reconceptualizing senses of place: social relations ... · 1997: reconceptualizing senses of place:...

RECONCEPTUALIZING SENSES OF PLACE: SOCIAL RELATIONS, IDEOLOGY AND ECOLOGY

Geografiska Annaler · 79 B (1997) · 1

1


1

byDavid Butz and John Eyles

Butz, D. and Eyles, J.,

1997: Reconceptualizing senses of place:social relations, ideology and ecology.

Geogr. Ann

. 79 B (1): 1Ð25.

ABSTRACT. The purpose of this paper is to reconceptualizesense of place through the examination, re-analysis and theoriza-tion of two case studies, one of an ex-urban community in Eng-land, the other a Himalayan farming and herding community.The paper begins by examining the traditional locus of sense ofplace research in humanistic geography with extensions to polit-ical geography and interpretive anthropology. Identifying threecore componentsÑsocial, ideological and ecologicalÑof sensesof place, the paper goes on to reconceptualize these elements us-ing HabermasÕs theory of communicative action and IngoldÕswork on environmental psychology. It then applies this recon-ceptualization to the case studies of Towcester and Shimshal.The paper concludes by emphasizing the ways these cases enrichour understanding of sense of place, by stressing the theoreticalcontributions of conceiving sense of place as rooted in theoriesof social organization and society, and as being variably and con-tingently ecologically emplaced.

Introduction

The purpose of this paper is to explore the natureof sense of place. It does this in three ways. First,it locates research on senses of place in the human-istic geographical literature, although it sees this asbeing extended in the political geography realm byexplorations of the intersection of place and pol-itics. This leads to a full explanation of a theory ofsenses of place and their social, ideological andecological dimensions through HabermasÕs (1984)theory of communicative action and IngoldÕs(1992) exploration of direct perception, the latterallowing for an explicit rendering of the ecologicalas well as the more usually conceptualized socialand ideological. Second, it explores senses of placethrough a process of initial analysis of two casestudies (the one building on the otherÑShimshalon Towcester), engagement with theoretical de-bates, reßection and re-analysis. Through this proc-ess, the results from different methodological ap-proachesÑconstruction of ideal types and the livedexperience of ÞeldworkÑare engaged and used toinform one another. Third, the paper presents, com-parative analysis of senses of place in these two dis-

tinct geographical and cultural cases or settings, theÞrst a commuter town in southern England (Tow-cester), the second a village and pastoral commu-nity in the Pakistan high Karakoram (Shimshal).What brings the two cases together is the use of asimilar interpretive framework for understandingsense of placeÑdeveloped in Towcester, reappliedin Shimshal. As with all comparative investiga-tions, such case selection allows a movement be-yond description to account for the phenomena ofinterest (cf. Durkheim, 1938). Such investigationsallow generalizations to be made, resulting fromthe interpretation, side by side, of different groups,collectivities, institutions and environments. AsHeclo (1972, p. 95) comments, Òto speak to com-parative analysis suggests not only that one will belooking at variables which actually vary, but alsothat one will be doing so in contexts which them-selves vary.Ó It is only through such comparativeanalysis that one can appreciate which are trulyunique and what are the more generic phenomena.

Thus it is our intention to examine the nature ofsense of place, empirically derived and refractedthrough a series of theoretic and reßective lensesfrom two very different contexts. In this way, it ispossible to begin to isolate potentially generaliz-able features of the phenomenon of interest fromthe study of the characteristics embedded in thecases themselves. But before the cases are present-ed, we wish Þrst to isolate the signiÞcance of senseof place in the geographical literature, seeing it asemerging as a given in humanistic geography butbeing necessarily grounded in both theories ofcommunicative action and research on the eco-logical basis of cultural attachment to place. Weshall conclude with a commentary on how our casestudies inform a theoretical understanding ofsenses of place.

