks a signiﬁcant difference from phenomenological theories of ‘temporality’,
in that its origin lies beyond subjectivity and consciousness, effectively rendering
them possible. However, it also accommodates them, as consciousness, or what Deleuze
terms the ‘active synthesis’, exists ‘within’ the ‘living present’, and is the location where
the lived experience of time – the subject of anthropology – is ultimately registered.
Now, for consciousness, the ‘active synthesis’, to become aware of past and future, it
must clearly access a dimension that is more than the mere continuous differentiation
accorded by the ﬁrst, passive synthesis of time. In accounting for this actuality, what
Boundas (1996: 85) terms ‘the formation of closed, “extended” or “cool” systems inside
the open-ended, intensive chaosmic virtual’, Deleuze calls upon the ‘second synthesis’ of
time. The ‘second synthesis’ is best revealed through confronting various fundamental
paradoxes arising from the insights of la durée, and linked to the resulting implausibility
of certain common-sense notions of, and metaphors for, grasping the workings of
time.
To begin with, following on from my description of la durée as qualitative multiplicity,
one must conclude that the ‘present’, as a discrete spatialization of time, no longer
‘is’, and no longer becomes what is ‘past’ when a new ‘present’ emerges from the future
to replace it. The nature of la durée is against such a conclusion, as it is against the notion
of time as a succession of instants, and one must in effect conclude that the ‘present’ is
not, the ‘living present’ being merely the continuous different/ciation of la durée.
Nevertheless, for consciousness, time continues to pass. How does this happen,
particularly as the past cannot pass once it is already past, and the present cannot be
replaced by another present that takes its place? The conclusion that both Deleuze and
Bergson draw is that the present must pass at the same time as it is present. This
apparently profound contradiction in terms of western common-sense notions of time,
and linguistic metaphors, is Deleuze’s ﬁrst paradox, which he terms the ‘paradox of
contemporaneity’: ‘the contemporaneity of the past with the present it was’ (1994: 81).
As Deleuze (1991: 59) writes:
The past would never be constituted if it did not coexist with the present whose past
it is. The past and the present do not denote two successive moments, but two
elements which coexist: One is the present, which does not cease to pass, and the
other is the past, which does not cease to be but through which all presents pass.
This then gives rise to a second paradox, what Deleuze terms the ‘paradox of coexistence’,
and which he expresses in similarly complex language intended, one imagines,
to undermine common-sense temporal terminology. ‘If each past is contemporaneous
with the present that it was, then all of the past coexists with the new present in relation
to which it is now past’ (Deleuze, 1994: 81–2). Or, in layman’s terms, the only place the
past can exist is in the present – it has nowhere else to be – even though the present is
not ‘present’, of course. This generalized past, therefore, does not exist in actuality, but
is the virtual form of the past, accessible through varied practices of remembering. With
HODGES Rethinking time’s arrow
411
399-429 096646 Hodges (D) 3/11/08 09:06 Page 411
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
respect to consciousness, it is this virtual form of the past and its actualization through
acts of remembering that is the precondition for our perception of the present as
‘passing’, and indeed our perception of the past itself, acting as a ground, the past ‘in
general’, against which consciousness can perceive la durée’s, and ultimately ‘time’s’ work.
It is the existence of this pure, virtual form of the past, the a priori past, which Deleuze
terms the ‘second synthesis’ of time.
We are now close to completing this elucidation. But ﬁrst there is a further, third
paradox to consider. Following on from the paradox of contemporaneity, where the past
is contemporary with the present it was, Deleuze points out that, as a result, we can no
longer conceive of the past as ‘past’, since it never actually came ‘after’. On the contrary,
in considering the second synthesis of time, the existence of the a priori past, Deleuze
writes: ‘each past is contemporaneous with the present it was, the whole past coexists
with the present in relation to which it is past, but the pure [virtual] element of the past
in general pre-exists the passing present’ (Deleuze, 1994: 82, my emphasis). Deleuze’s
third paradox is therefore entitled the ‘paradox of pre-existence’ – and one should note
that it evidently works to further challenge the most fundamental metaphors many
languages use for temporal perception.
A summary of these complex propositions will now pave the way for some preliminary
conclusions.The ﬁrst synthesis of time forms the ‘living present’, and is that passive,
connective synthesis which lays the foundation for existence – effectively the physical
ground for the universe as we know it. Within this ﬁrst synthesis we ﬁnd the ‘active’
synthesis (consciousness), which can then differentiate between past and future due to
the second synthesis of time, once again a passive synthesis comprising the existence of
the a priori past. But there is still, of course, the ‘third synthesis’ of time to be accounted
for. This is the driving force of different/ciation, indeed it consists of the operation of
different/ciation itself, in which ‘[t]ime splits into two heterogeneous dissymmetrical
emissions, one toward the future, making the present pass, and another toward the past,
coexisting wholly with the present it was’ (Turetzky, 1998: 217). It is the eruption of a
future that subordinates the actualized to its novelty. Bergson grounds the differentiating
quality of la durée in what he called the élan vital; Deleuze in a more creditable
Nietzschean eternal return which, as Ansell Pearson (2002: 200) points out, ‘does not
speak of a return of the same but only of difference’. We might equally read this as a
metaphor for the vital momentum that physicists locate in the well-springs of matter,
or in the mysterious operations of sub-atomic particles.
It is important to note that, despite an apparent similarity between time as la durée
and the notion of ﬂuid time identiﬁed earlier – which characteristically spatializes time
using metaphors such as the ‘current of time’ – Deleuze’s reworking of Bergson’s theory
is fundamentally more speciﬁc. First, with respect to la durée itself, which categorically
has no ‘direction’, and is non-chronological in nature; second, regarding the paradoxical
nature of the past in the second synthesis; and third, in terms of the integrated
relationship between time and space. Regarding the latter, it must be noted that la durée
is thoroughly spatio-temporal in character, and following philosophers such as Ansell
Pearson (1999, 2002) and Boundas (1996: 94–9), it is potentially compatible with
dominant contemporary social, and scientiﬁc theories, and paradigms in modern physics
and biology regarding energy, matter, or genomics.30 It is composed of all the seemingly
‘continuous’, but in effect different/ciating, pulsations of organic and inorganic matter
ANTHROPOLOGICALTHEORY 8(4)
412
399-429 096646 Hodges (D) 3/11/08 09:06 Page 412
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
that constitute the universe and life in it – the multiple strands of bifurcating materiality
which, for anthropologists, are of interest in their ethnographic guise.31 Continuity
in this respect is reduced to the perceived occurrence of ‘habit’ within processes, as with
respect to la durée no repetition is ever ‘the same’; habit itself is continuously open to
the eruption of novelty; and different/ciation ultimately triumphant.32 As for the
notion of process itself, in ontological terms it comprises the myriad ‘sequences’ or
‘lines’ of actualization within la durée that evince contingent interrelationships while
encompassing different/ciation.