Senses of place and humanistic geography

As Cloke

et al

. (1991) graphically express it in theirreview of approaches in human geography, the hu-

DAVID BUTZ AND JOHN EYLES


2

manistic turn ÒpeoplesÓ the discipline. The turn al-lows for a focus on place and the experience ofplace by people through phenomenological and ex-istential arguments (Relph, 1970; Seamon, 1979).Place always implies a sense of place even if thatexperience was not particularly pleasant. Thus,Relph (1985, p. 26) argues that place (and sense ofplace) are qualitatively different from that of land-scape or space. ÒThe latter are part of any immedi-ate encounter with the world, and so long as I cansee I cannot help but see them no matter what mypurpose. This is not so with places, for they are con-structed in our memories and affections through re-peated encounters and complex associations.ÓPlace is where one is known and knows others.Sense of place involves sense of being.

Before any choices there is this ÒplaceÓ,where the foundations of earthly existenceand human condition establish themselves.We can change locations, move, but this is stillto look for a place; we need a base to set downour Being and to realize our possibilities, a

here

from which to discover the world, a

there

to which we can returnÓ.(Dardel, 1952, p. 56; translated byRelph, 1985, p. 27)

Such sentiment is also expressed by anthropolo-gists who suggest that consciousness of the worldbeyond place is the catalyst for the recognition ofoneÕs own community as a distinct entity (Cohen,1982). Places are thus seen as centres of felt value(Tuan, 1977), centres of experience and aspirationsof people (Tuan, 1976). To be attached to a placeis an important human need, perhaps the least rec-ognized one (Weil, 1955). Place is a profound cen-tre for human existence (Relph, 1976), importantfor identity of the individual (with the group) (Dun-can, 1973). It is of course not the only basis of iden-tity or attachment but it provides a grounding forother dimensions beyond the household. Even inthe electronic era (Meyrowitz, 1985), sense ofplace-boundedness is strong (Pred, 1983).

Place identity remains strong even if the attach-ment is not positive (Hummon, 1992), even if it isdisrupted (Brown and Perkins, 1992) and at differ-ent scales from dwelling to region (Cuba and Hum-mon, 1993). As Cuba and Hummon (1993) note,place identity, as expressed by a sense of feeling-at-home, is widespread, rich in its attachment tomultiple locales and complex in spatial structuresand in its determination. Indeed, in a review of ten

deÞnitions of place, Brown and Perkins (1992, p.284) conclude that place attachments are integralto self-deÞnitions; they provide stability and non-threatening changes; they are holistic and multi-faceted and multilevel. Brown and Perkins arguedthat Òplace attachment involves positively experi-enced bonds, sometimes occurring without aware-ness, that are developed over time from the behav-ioral, affective and cognitive ties between individ-uals and/or groups and their sociophysical environ-ment.Ó

Sense of place underpins sense of (well)being.This is noted by Foucault (1980, p. 70): ÒA critiquecould be carried out of this devaluation of space[

sic

] that has prevailed for generations... to tracethe forms of implantation, delimitation and demar-cation of objects, the modes of tabulation, the or-ganization of domains meant throwing into reliefprocessesÑhistorical ones, needless to sayÑofpower.Ó The denial of sense of place, criticized byFoucault, is taken up by Harvey (1982) in his de-scription of the homogeneous nature of places thatare all subject to the logic of the economy (al-though Harvey modiÞes his position somewhat inlater works). Thrift (1987) and Entrikin (1991) takeissue with this approach, arguing that subjectsmake-up and are identities localized even if theÒmaterialÓ is not all local in origin (cf. essays inDuncan and Ley, 1993; Cosgrove and Daniels,1988; Gregory and Walford, 1989; Philo, 1991;Keith and Pile, 1993; Pile and Thrift, 1995). Fur-ther, the argument that political alignments havecrystallized largely around national social differ-ences to produce national patterns of political mo-bilization and partisan support has also been chal-lenged. For example, Dunleavy (1979) points to theimportance of the social and the local in shaping in-terest perceptions and value formations to charterthe growth of Òconsumption cleavagesÓ in urbanpolitical alignments. Agnew (1987) examines theimportance of the localÑsocial, historic and per-ceptualÑin shaping the patterns of support forScottish nationalism.

From such studies the signiÞcance of the localÑof placesÑemerges. But it must be recognized thatplaces have both individuality and interdepend-ence. This is superbly expressed by Entrikin (1991,p. 134), addressing the apparent divide betweenplaces as existence and places in nature:

The closest that we can come to addressingboth sides of this divide is from a point in be-tween, a point that leads us into the vast realm



3

of narrative forms. From this position we gaina view from both sides of the divide. We gaina sense both of being Òin a placeÓ and Òat a lo-cationÓ, of being at the centre and being at apoint in a centreless world. To ignore eitherargument of this dualism is to misunderstandthe modern experience of place.

He goes on to make a spirited defence for the sig-niÞcance of place, proclaiming its basis for Òcom-munityÓ and signiÞcance for democracy (its nor-mative signiÞcance), its utility as a ÒmodelÓ ineveryday understandings about the world (its epis-temological signiÞcance), and arguing that it is ofcontinued relevance in a mobile world to ÒplaceÓpeople (its empiricalÑtheoretical signiÞcance). Insome respects, EntrikinÕs assertions parallel Ag-newÕs (1987) three dimensions of place: location(the spatial distribution of activitiesÑthe impact ofthe wider world on place), locale (the setting inwhich social relations are constitutedÑakin toÒcommunityÓ) and sense of place (place attach-ment and the structure of feelings that are used inthe everyday).

For us, place is not usurped by other discoursessuch as those that emphasize the non-place self andthe non-place community. Walter (1988, p. 97) ar-gues that the dominance of FreudÕs thought helpsunground the self:

Freud moved theory of the mind away fromgrounded experience and helped to build thecouch as a vehicle abstracting patient fromplace. Despite his own existential recognitionof the inner need for place, FreudÕs psycho-logy never integrated personal identity withthe sense of belonging, and the real power ofplace.

We may add that this power of places was furtherattenuated as the mass of people in industrial soci-ety came to reside in urban places. Despite our in-terdependence, we are individualized, blas� aboutinteractions, calculating about events and integrat-ed around the self rather than signiÞcant others.SimmelÕs (1950) description of the mental life ofthe metropolis and WirthÕs (1964) of urbanism asa way of life encapsulate this world of independent,isolated individuals living in densely settled, het-erogeneous settlements where only we know ouridentity. These arguments also have relevance forthe Òloss of communityÓ and hence non-place com-munities.

Agnew (1987) in fact argues that there are twostages to the devaluation of sense of place as asigniÞcant cognitive and social structure. First, itstems from the ambiguity in the language of com-munity, in which the term is used to describe bothphysical setting and a morally valued way of life(Nisbet, 1966; Calhoun, 1980; Agnew 1989). T�n-niesÕs (1957) ideas of the transition from a place-based community to a place-less or national societylink this stage to the second devaluation of place,namely the eclipse of community and with it, byimplication, places as ÒhistoryÓ. This allows for theevolution of society into associations of interde-pendent but autonomous workers and consumers.In modern society, individuals become dissociatedfrom place as social networks and interaction makegeography less relevant than in the past. Institu-tions, not places, guide interactions (Stacey, 1969),and it is accessibility rather than propinquity thatis the important spatial referent (cf. Webber, 1964).While it is not possible to deny the importance ofinstitutions or accessibility, Tilly (1973, p. 236) ex-presses our concerns well. He argues that placeshave persisted in importance but there has been arelative decline in such localized communities asthe bases of collective action: Òlocal ties have di-minished little or not at all, extra local ties have in-creased.Ó Further, local ties and sense of locale inthe past may have been political rather than social,as the Lee

et al.

(1984) study of Seattle demon-strates. And while the inßuence of national, massphenomena cannot be doubted (cf. Pahl, 1970),they are just as likely to stimulate dissimilar behav-iour by individuals in distinct places as similar ac-tions (cf. Claggett

et al

., 1984). From this litera-ture, we note that place and our sense of place can-not be taken for granted and must be grounded.How might that grounding occur?