And so one can propose that la durée is the manifold substance of ‘history’ itself. In
this respect, evoking and interpreting signiﬁcant ‘habitual processes’ or ‘lines’ of actualization
within it – of short duration or longue durée – of relevance to selective, smallscale
domains of human activity, with the precise yet ultimately restricted tools of human
representation – is what the anthropologist’s task evidently comprises.33 One can thus
begin to grasp how the relational concept of la durée is a philosophical cousin to anthropological
concepts such as Tonkin’s (1992: 72–5) theory of ‘co-existent times’, Thomas’s
(1996: 121) notion of history as a ‘systemic process’, or Ingold’s proposition that
anthropological subjects are integrated in ‘the unfolding of a total relational ﬁeld’ (1990:
225); the multi-layered perspectives on history that have developed from the work of
the Annales school and related debates (e.g. Braudel, 1989, 1990; Gurvitch, 1964); or
Adam’s (1998) theory of timescape. Similarly, one can infer how la durée might underpin
models of sociality and socio-historical practice informed by the principally metaphorical
notions of ﬂuid time as advocated by Bourdieu, Giddens, or Munn, with ‘time’
experience now a central, dynamic dimension. Finally, we can note parallels with
approaches in anthropology which argue that social scientiﬁc understanding is an
emergent, relational process of conceptual and narrative creation (Strathern, 1988;
Viveiros de Castro, 2002; Wagner, 1981; cf. Henare et al., 2007: 13); or posthumanist
perspectives in the sociology of science, that seek to establish paradigms grounded in an
explicit theory of temporal emergence (Latour, 2007; Pickering, 1995; Rheinberger,
1997). With this multi-layered, relational ontology of historical time before us, we shall
now return to work reviewed on the anthropology of time earlier, and extend it
critically in the light of the model just outlined.34
RE-THINKINGTHE ANTHROPOLOGY OF ‘SPACETIME’
Of the studies addressed, our objections to them are now clearer. Both Gell’s model and
the theory of la durée allow for articulation of A-series and B-series time. And Gell’s
assessment of ‘real time’ from one angle appears similar to that of Bergson and Deleuze:
both are inaccessible in a direct, objective sense to human perception. But, importantly,
Gell considers ‘real time’ to nonetheless ‘reﬂect the temporal [dated] relationships
between events as they really are’ (Gell, 1992: 165), a position fundamentally incompatible
with the phenomenon of la durée, which consists of pure different/ciation, and
to which Gell’s abstracted notion of dated relationships is clearly inapplicable. In this
respect, Gell and Mellor rely to a much greater extent on western ‘common-sense’
conceptions of time than Bergson and Deleuze. Indeed, the latter two provide a challenging
but workable non-chronological framework that ultimately affords a distinctive
vantage-point on cross-cultural temporal perspectives – though it is still, of course,
grounded in western (French) cultural terminology. If our different models are broadly
HODGES Rethinking time’s arrow
413
399-429 096646 Hodges (D) 3/11/08 09:06 Page 413
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
compatible, it is precisely Gell’s emphasis on the temporal, datable nature of B-series
‘real time’ that constitutes our most notable point of disagreement. When worked
through, however, such general differences evidently become signiﬁcant. Gell seeks to
identify dated ‘real time’ as the basis for the study of human time experience; I seek to
establish la durée as a basis for the study of lived experience and sociality, as we shall see.35
La durée, as I have shown, provides a model of non-chronological spacetime as
different/ciation, and allows for insertion into this model a notion of the intersubjective
nature of human experience that articulates with it through the three syntheses. In
anthropological terms, this is accordant with a critique of linear time (cf. Greenhouse,
1996); and it enables the adoption of an anthropological apparatus inﬂuenced by
phenomenology as advocated by Munn, and its development in novel ways through a
consideration of the relationship of lived experience to its concrete, non-human context.
By contrast with Munn, therefore, and in keeping with Deleuze’s approach, lived experience
lies in a complex relationship with la durée, which encloses and enables it while
remaining, in the last instance, inaccessible to objective human representation. This
reveals the partially determined, but inherently relational, cultural and individualized
nature of sociality and lived experience, as la durée underpins human existence and the
physical conditions which shape it; while the experience, appropriation and representation
of ‘time’ reside largely in the domain of everyday practice (Lambek, 2002; Munn,
1992), neurological, cognitive and embodied processing (Bloch, 1998; Damasio, 2000:
219–26) and, ultimately, individual ‘moments of being’ (Chodorow, 1999: 216–17,
Ortner, 2005; Rapport and Overing, 2000: 260–1). And in this respect it should also
be noted that to conceptualize these processes in language is a task that can complement
and quantify, but not represent the complex qualitative experience of la durée that constitutes
our lives – as Bergson and Deleuze have made clear, conceptual thought can only
constitute a spatialization of la durée.36
Turning now to rough out the implications for anthropological models, one must ﬁrst
note that the process of different/ciation which lies at the root of what is commonly
conceptualized as ‘time’ must sit squarely at the foundation of any notion of sociality or
historical practice. If this places sociality in a broader, multi-layered, relational context,
it simultaneously underlines that the temporal is its correlate – indeed, in this respect,
terms such as sociality and temporality can effectively be proposed to be interchangeable
(cf. Hirsch and Stewart, 2005a: 262–3; Munn, 1992: 116). This prompts several
observations. As a general statement, we can propose that any considered anthropological
analysis should integrate discussion of historical processes with consideration of
the lived experience of time – each potentially reconceived on an ontological level as
emergent features of la durée. With regard to sociality, however, it has frequently been
the case that the notion of temporality has been evoked by anthropologists as a means of
focusing explicitly on the temporal dimension of social life. One must therefore stress
that all such ‘discrete’ temporal experience should be reconceived as a dynamic feature
of la durée fully integrated with other pertinent historical process. In this sense, we are
extending to its logical conclusion what has always been latent in notions such as sociality,
which along with other concepts ending in -ality was conceived as a result of the
shift towards viewing human experience as constituted intersubjectively ‘in time’ during
the 1980s (cf. Hirsch and Stewart, 2005a: 262). That said, it should be noted that while
the majority of contemporary anthropologists employ the terminology of constitutive
ANTHROPOLOGICALTHEORY 8(4)
414
399-429 096646 Hodges (D) 3/11/08 09:06 Page 414
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
ﬂux, and related concepts such as sociality and habitus, along with paradigms rooted in
processual and historical analysis – fewer engage in explicit analyses of how the temporal
is concretely experienced and dynamically implicated in everyday practice. So the precise
task then becomes to elide the marginal disciplinary status of the anthropology of time,
arguing on the basis of an explicit temporal ontology for the centrality of integrated
discussion of lived time and historical process to all social analysis; and ultimately,
perhaps, for the need to imagine sociality in terms of different/ciation.
Evidently, therefore, adoption of la durée produces a re-evaluation of the anthropology
of time. For on the one hand, one can still address varied notions and experiences
of ‘time’, ‘history’, ‘past’, ‘present’, ‘future’, ‘tempo’, ‘time reckoning’ – those enduring
topics of the sub-discipline now underwritten by the ‘structure’ of la durée – as they
emerge in different historical contexts.37 But on the other, one must deﬁnitively collapse
any discrete notion of ‘lived time’, or ‘temporality’, into an inclusive notion of sociality
conceived as an emergent element of la durée – that is to say sociality would be melded
with a dynamic, regrounded notion of lived time. If I therefore raise the prospect of
collapsing the anthropology of ‘time’ back into ‘all the other dimensions and topics
anthropologists deal with in the social world’ (Munn, 1992: 93) – which was a ﬂaw in
Munn’s eyes – I am now doing so conscious of the fact that all dimensions of sociality
require analysis of their contingent, dynamic orientation towards la durée.38
With respect to the charge of totalization that can be levelled at the notion of ﬂuid
time, we should also note that la durée is open to the many divergent and convergent,
multi-layered features of human experience. One of the further problems, of course, with
a metaphorical terminology that treats of currents and ﬂows and ﬂuxes but remains illdeﬁned
is that the poetic suggestiveness of such language lends a totalizing form to
interpretations of everyday practice. Through deconstructing and redeﬁning some of
these metaphors, we are now well-placed to grasp how, with reference to la durée, they
can be complexly conceived so as to implicate different/ciation – in other words, there
is no direction in which ﬂux or process is moving, and there is no one river of time that
ﬂows. In this sense, la durée retains a universal applicability while remaining free of the
totalizing impulse identiﬁed earlier (cf. Osborne, 1995: 28). It comprises an integrative,
emergent pluralism.39 To the extent that it is possible, it also presents a foundation for
cross-cultural study that is wholly non-chronological, avoiding any obvious resemblance
to western common-sense notions of time – and even the hint of a spatialized model of
time as linear. La durée therefore provides one precise and speciﬁc deﬁnition of what
anthropologists might understand by the notion that lived experience is grounded in an
emergent, ﬂux-like historical process or total relational ﬁeld (Ingold, 1990; Thomas,
1996) – while obviating the need to try and intricately articulate such an anthropological
approach with the plethora of competing theories about bio-physical spacetime from
other ﬁelds. In this respect, we might note its apparent compatibility with ‘complexity
theory’ and dominant paradigms in the philosophy of biology (cf. Delanda, 2005;
Goodwin, 2001; Serres and Latour, 1995: 57–62).
However, what is not yet explicit from this enquiry is whether ultimately one is
culturally ‘trapped’ within even such a complex notion of ﬂux – indeed if there is an
outside perspective that will afford a different vision of lived experience; or if the nonhuman
features of la durée, in the guise of ‘physical time’, are perhaps a cultural construct
it is possible to side-step, as certain anthropologists have tried unsuccessfully, to my
HODGES Rethinking time’s arrow
415
399-429 096646 Hodges (D) 3/11/08 09:06 Page 415
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
mind, to suggest (e.g. Greenhouse, 1996). In this respect, some writers have proposed
that current theoretical preoccupations with change and ﬂuidity are a symptom of our
contemporary ‘runaway’ civilization – modernity, postmodernity – and in the case of
anthropologists, the preoccupation of an intellectual elite who for the main part are
themselves based in an intensely ﬂuid environment, the modern western university.40 In
the same way that Jameson has written of the impossibility of imagining new, ‘postindividualistic’
life-worlds, it may therefore be argued that it is only possible to think
beyond ﬂux in a future in which our current historical conditions – of rapid historical
change and rupture – have been surpassed.41 My proposition, however, is that la durée
offers a precise perspective on ﬂux and change that allows one to clarify the relationship
between historical change that anthropologists habitually curtail and deﬁne ethnographically,
and the different/ciating ﬂux of la durée, which is all-pervasive and provides
an explicit spatio-temporal ontology with which to ground practice methodologies.
Flux, in this sense, is not the unconscious mirror of our ﬂuid times, not coupled loosely
to historical change, but a distinct, wider-ranging concept, of greater purchase, with the
potential for precise cross-cultural applicability. It might be said to inherently invite both
novelty at an ontological level, and variation in practices, through the indeterminate
process of the actualization of the virtual (cf. Ansell Pearson, 2002: 2–3). But it exists
in no causal or enabling relationship with rapid historical change. Indeed, different/
ciation can be said to ontologically underpin so-called ‘traditional’ societies just as much
as the turbulent globalized environments of modernity. The concrete temporal dynamics
and character of social change, therefore, is a matter for contingent elucidation, rather
than being indexed to the temporal ontology of la durée. We now have a novel, nuanced
temporal perspective on what Giddens (1979: 210) might mean when he states that ‘the
possibility of change is . . . inherent in every circumstance of social reproduction’.
At this point, therefore, one might signal a ﬁnal, signiﬁcant, if logical innovation in
analytical procedures, to acknowledge this fuller integration between previously distinct
analytical considerations of sociality and temporality. The extent to which the activity
of temporalization, and the verb to temporalize, can be effectively elided into any process
of symbolization in everyday practice has become a point of contention. In this respect,
and given the queries outlined in this article, I have challenged the use of temporality
and related terms as analytical tools to explore the experience of ‘time’ in everyday practices.42
When writing of everyday practice, following Bergson and Deleuze, one might
therefore use actualization and the verb to actualize – which fully implicate the
virtual–actual axis – in place of common terms such as temporalization; or even, with
further qualiﬁcation, symbolization. While this image of human sociality as a partlydetermined,
fully socio-temporal practice reﬂexively open to historical analysis is abbreviated,
and would beneﬁt from development at length in relation to ethnographic
materials, it provides one workable philosophical foundation for future research in this
area, and incorporates a vision of the relationship between analysis and ﬂux that could
be further developed. In particular, one can argue that the centrality of temporal experience
to sociality demands a precise set of tools that enable consistent and sustained
analysis of the temporal modalities and outlooks of everyday practice.This might involve
a more nuanced attention to everyday indigenous experiences of change, continuity,
epoch and rupture, alongside more conspicuous ‘local traditions’, and processes of
objectiﬁcation and codiﬁcation of history practised by those in power (including anthroANTHROPOLOGICALTHEORY
8(4)
416
399-429 096646 Hodges (D) 3/11/08 09:06 Page 416
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
pologists); novel theorization of ‘conﬂicts in time’ within the politics of lived experience;
and elucidation of an emergent concept of ‘living traditions’ where the process of
different/ciation can be seen to be ‘transcended’ and afforded graded cultural continuity
(cf. Harris, 1996: 13; Lambek, 2002: 273; MacIntyre, 1981: 206–7). Indeed, ethnographic
work problematizing the handling of time experience and emergence in
anthropological models is already in progress (e.g. Hirsch and Stewart, 2005b; Lambek,
2002; Mosko and Damon, 2005), and this article has focused on clarifying the
philosophical and ontological basis for such a project. In this respect, we can conclude
by endorsing la durée as a candidate for the temporal ontology of this more temporallyaware
social science.
Notes
1 Quid est ergo tempus? Si nemo ex me quaerat, scio; si quaerenti explicare velim, nescio . . .
(‘What then is time? If no one asks me, I know; if I wish to explain it to him who
asks, I know not . . .’) in St Augustine, Confessions: Book 11, Chapter XIV. Compare
Derrida (1994: 6): ‘Time . . . gives nothing to see. It is at the very least the element
of invisibility itself. It withdraws whatever could be given to be seen. It itself withdraws
from visibility. One can only be blind to time, to the essential disappearance
of time even as, nevertheless, in a certain manner, nothing appears that does not
require and take time.’
2 Reviews by Gell (1992: 3–148) and Greenhouse (1996: 19–83) draw comparable
conclusions.
3 It would be difﬁcult to do more than summarize in an article of this length the use
such theorists make of this notion, which is what I do here. Nevertheless, this is
concordant with the point I am making, namely that they themselves do not provide
explicit deﬁnitions of time.