Much recent work in the ÒnewÓ cultural and so-cial geography locates this groundingÑand the ar-gument for contextualization is now surely wellmadeÑin psychoanalytic and/or poststructuralist(i.e. anti-humanist) notions of Òthe selfÓ, especiallyin the connections among space, place, subjectivityand identity (cf. Pile, 1993; Keith and Pile, 1993;Pile and Thrift, 1995). Those who follow a psycho-analytic model of subjectivity see a personÕsgrounding in place as rooted largely in the uncon-scious (cf. Thrift, 1993), while poststructuralist ap-proaches understand subject positions and placesto be constituted and linked within and through dis-courses. These approaches provide legitimategrounds from which to critique both the existence



4

and emancipatory potential of shared place atti-tudes and attachments, arguing that such place-ori-ented communities of sentiment repress differenc-es within, and exclude differences from outside (cf.Young, 1990). While we recognize the contribu-tions of this strand of socio-cultural toward recon-ceptualising place, we nevertheless wish to devel-op our analysis in a direction that recognizes thecontinuing salience of community in

our

subjectsÕlives.

Social and ideological dimensions of senses of place

The grounding of human beings in place may beexplored in the context of studies of ÒcommunityÓin which sense of place often resides. Eyles iden-tiÞes three salient elements of community: Òplaceor area, people and their institutions, and sense ofbelonging, which helps enrich our notion of placeÓ(1985, p. 63). He suggests that the concept of com-munity, as constituted in its three attendant ele-ments:

can provide insights into the importance androle of place in social and material life. It is inthis respect that the three aspects of com-munity-place, people and mindÑare takenand discussed [in terms of] community as ec-ological structure, social structure and ideo-logical structure respectively.

(Eyles, 1985, pp. 63Ð64)

In this section we want to explore the relationshipsbetween sense of place and each of these aspects,or dimensions, of community. The

social

com-ponent provides the basic material for everyday lifein a community. Community consists, largely, ofgroups of individuals and their relationships withone another. This social life does not necessarily in-volve place. However, place necessarily locates ac-tivities and has meaning as an area for social activ-ities or for the expression of sentiments. Thus, plac-es are often constituted by the people who live inthem. The conjoining of people and places leads tothe latterÕs constitution as matrices of symbolswhich comprise the

ideological

component ofcommunity and place. Ideological structure consti-tutes community as an expression of collective sen-timent and as a device for the protection and pro-motion of sectional interests. Matrices of symbolspertaining to places can engender a sense of be-longing and identity; individuals identify with a

place, and feel they belong to it, because they sharesocial values and sentiments with others in thatplace. The place comes to represent a set of sharedvalues. This place-based sense of community ex-ists in the mind, but is not a product of the mindalone. The mental representation is based on theenvironmental, social and material conditions inwhich the individual is located. We address the re-lationship between social and ideological aspectsof community through a discussion of HabermasÕstheory of communicative action, in which (1) col-lective agency toward intersubjective understand-ing is understood as occurring Òin placeÓ, and isfounded on shared elements of an ÒemplacedÓ lifeworld; and (2) the symbolic constitution of placeis itself conceived as materially and socially con-structed. We then attempt to root

ecologically

therelationship between sense of place and everydaylife, by incorporating a discussion of work in cul-tural ecology.

Community, whatever its other characteristics,is Þrst and foremost a set of social and cultural re-lations. It basically consists of a set of social inter-actions grounded in shared meanings, values andinterests. HabermasÕs (1984) distinction betweeninstrumental and communicative action helps toconceptualize that link between the social and theideological. In his theory of communicative action,Habermas argues that members of speech commu-nities (a concept that incorporates a more conven-tional notion of community) are occupied contin-uously in two types of agency. The Þrst is gearedtoward speciÞc instrumental or technical out-comesÑto get to work, to make money, to meetpeople, to extract produce from pastures, etc.Ñwhich he calls

instrumental action

. Habermas callsthe second

communicative action

, and describes itas a continuous struggle to understand one another,to negotiate a set of common meanings, to reach anintersubjective understanding. Communicative ac-tion occurs as individuals challenge and eventuallyaccept the legitimacy of each anotherÕs arguments,in terms of one or more of three validity claims,which Habermas describes as exhaustive and irre-ducible: truth, appropriateness, and authenticity.We are not convinced that these validity claims areeither exhaustive or irreducible, or can ever resultin truly intersubjective understanding (Butz, 1995;Giddens, 1982; Fraser, 1987; Honneth and Joas,1991). Nevertheless, HabermasÕs conceptualiza-tion of these two action-orientations, and especial-ly the relationship between them, is helpful forconceptualizing senses of place.