4 See Bourdieu (1977: 9):
The detemporalizing effect (visible in the synoptic apprehension that diagrams
make possible) that science produces when it forgets the transformation it
imposes on practices inscribed in the current of time, i.e. detotalized, simply by
totalizing them, is never more pernicious than when exerted on practices deﬁned
by the fact that their temporal structure, direction, and rhythm are constitutive
of their meaning. (Emphasis in original)
5 As Wolf (1982: 4) writes in his very ﬁrst sentence: ‘The central assertion of this book
is that the world of humankind constitutes a manifold, a totality of interconnected
processes, and inquiries that disassemble this totality into bits and then fail to
reassemble it falsify reality.’
6 The term ‘temporal ontology’ applies to the implicit or explicit theory of the nature
of time and temporal experience underpinning a paradigm, which is necessarily
metaphysical in character, and an inescapable component of any social theory – a
sociological counterpart to what Bakhtin (1981), for example, in referring to texts,
termed a chronotope.
7 A well-known exception is Bourdieu’s (1977: 97–109, 1963) analysis of the agrarian
calendar. However, while offering characteristic insights, he still subordinates the
HODGES Rethinking time’s arrow
417
399-429 096646 Hodges (D) 3/11/08 09:06 Page 417
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
discussion of a more general approach to time to his desire to utilize the calendar to
illustrate the workings of habitus, and in this respect his emphasis on time as constitutive
obviates the possibility of examining more intricately the foundations for the
human experience of time. See Gell (1992: 263–305) for a detailed appraisal of
Bourdieu’s approach to time. Thomas (1996: 5–6) also analyses the temporal
dynamics of such theories, but does not press home an argument for the theorization
of time itself (i.e. the provision of a temporal ontology), indicating rather that
his own emphasis is on the ﬁeld of ‘historical time’ – an elision that I would wish
to query.
8 For example O’Brien and Roseberry (1991). Taussig (1987) offers an inﬂuential and
aggressive critique of the ‘Euro-centric’ features of the temporal outlook of Wolf
(1982), Mintz (1985), and like-minded political economists, which played its part
in prompting these more reﬂexive approaches. His approach is complemented by
Ortner (1984).
9 As Giddens (1979: 210) states: ‘[I]n the replacement of the synchrony/diachrony
opposition with a conception of structuration, the possibility of change is recognized
as inherent in every circumstance of social reproduction’. It is a statement that
sounds eminently unproblematic until one considers more closely its temporal
ontology; and the temporal implications of widespread concepts of globalization and
social change are also widely unclariﬁed.
10 Given that ﬂow, ﬂux and process are used so freely by social analysts, strictly speaking,
in dictionary terms, all imply change. According to Chambers, ﬂowing is immediately
deﬁned as ‘to move or change form like a ﬂuid’; ﬂux as ‘the act of ﬂowing’;
process as ‘a sequence of operations or changes undergone’; and change, of course, as
‘to pass from one state to another’. While these are not technical deﬁnitions, my
argument is precisely that neither are these words used, on the whole, in a technical
sense by social analysts.
11 In this respect, pre-Socratic philosophers such as Heraclitus have become a common
touchstone for postmodernists (he was also a favourite of Heidegger), with assertions
such as the following: ‘For it is not possible to step twice into the same river, according
to Heraclitus, nor to touch mortal substance twice in any condition; by the swiftness
and speed of its change, it scatters and collects itself again – or rather, it is not
again and later but simultaneously that it comes together and departs, approaches
and retires.’ (From Plutarch [c. AD 46–127], On the E at Delphi 392B, quoted in
Barnes, 1987: 117.)
12 Heise suggests that postmodernist writers often present exaggerations or satirical
depictions of what writers such as Harvey (1990: 284–307) have characterized as
‘postmodern’, ‘post-Fordist’, time-space compression. In relation to the temporal
ontologies of other ﬁelds of western intellectual production, therefore, postmodern
anthropological thought arguably bears comparison with modernist trends, a suggestion
that acquires some weight if one considers, for example, the preoccupation with
ﬂux in modernist literature (cf. Stevenson, 1998: 87–158); and even, in the work of
political economists, reﬂects aspects of 19th-century historical realism (cf. Lukács,
1972) – which is perhaps not surprising given the Marxian genealogy of such models.
13 This would also appear to be a feature of literary criticism, particularly of the
novel. See Currie (2007) on how attention to the philosophy of time has been largely
ANTHROPOLOGICALTHEORY 8(4)
418
399-429 096646 Hodges (D) 3/11/08 09:06 Page 418
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
overlooked in this discipline, even among writers focusing on time itself (e.g. Heise,
1997).
14 Terdiman (1993) provides a historical analysis of relations between modernity, ﬂuid
time, and modernist and postmodernist representation that touches on such issues.
15 See Gell (1992: 149–74 and 221–41) for the B-series and A-series theories of time
and his presentation of the relations between them.
16 Agamben (1993) provides a concise philosophical critique of linear time (see also
Berger, 1984 and Merleau-Ponty, 1989). For Agamben,
[s]ince the human mind has the experience of time but not its representation, it
necessarily pictures time by means of spatial images . . . The modern concept of
time is a secularization of rectilinear, irreversible Christian time, albeit sundered
from any notion of end and emptied of any other meaning but that of a
structured process in terms of before and after. (1993: 91, 96)
His critique is particularly aimed at the Marxist use of history:
Modern political thought has concentrated its attention on history, and has not
elaborated a corresponding concept of time . . . The vulgar representation of time
as a precise and homogeneous continuum has thus diluted the Marxist concept
of history: it has become the hidden breach through which ideology has crept
into the citadel of historical materialism. (Agamben, 1993: 91)
Thompson (1967), in a well-known analysis, locates the secularization and spread
of this model of time to the growth of industrial capitalism.
17 If you introduce Gell’s image of B-series time as ‘dated’ into the types of statement
quoted from Bourdieu, Giddens or Wolf, earlier in this article, you begin to grasp
its incongruous character in the anthropological context. In this respect, the apparently
‘natural’ character of metaphors of ﬂow, ﬂuidity and so on, devoid of overt
cultural references to time, suggests they are less glaringly problematic to deploy and
so have attracted less attention, in the cross-cultural context, despite bearing their
own set of problems.
18 This position has led Munn in her earlier ethnographic work to err towards a
position similar to the A-series ‘reductionism’ championed by Greenhouse, although
this does not appear to be the case in her 1992 essay. Incidentally, Adam (1990,
1998) has been the theorist most concerned to articulate social theory with insights
from physical and ecological science. Her work makes for fascinating reading and is
particularly insightful regarding time ecology, but is more limited regarding the
apparatus required for an analysis of the everyday human experience of time.
19 The problem, clearly, is one that has affected symbolic anthropology in more general
terms, and has received much attention in other areas of anthropological theory,
concerning the rapprochement and articulation between materialist and symbolic
perspectives (e.g. Foster, 1995; O’Brien and Roseberry, 1991; Parmentier, 1987;
Roseberry, 1989).
20 Bergson and Deleuze’s approaches are also comparable to other theories of time in
the ‘distaff’ tradition: most notably, Nietzsche’s notion of eternal recurrence, in
which time ‘is neither an aspect of change nor an absolute extension; it is a
HODGES Rethinking time’s arrow
419
399-429 096646 Hodges (D) 3/11/08 09:06 Page 419
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
constitutive intensity’ (Turetzky, 1998: 114); or the Stoics’ notions of Chronos and
Aion (cf. Turetzky, 1998: 38–42), which Deleuze develops elsewhere (Deleuze, 1990
[1969]). One can also invoke broad parallels with key works in ‘process’ philosophy
such as Whitehead (1979) (Clark, 1999, analyses the relationship between Deleuze
and Whitehead); or with writers in the American pragmatist tradition, particularly
James (1996) on ﬂux, stream of consciousness and radical empiricism (as have been
drawn for other thematic contexts by Rorty, 1989, 1991). There are also parallels to
be noted with work in the phenomenological tradition, such as Levinas’ (1989
[1946]) concept of the il y a. Bergson and Deleuze, however, are particularly enlightening
for their efforts to correlate human experience (‘A-series’ time) with nonhuman
material processes (‘B-series’ time), and for having most thoroughly worked
through the implications of integrating ‘ﬂux’ into analysis. For related reasons, their
work has been the subject of much attention by philosophers in recent years, which
have also seen a revision of Bergson’s reputation (cf. Ansell Pearson, 1999, 2002;
Mullarkey, 1999).
21 Respectively translated as Matter and Memory (1988) and Creative Evolution (1944).
22 Deleuze’s three books are available in translation as Bergsonism (1991 [1966]), The
Logic of Sense (1990 [1969]) and Difference and Repetition (1994 [1968]).
23 The presentation therefore takes a pragmatic, selective approach, and makes modest
claims to represent the full complexity of this multiform theory. Deleuze alone, for
example, presents several versions, in each case with novel and complex jargon, and
anyone familiar with his work will also be aware of its occasionally convoluted
nature. I should also note that for the sake of intelligibility I retain the term la durée,
which Deleuze dispenses with in his later work. Disclaimers aside, however, my
exposition elucidates the main components of this approach to time with precision.
For critiques of Bergson, see Sartre (1956: 192–4), Merleau-Ponty (1989: 413–4),
and Bachelard (1950). For a related critique of Deleuze, see Wyschogrod (1990:
189–229) and for a reply, Boundas (1992). For a brief overview of critiques of
Bergson and Deleuze, and counter-critique, see Boundas (1996: 98–103).
24 See Bergson (1960, 1988), Deleuze (1991: ch. 2), Turetzky (1998: chrs 13 and 14),
and Boundas (1996: 82–5). ‘Multiplicity’ is the established translation of the French
‘multiplicité’ in this literature, although English language mathematicians would use
the term ‘manifold’.
25 ‘In the terms of another discourse, actualization is individuation, the creation of
singularity (whether physical, psychical, or social), insofar as the processes of
individuation predate the individual yet the individual is a somehow open-ended
consequence of these processes’ (Grosz, 1999: 27).
26 Grosz (1999: 28) writes:
Duration proceeds not by continuous growth, smooth unfolding, or accretion,
but through division, bifurcation, dissociation – by difference – through sudden
and unexpected change or eruption. Duration is a mode of infecting selfdifferentiation:
difference is internal to its function, its modes of elaboration and
production, and is also its ramifying effect on those objects located ‘within’ its
milieu. This means that not only must concepts of time (in physics, biology,
philosophy, cultural studies, and social theory) be opened up to their modes of
ANTHROPOLOGICALTHEORY 8(4)
420
399-429 096646 Hodges (D) 3/11/08 09:06 Page 420
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
differentiation, but also that our very concept of objects, matter, being . . . needs
to be open to the differentiations that constitute and continually transform it.
27 Deleuze is invoking the difference in French between différencier, to make or become
different, and différentier, which means the same but in a mathematical context. As
Boundas (1996: 91) glosses: ‘Different/ciation refers to the complex relations
between problems and solutions, questions and answers, virtual Idea-structures and
their actualizations. Deleuze calls “differentiation” the totality of the diacritic
relations which occur “inside” an Idea-structure, and “differenciation”, the process
of actualization of such a structure.’
28 Ansell Pearson (2002) provides a lengthy elucidation of the key concept of the
virtual which, he writes, ‘presents an ontological challenge to our ordinary conceptions
of perception and memory, of time and subjectivity, and of life in its evolutionary
aspect . . . [T]he virtual is by nature something intrinsically vague and
indeterminate [but] wholly real and the real is, in fact, unencounterable and
unthinkable without it’ (2002: 2–3). See also Shields (2002).
29 See Deleuze (1994: ch. 2). The following exposition also draws on Deleuze (1991),
Turetzky (1998: ch. 14), and Boundas (1996).
30 Ansell Pearson (2002: 66) writes:
The single time common to all times is the time of duration, where duration is
conceived as a virtual multiplicity . . . My claim is that this is not simply the time
of the philosopher. This is why it is important that duration qua a virtual multiplicity
is not restricted to the solely psychological or phenomenological but also
encompasses the vibrating rhythms of matter.
This ‘single time’, or ‘open whole’, is also referred to by Deleuze (1986: 59) as the
plane of immanence (see Ansell Pearson, 2002: 41–2 for a succinct discussion of this
‘pluralist monism’).
31 As Grosz writes (1999: 17):
[La durée] functions simultaneously as singular, uniﬁed, and whole, as well as in
speciﬁc fragments and multiplicitous proliferation. There is one and only one
time, but there are also numerous times: a duration for each thing or movement,
which melds with a global or collective time. As a whole, time is braided, intertwined,
a unity of strands layered over each other; unique, singular, and individual,
it nevertheless partakes of a more generic and overarching time, which makes
possible relations of earlier and later.
It is at this point, however, that I would advocate a return to a regrounded anthropological
perspective and theoretical apparatus. My approach therefore stops short
of adapting for anthropology Deleuze’s more well-known philosophical work with
Guattari (e.g. Deleuze and Guattari, 1984, 1988) – which was translated in the
1970s and 1980s, and whose fashionable status in postmodern debates has tended
to obscure the more signiﬁcant work contained in Difference and Repetition, and on
Bergson more generally. Indeed, since its publication in English in 1994, Difference
and Repetition has increasingly been recognized as Deleuze’s distinctive contribution
HODGES Rethinking time’s arrow
421
399-429 096646 Hodges (D) 3/11/08 09:06 Page 421
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
to modern continental philosophy. For philosophical discussion of how la durée
might be concretely imagined, see Boundas (1996: 96–7), and Deleuze (1991:
103–5). See also Boundas (1996, note 48) for further references.
32 As Deleuze writes: ‘we have no other continuities apart from those of our thousands
of habits’, yet ‘[h]abit draws something new from repetition – namely difference’
(1994: 94, 93).
33 Bergson (1988) and Deleuze (1994) both explicate how conceptual discourse can
only constitute a ‘spatialization’ and mystiﬁcation of la durée, if frequently giving
rise to utilitarian and successful ways of manipulating it.
34 It is worth emphasizing that this new deﬁnition renders the core tropes of ﬂuid time
– with their ﬂuvial overtones – redundant. Indeed there is a case for eliding the
terminology of ‘ﬂux’ and ‘process’ as well, given their association with common-sense
notions (time, as we have evoked it, is more of a ‘pulse’). But let us opt instead to
redeﬁne this terminology, to enable a convergence between anthropological and
philosophical notions.
35 Gell (1992: 317–19) regrettably gives short shrift to Bergson, whom he classiﬁes
erroneously as an A-theorist. He does not engage with Deleuze’s philosophy of time,
which was only translated into English in the 1990s, and not widely debated outside
French philosophical circles while he was researching The Anthropology of Time.