5

Habermas contends that instrumental and com-municative action are integrally related, in that in-strumental action relies on prior and ongoing com-municative action (Giddens, 1982). It is throughcommunicative action that speech communitiesnegotiate both the rules for decision making, andspeciÞc decisions themselves. The products ofboth negotiations shape instrumental action: theÞrst determines what is accepted as appropriate in-strumental action; the second, what is consideredto be successful action. In this way, Habermas con-ceives technical activities as the outcome of instru-mental action, but determined communicatively.At the same time, the technical success of instru-mental action may be used as a claim to validatefurther communicative action.

What makes communicative actionÑthis con-tinual challenging, arguing and validatingÑlessarduous than it sounds, especially in small face-to-face communities, is the existence of an intersub-jectively shared

life world

, which consists for Hab-ermas of a loose and shifting set of non-problem-atic background convictions that provide the foun-dations for what we consider to be true, appropriateand truthful; in industrial societies, routinization(Stehr, 1994). Life world is therefore the symboliccontext for determining the rules of decision mak-ing. The shared assumptions of life world easecommunicative action. But the relationship is notone-way, because communicative action constant-ly validates, alters and reproduces elements of lifeworld by calling the unquestioned into question,and requiring its validation. This occurs as individ-uals attempt to utilize the convictions of life worldto validate their claims. In short, shared life worldunderlies communicative action even as communi-cative action rationalizes life world.

The relationships Habermas posits betweencommunicative action, instrumental action and lifeworld help clarify the ways that place, communityand senses of place are integrated. First, the com-municative efforts of a speech community neces-sarily occur Òsome placeÓ. Places provide the spe-ciÞc sites and larger context for communicative ac-tion as a form of social interaction, and thereforebecome associated with, and to some degree con-stitutive of, that interaction. The reverse is alsotrue. The places where communicative action oc-curs become associated with and constitutedthrough that form of interaction. This is obviouslytrue of face-to-face interaction among members ofa speech community, whether formally in councilchambers or informally over the back fence. How-

ever, more obviously mediatedÑless face-to-faceÑspeech communities are also emplaced inthis way, although not always in the sense of con-tiguous spatial boundedness. An Internet discus-sion group is an extreme example of a speech com-munity without propinquity, the participants ofwhich nevertheless associate it with, and constituteit through, their speciÞc places of access, and per-haps with and through the larger and less tangibleimaginative ÒcyberplaceÓ of the Net where theseindividual places of entry meet. We would arguethat such a conceptualization diminishes the dan-ger that associating sense of place with territoriallybounded community renders sense of place obso-lete in an era of time-space distanciation. Certainly,the notion of community without propinquity re-quires some rethinking of senses of place and theway they are shared, but it does not make senses ofplace less signiÞcant either as an abstract concept,or as constitutive of individualsÕ daily lives. Socialinteraction geared toward intersubjective under-standingÑwhether in contiguous or non-contigu-ous speech communitiesÑis thus integrally asso-ciated with the particular sites of interaction, bothlending those sites signiÞcance and deriving mean-ing from them. In other words, the social processof communicative action, to the extent that it is em-placed, engenders senses of place on a very small-scale, which links the place of social interactionwith the form of interaction, and lends both placeand social interaction signiÞcance.