36 Merleau-Ponty offers an interesting opening here for a durée-inﬂected aesthetics of
anthropological writing. In Signs (1964), he suggests there is a distinction between
what he terms the ‘primary expression’ of corporeal communication, or the language
of the senses, and the ‘secondary expression’ of conceptual assertion. The latter, he
asserts, is founded on extension and translation of the former. Extending primary
expression to include ‘lived experience’ as it is manifested in la durée, secondary
expression can therefore be understood as the conceptualization of this lived
experience that occurs ‘within’ it. Artistic expression, Merleau-Ponty proposes, has
a privileged access to primary expression that conceptual language does not, through
its ‘tacit and implicit accumulation of meaning rather than by abstracting meaning
into a pure state of clarity’ (Kearney, 1994: 80). Anthropology is therefore ideally
placed among disciplines to evoke the various dimensions of lived experience in la
durée, through combinations of artistic and conceptual expression.
37 To a degree, I am in agreement here with writers such as Barth (1987) or Scott
(2007) who argue that ‘we must always struggle to get our ontological assumptions
right: to ascribe to our object of study only those properties and capabilities that we
have reasonable ground to believe it to possess’ (Barth quoted in Scott, 2007: 3,
emphasis retained), so as to avoid ‘the artiﬁcial production of rubrics . . . or recourse
to isolated socio-cultural phenomena as the topical foci for local and cross-cultural
study’ (Scott, 2007: 3). The anthropology of time should of course be grounded
primarily in contextualized lived experience, and thus reference how la durée is
manifested ethnographically. Nevertheless, the latter’s temporal ontology is not open
to cultural remodelling, however local people might seek to conceive of it (which is
of course of primary interest) – or social scientists, for that matter. In this sense, from
the perspective of the anthropology of time, an addendum to Barth’s formula might
run: ‘we must always struggle to make our own ontological assumptions explicit’ (cf.
Barth, 1987: 8; Scott, 2007: 3).
ANTHROPOLOGICALTHEORY 8(4)
422
399-429 096646 Hodges (D) 3/11/08 09:06 Page 422
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
38 Hirsch and Stewart (2005a: 271) reach a related conclusion in their discussion of
‘historicity’, in some ways a terminological equivalent to ‘temporality’: ‘Historicity
in this anthropological or ethnographic sense is a concomitant of sociality.’
39 For extended discussion of how Deleuze’s approach admits a fundamental pluralism
within the apparently unifying concept of la durée, see Ansell Pearson (2002, ch. 4).
40 Berger (1984) develops such a thesis, for example, suggesting that ‘[h]istory . . . no
longer speaks of the changeless but, rather, of the laws of change which spare
nothing’ (1984: 12). It is a theme pursued in the extensive literature on modernity
and postmodernity, particularly among writers of a Marxian persuasion, and in
this respect is also present in Marx and Engels’ (1952 [1848]) Manifesto of the
Communist Party.
41 Jameson (2002: 111–12) writes:
For Marxism, indeed, only the emergence of a post-individualistic social world,
only the reinvention of the collective and the associative, can concretely achieve
the ‘decentering’ of the individual subject . . . only a new and original form of
collective social life can overcome the isolation and monadic autonomy of the
older bourgeois subjects in such a way that individual consciousness can be lived
– and not merely theorized – as an ‘effect of structure’.
Chodorow (1999) provides an effective rejoinder to such assertions of cultural determinism;
and indeed Berger himself (note 39, this article) is arguably confounding
talk about time and social change with a sense of the inevitability and annihilating
power of secular death (as Heideggerians and existentialists also do in their different
ways, cf. Osborne, 1995: 69–112) which is not to say that his observations do not
have ethnographic applicability.
42 For Fabian (1983: 73), for example, temporalization
connotes an activity, a complex praxis of encoding Time. Linguistically, temporalization
refers to the various means a language has to express time relations.
Semiotically, it designates the constitution of sign relations with temporal
referents. Ideologically, temporalization has the effect of putting an object of
discourse into a cosmological frame such that the temporal relation becomes
central and topical (e.g. over and against spatial relations). Finally, temporalizing,
like other instances of speech, may be a deictic function. In that case a temporal
‘reference’ may not be identiﬁable except in the intention and circumstances of
a speech-act.
It is the proposition that a ‘temporal relation’ is discretely identiﬁable that, once
brought into question, problematizes this, and ultimately any approach that wishes
to demarcate ‘time’ or ‘temporalization’ as a discrete ﬁeld of enquiry; and arguably
reinforces the marginality of the study of ‘time experience’.
HODGES Rethinking time’s arrow
423
399-429 096646 Hodges (D) 3/11/08 09:06 Page 423
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
References
Adam, B. (1990) Time and Social Theory. Cambridge: Polity Press.
Adam, B. (1998) Timescapes of Modernity: The Environment and Invisible Hazards.
London and New York: Routledge.
Agamben, G. (1993) Infancy and History: Essays on the Destruction of Experience.
London and New York: Verso.
Ansell Pearson, K. (1999) Germinal Life: The Difference and Repetition of Deleuze.
London: Routledge.
Ansell Pearson, K. (2002) Philosophy and the Adventure of the Virtual: Bergson and the
Time of Life. London: Routledge.
Bachelard, G. (1950) La Dialectique de la Durée. Paris: Les presses universitaires de
France.
Bakhtin, Mikhail (1981) The Dialogical Imagination. Austin: University of Texas
Press.
Barnes, J. (1987) Early Greek Philosophy. London: Penguin.
Barth, F. (1987) Cosmologies in the Making: A Generative Approach to Cultural Variation
in Inner New Guinea. Cambridge: Cambridge University Press.
Berger, J. (1984) And Our Faces, My Heart, Brief As Photos. London: Writers and
Readers.
Bergson, H. (1944 [1907]) Creative Evolution. New York: Random House.
Bergson, H. (1960) Time and Free Will: An Essay on the Immediate Data of
Consciousness. New York: Harper and Row.
Bergson, H. (1988 [1896]) Matter and Memory. New York: Zone Books.
Bloch, M. (1998) How We Think They Think: Anthropological Approaches to Cognition,
Memory, Literacy. Oxford: Westview.
Borges, Jorge Luis (1970 [1941]) ‘The Garden of Forking Paths’, in Jorge Luis Borges
Labyrinths, pp. 44–54. Harmondsworth: Penguin.
Boundas, C.V. (1992) ‘Gilles Deleuze: The Ethics of the Event’, in D. Goicoechea and
M. Zlomislic (eds) Joyful Wisdom, pp. 169–99. St Catharine’s, Ontario: Thought
House.
Boundas, C.V. (1996) ‘Deleuze-Bergson: An Ontology of the Virtual’, in P. Patton
(ed.) Deleuze: A Critical Reader, pp. 81–106. Oxford: Blackwell Publishers.
Bourdieu, P. (1963) ‘The Attitude of the Algerian Peasant Toward Time’, in
J. Pitt-Rivers (ed.) Mediterranean Countrymen, pp. 55–72 (trans. G.E. Williams).
Paris and the Hague: Mouton.
Bourdieu, P. (1977 [1972]) Outline of a Theory of Practice. Cambridge: Cambridge
University Press.
Braudel, F. (1989) The Identity of France. Volume One: History and Environment.
London: Fontana Press.
Braudel, F. (1990) The Identity of France. Volume Two: People and Production. London:
Fontana Press.
Chambers English Dictionary (1990) Edinburgh: W. and R. Chambers Ltd.
Chodorow, N.J. (1999) The Power of Feelings: Personal Meaning in Psychoanalysis,
Gender, and Culture. New Haven, CT and London: Yale University Press.
Clark, T. (1999) ‘A Whiteheadian Chaosmos: Process Philosophy from a Deleuzean
Perspective’, Process Studies 28(3–4): 179–94.
ANTHROPOLOGICALTHEORY 8(4)
424
399-429 096646 Hodges (D) 3/11/08 09:06 Page 424
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
Currie, M. (2007) About Time: Narrative, Fiction, and the Philosophy of Time.
Edinburgh: Edinburgh University Press.
Damasio, A. (2000) The Feeling of What Happens: Body, Emotion, and the Making of
Consciousness. London: Vintage.
Delanda, M. (2005) Intensive Science and Virtual Philosophy. London: Continuum
Books.
Deleuze, G. (1956a) ‘Bergson (1859–1941)’, in M. Merleau-Ponty (ed.) Les philosophes
celebres, pp. 292–9. Paris: Mazenod.
Deleuze, G. (1956b) ‘La Conception de la différence chez Bergson’, Études
bergsoniennes 4: 77–112.
Deleuze, G. (1986) Cinema 1: The Movement-Image. London: Athlone Press.
Deleuze, G. (1990 [1969]) The Logic of Sense. New York: Columbia University Press.
Deleuze, G. (1991 [1966]) Bergsonism. New York: Zone Books.
Deleuze, G. (1994 [1968]) Difference and Repetition. London: Athlone Books.
Deleuze, G. and F. Guattari (1984) Anti-Oedipus: Capitalism and Schizophrenia.
London: Athlone Press.
Deleuze, G. and F. Guattari (1988) A Thousand Plateaus: Capitalism and Schizophrenia.
London: Athlone Press.
Derrida, J. (1994) Given Time: 1. Counterfeit Money. Chicago, IL: Chicago University
Press.
Durkheim, E. (1915) The Elementary Forms of the Religious Life. London: Allen &
Unwin.
Evans-Pritchard, E.E. (1939) ‘Nuer Time Reckoning’, Africa 12(2): 189–216.
Evans-Pritchard, E.E. (1940) The Nuer. Oxford: Clarendon.
Fabian, J. (1983) Time and the Other: How Anthropology Makes its Object. New York:
Columbia University Press.
Faulkner, William (1993 [1937]) Absalom, Absalom! (The Corrected Text). New York:
Modern Library.
Foster, R.J. (1995) Social Reproduction and History in Melanesia: Mortuary Ritual, Gift
Exchange, and Custom in the Tanga Islands. Cambridge: Cambridge University Press.
Geertz, C. (1973) ‘Person, Time, and Conduct in Bali’, in C. Geertz, The
Interpretation of Cultures, pp. 360–411. New York: Basic Books.
Gell, A. (1992) The Anthropology of Time: Cultural Constructions of Temporal Maps and
Images. Oxford: Berg Books.
Giddens, A. (1979) Central Problems in Social Theory: Action, Structure and
Contradiction in Social Analysis. London: The Macmillan Press.
Giesen, B. (1992) ‘The Temporalization of the Social Order: Some Theoretical
Remarks on the Change in “Change”’, in H. Haferkamp and N.J. Smelser (eds)
Social Change and Modernity, pp. 294–319. Berkeley and Los Angeles: University of
California Press.
Goodwin, B. (2001) How the Leopard Changed its Spots: The Evolution of Complexity.
Princeton, NJ: Princeton University Press.
Greenhouse, C. (1996) A Moment’s Notice: Time Politics Across Cultures. Ithaca, NY and
London: Cornell University Press.
Grosz, E. (1999) Becomings: Explorations in Time, Memory, and Futures. Ithaca, NY and
London: Cornell University Press.
HODGES Rethinking time’s arrow
425
399-429 096646 Hodges (D) 3/11/08 09:06 Page 425
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
Gupta, A. and J. Ferguson, eds (1997) Culture, Power, Place: Explorations in Critical
Anthropology. Durham, NC and London: Duke University Press.
Gurvitch, G. (1964) The Spectrum of Social Time. Dordrecht: D. Reidel Publishing
Company.
Harris, O. (1996) ‘The Temporalities of Tradition: Reﬂections on a Changing
Anthropology’, in V. Hubinger (ed.) Grasping the Changing World: Anthropological
Concepts in the Postmodern Era, pp. 1–16. London: Routledge.
Harvey, D. (1990) The Condition of Postmodernity. Oxford: Blackwell Publishers.
Heidegger, M. (1993 [1927]) Being and Time. Oxford: Blackwell Publishers.
Heise, U.K. (1997) Chronoschisms: Time, Narrative, and Postmodernism. Cambridge:
Cambridge University Press.
Henare, A., M. Holbraad and S. Wastell (2007) ‘Introduction: Thinking through
Things’, in A. Henare, M. Holbraad and S. Wastell (eds) Thinking through
Things: Theorising Artefacts Ethnographically, pp. 1–31. London and New York:
Routledge.
Hirsch, E. and C. Stewart, C. (2005a) ‘Introduction: Ethnographies of Historicity, in
“Ethnographies of Historicity”’, Special Issue of History and Anthropology 16(3):
261–74.
Hirsch, E. and C. Stewart, eds (2005b) ‘Ethnographies of Historicity’, Special Issue of
History and Anthropology 16(3).
Husserl, E. (1966 [1887]) The Phenomenology of Internal Time-Consciousness.
Bloomington and Indianapolis: Indiana University Press.
Ingold, T. (1990) ‘An Anthropologist Looks at Biology’, Man (New Series) 12(2):
208–29.
James, W. (1996) Essays in Radical Empiricism. Lincoln and London: University of
Nebraska Press.
James, W. and D. Mills, eds (2005a) The Qualities of Time: Anthropological Approaches.
Oxford: Berg Books.
James, W. and D. Mills (2005b) ‘Introduction: From Representation to Action in the
Flow of Time’, in W. James and D. Mills (eds) The Qualities of Time:
Anthropological Approaches, pp. 1–15. Oxford: Berg Books.
Jameson, F. (2002) The Political Unconscious: Narrative as a Socially Symbolic Act.
London and New York: Routledge.
Kahn, C.H., ed. (1981) The Art and Thought of Heraclitus. Cambridge: Cambridge
University Press.
Kearney, R. (1994) Modern Movements in European Philosophy: Phenomenology, Critical
Theory, Structuralism. Manchester: Manchester University Press.
Kirk, G.S., J.E. Raven and M. Schoﬁeld (1983) The Presocratic Philosophers.
Cambridge: Cambridge University Press.
Lambek, M. (2002) The Weight of the Past: Living with History in Mahajanga,
Madagascar. New York and Basingstoke: Palgrave Macmillan.
Latour, Bruno (2007) Reassembling the Social: An Introduction to Actor-Network Theory.
Oxford: Oxford University Press.
Leach, E. (1961) Rethinking Anthropology. London: Athlone Press.
Levinas, E. (1989 [1946]) ‘There is: Existence without Existents’, in E. Levinas The
Levinas Reader, pp. 29–36. Oxford: Blackwell Publishers.
ANTHROPOLOGICALTHEORY 8(4)
426
399-429 096646 Hodges (D) 3/11/08 09:06 Page 426
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
Lévi-Strauss, C. (1963) Structural Anthropology I. New York: Basic Books.
Lukács, G. (1972) Studies in European Realism. London: Merlin Press.