Second, the ideological contexts (life worlds)that members of a speech community bring to com-municative action, and which they use as the basisfor assessing othersÕ validity claims, are them-selves grounded in the places in which memberslive their lives. Participants in communicative ac-tion associate the norms, attitudes, suppositions,assumptions with which they evaluate othersÕ va-lidity claims with particular environmental/cor-poreal settings. We all live our background convic-tions in place, and they take shape in our minds asguidelines for a material existence, lived in a place.Our practical understanding of the world is rootedin our life places. It is this relationship betweenplace and life world which comprises the core ofsenses of place. Place, to the extent that it is

shared

by members of a speech community (and the co-presence implied by the notion of communicativeaction makes it shared), becomes a basis for com-monality in the life worlds of participants, whichhelps make their validity claims recognizable, tan-gible, indeed real to one another. In many instanc-



6

es, then, shared senses of place can facilitate effortsto achieve intersubjective understanding amongmembers of a speech community.

Third, the relationship between places and so-cial interaction geared toward intersubjective un-derstanding also works in the opposite direction, inthat participants in discourse create and re-createplace symbols as they interact and attempt to un-derstand one another. The process of communica-tive action ensures that life world is as much a so-cial as it is a mental construct: communicative ac-tion rationalizes life world, and in so doing also ra-tionalizes sense of place. In that way, shared sensesof place may be outcomes of communicative ac-tion, as well as constituent elements of it. In short,much of the life world shared by members of ter-ritorially based communities is likely to be boundup in a shared experience of and in place. Peoplewho have a common history in a place are likely toshare certain orientations toward that place whichare reproduced through communicative action, andwhich are represented in things that exist at the in-tersection of local meaning and local knowledge.This implies that senses of place are dynamic andcontingent, not static or originary in any sense.

Fourth, not only is the ideological component ofplace (or senses of place) socially constituted, butso are its material aspects. Communicative actionregulates what can be accepted as legitimate instru-mental action, some of which is geared toward thedeÞnition, use and reproduction of the sets of ma-terial resources that comprise a physical place.Places are constructed symbolically and physicallyas the products of communicative and instrumentalaction respectively. Places are produced materiallyin response to the outcome of communicative ac-tion, which is itself regulated by a place-embeddedlife world. Again, the relationship is two-way, forthe symbolic component of place is materially con-stituted. Place is where life is lived instrumentally(as well as symbolically); it is the corporeal settingfor individual life worlds. And for many speechcommunities it provides a large and tangible (yetlimited) component of what is shared among par-ticipants. Thus, place can provide a common andmaterial foundation for shared elements of lifeworld, and a basis for social interaction throughcommunicative action.

Several summary points emerge which helpmove the sense of place concept beyond its human-ist roots. First, and most obvious, social interac-tion, place and sense of place are mutually consti-tutive. They are all necessarily implicated in one

another, and none of them can be conceived as orig-inary. Second, and following from the Þrst point,senses of place are never purely individual or pure-ly collective. They are never purely individual be-cause life world is always reproduced, negotiatedand rationalized through a social process of com-municative action. Place meanings may be highlyprivate, but they are nevertheless grounded in acommunicatively rationalized life world. Senses ofplace are never purely collective, in the sense thatwe can identify the deÞnitive sense of place of acommunity. All individuals participate to varyingdegrees in numerous speech communities (severalof which may coalesce to approximate what weconventionally conceive as the spatially-boundedÒcommunityÓ), occupy particular subject positionsin each, and bring particular place experiences toeach, so that the discursive constitution of any in-dividualÕs senses of place will overlap with, but notduplicate, that of other individuals. Third, follow-ing from the previous point, an individualÕs sensesof place are unlikely to be stable or unitary. Theyare not stable, just as life world is not stable, be-cause the places and social processes throughwhich they are constituted are continuously chang-ing. They are not unitary, because individuals par-ticipate in numerous speech communities, and oc-cupy several subject positions, all of which suggestdifferent, often overlapping, often contradictory,attitudes toward place. What all of this suggests forthe study of senses of place, is that we abandon allattempts to describe unitary senses of place whichare deÞnitive of the relationship between groups ofpeople and their places. It would seem more fruitfulto view senses of place, both within and among in-dividuals, as necessarily tentative and contingent,particularistic, dynamic, and at least potentiallycontradictory. From that point of departure, thetask becomes one of recognizing commonalitiesand overlaps where they exist, and tracing the so-cial, ideological and ecological conÞgurationswhich account for that overlap. The discussion thusfar provides an adequate sense of the relationshipbetween sense of place and the Þrst two of thoseconÞgurations. It remains to integrate fully an ec-ological component into our understanding ofsenses of place.