MacIntyre, A. (1981) After Virtue: A Study in Moral Theory. London: Duckworth.
McTaggart, J.E. (1908) ‘The Unreality of Time’, Mind (New Series) 17(68): 457–74.
Malinowski, B. (1927) ‘Lunar and Seasonal Calendar in the Trobriands’, Journal of the
Royal Anthropological Institute of Great Britain and Ireland 57(January–June):
203–15.
Marx, K. and F. Engels (1952 [1848]) Manifesto of the Communist Party. Moscow:
Progress Publishers.
Mellor, D.H. (1981) Real Time. Cambridge: Cambridge University Press.
Merleau-Ponty, M. (1964) Signs. Evanston, IL: Northwestern University Press.
Merleau-Ponty, M. (1989 [1945]) Phenomenology of Perception. London: Routledge.
Mintz, S. (1985) Sweetness and Power: The Place of Sugar in Modern History. New York:
Viking Press.
Mosko, M. and F. Damon, eds (2005) On the Order of Chaos: Social Anthropology and
the Science of Chaos. New York and Oxford: Berghahn Books.
Mullarkey, J., ed. (1999) The New Bergson. Manchester: Manchester University Press.
Munn, N. (1983) ‘Gawan Kula: Spatiotemporal Control and the Symbolism of
Inﬂuence’, in J.W. Leach and E. Leach (eds) The Kula: New Perspectives on Massim
Exchange, pp. 277–308. Cambridge: Cambridge University Press.
Munn, N. (1986) The Fame of Gawa: A Symbolic Study of Value Transformation in a
Massim (Papua New Guinea) Society. Durham, NC and London: Duke University
Press.
Munn, N. (1992) ‘The Cultural Anthropology of Time: A Critical Essay’, Annual
Review of Anthropology 21: 93–123.
O’Brien, J. and W. Roseberry (1991) Golden Ages, Dark Ages: Imagining the Past in
Anthropology and History. Berkeley: University of California Press.
Ong, A. and S.J. Collier (2004) Global Assemblages: Technology, Politics and Ethics as
Anthropological Problems. Oxford: Blackwell Publishing.
Orlove, B.S., ed. (2002) Repertoires of Timekeeping in Anthropology, Special Issue of
Current Anthropology 43(S4).
Ortner, S.B. (1984) ‘Theory in Anthropology since the Sixties’, Comparative Studies in
Society and History 26(1): 126–66.
Ortner, S.B. (2005) ‘Subjectivity and Cultural Critique’, Anthropological Theory 5(1):
32–52.
Osborne, P. (1995) The Politics of Time: Modernity and Avant-Garde. London and New
York: Verso.
Parmentier, R.J. (1987) The Sacred Remains: Myth, History and Polity in Belau.
Chicago, IL: University of Chicago Press.
Pickering, A. (1995) The Mangle of Practice: Time, Agency and Science. Chicago, IL:
Chicago University Press.
Proust, M. (1999) Time Regained (In Search of Lost Time Vol. 6. New York: Modern
Library.
Pynchon, T. (1973) Gravity’s Rainbow. London: Picador.
Rapport, N. and J. Overing (2000) Social and Cultural Anthropology: The Key Concepts.
London and New York: Routledge.
HODGES Rethinking time’s arrow
427
399-429 096646 Hodges (D) 3/11/08 09:06 Page 427
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
Rheinberger, H.-J. (1997) Toward a History of Epistemic Things: Synthesizing Proteins in
the Test-Tube. Stanford: Stanford University Press.
Rorty, R. (1989) Contingency, Irony, and Solidarity. Cambridge: Cambridge University
Press.
Rorty, R. (1991) Essays on Heidegger and Others. Philosophical Papers, Vol. 2.
Cambridge: Cambridge University Press.
Roseberry, W. (1989) Anthropologies and Histories: Essays in Culture, History, and
Political Economy. New Brunswick, NJ and London: Rutgers University Press.
Sartre, J.-P. (1956) Being and Nothingness: An Essay in Phenomenological Ontology. New
York: Pocket Books.
Schutz, A. (1962) The Problem of Social Reality. The Hague: Martinus Nijhoff.
Scott, M. (2007) The Severed Snake: Matrilineages, Making Place, and a Melanesian
Christianity in Southeast Solomon Islands. Durham, NC: Carolina Academic Press.
Serres, M. and B. Latour (1995) Conversations on Science, Culture, and Time. Ann
Arbor: University of Michigan Press.
Shields, R. (2002) The Virtual (Key Ideas) London: Routledge.
Stevenson, R. (1998) Modernist Fiction. Hemel Hempstead: Prentice Hall.
Strathern, M. (1988) The Gender of the Gift: Problems with Women and Problems with
Society in Melanesia. Berkeley, Los Angeles and London: University of California
Press.
Taussig, M. (1987) ‘The Rise and Fall of Marxist Anthropology’, Social Analysis 21:
101–13.
Terdiman, Richard (1993) Present Past: Modernity and the Memory Crisis. Ithaca, NY
and London: Cornell University Press.
Thomas, E.J. and G. Hjort, eds (1996) Early Buddhist Scriptures. New Delhi: Aryan
Books International.
Thomas, N. (1996) Out of Time: History and Evolution in Anthropological Discourse
(second edition). Ann Arbor: University of Michigan Press.
Thompson, E.P. (1967) ‘Time, Work-Discipline, and Industrial Capitalism’, Past and
Present 38(December): 56–97.
Tonkin, E. (1992) Narrating our Pasts: The Social Construction of Oral History.
Cambridge: Cambridge University Press.
Turetzky, P. (1998) Time. London and New York: Routledge.
Viveiros de Castro, E. (2002) ‘O nativo relativo’, Mana 8(1): 113–48.
Wagner, R. (1981) The Invention of Culture. Chicago, IL and London: University of
Chicago Press.
Whitehead, A. (1979 [1927–8]) Process and Reality. New York: The Free Press.
Wolf, E. (1982) Europe and the People without History. Berkeley and Los Angeles:
University of California Press.
Woolf, V. (2000 [1925]) Mrs Dalloway. London: Penguin.
Wyschogrod, E. (1990) Saints and Postmodernism. Chicago, IL: Chicago University
Press.
ANTHROPOLOGICALTHEORY 8(4)
428
399-429 096646 Hodges (D) 3/11/08 09:06 Page 428
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from
MATT HODGES focused his doctoral research on temporality and social change in a southern French village;
subsequently he carried out ﬁeldwork in the Basque Country on time and sustainability. Publications have
concerned time, modernity, social change and historical rupture, heritage tourism, and have included ethnographic
ﬁction. He has lectured at the Department of Anthropology, Goldsmiths College, where he was also
Visiting Fellow, and was AHRC Fellow in the Creative and Performing Arts in the School of Literature and
Creative Writing at the University of East Anglia, where he worked on a public anthropology project translating
ﬁeldwork experience into literary ﬁction. He is now Research Fellow in the Department of Sociology
and Philosophy, University of Exeter. Address: Egenis, Department of Sociology and Philosophy, University of
Exeter, Byrne House, St Germans Rd, Exeter, EX4 4PJ, UK. [email; m.hodges@ex.ac.uk]
HODGES Rethinking time’s arrow
429
399-429 096646 Hodges (D) 3/11/08 09:06 Page 429
by Csaba Szalo on November 13, 2008http://ant.sagepub.comDownloaded from