Ecological dimensions of senses of place

We wish to develop our treatment of an ecological-ly grounded sense of place around IngoldÕs (1992)critique of cognitivist theories of environmental



7

perception, and his arguments in favour of an alter-native theory of direct perception. First, we willsummarize the main points of IngoldÕs argument.It will be evident that his approach conßicts insome important ways with the discursive model wehave developed so far. Our second task, therefore,is to clarify which parts of IngoldÕs model we wishto reject, and which we think are useful for extend-ing our understanding of senses of place into an ex-plicitly ecological realm.

If senses of place are attitudes toward place, orcultural representations of place, it follows that aneffort to conceptualize individualsÕ senses of eco-logically grounded aspects of place can beneÞtfrom some attention to one of the central questionsof cultural ecology: what is the relationship be-tween culture and ecological setting? (Steward,1955, p. 33). According to Ingold, cultural ecolo-gists have conventionally answered this question inan internally inconsistent way, at once insistingthat (1) all meaning is culturally constructed, and(2) that culture is human beingsÕ means of adaptingto the environment. He describes the essence of thecontradiction as follows:

Cultures, it is supposed, are systems of sym-bols. As meaning-making animals, humansimpose their symbolically constituted designsupon the external world... . If all meaning isthus culturally constructed, then the environ-ment on which it is imposed must originallybe

empty

of signiÞcance. But if we hold thatculture is manÕs [

sic

] means of adaptation tothe environment, and if the environmentÑprior to its ordering through cultural catego-riesÑis mere ßux, devoid of all form andmeaning, it follows that culture is an adapta-tion to nothing at all... . Either we must aban-don the notion, central to ecological anthro-pology, that culture is an adaptive system at-tuned to given environmental constraints, orwe have to abandon the idea that human be-ings inhabit worlds that are themselves cultur-ally constructed.

(Ingold, 1992, p. 39; emphasis inoriginal)

Ingold views this dilemma as stemming from acognitivist approach to environmental perception,which Òerects an impermeable barrier between theÔinterior worldÓ of human subjects and their exte-rior conditions of existenceÓ (Ingold, 1992, p. 40),so that Òwe must know the world before we can act

in it, and knowing consists in the organisation ofsensations impinging upon the passively receptivehuman subject into progressively higher-orderstructures or ÔrepresentationsÕÓ (Ingold, 1992, p.45). According to a cognitivist model, raw percep-tions of environmental characteristics are mean-ingless until organized into cultural categories andrepresentations through a process of cognition. In-gold rejects this cognitivist dichotomy betweensensation and intellect, and argues that Òthere is nodistinction between seeing and Ôseeing asÕÓ (Reed,1987, p. 105). His central claim is that Òit is possi-ble for persons to acquire

direct

knowledge of theirenvironments in the course of their practical activ-itiesÓ (Ingold, 1992, p. 40; emphasis in original).He supports his claim with a reading of GibsonÕs(1979) Òecological psychologyÓ, especially histheory of

direct perception

.The theory of direct perception relies on the no-

tion of environmental

affordances

, which Gibson(1979, p. 127) describes as what an environmentÒ

offers

the animal, what it

provides

or

furnishes

, ei-ther for good or illÓ (Gibson, 1979, p. 127; empha-ses in original) for the consummation of behaviour.These affordances exist as inherent potentials ofenvironmental objects themselves, independent ofwhether or how a subject uses them. Thus, environ-mental objects are not neutral objects waiting forindividuals to assign them meaning; their meaningis in what they afford. Human beings perceive theenvironment as sets of affordances for particularpractical purposes, and Òperceiving is,

ipso facto

,knowingÑto have seen something is to havesought out the information that enables one toknow itÓ (Ingold, 1992, p. 46). Environmentalknowledge is thus essentially practical; it is knowl-edge about what the environment affords. Ingoldacknowledges that human beings have the abilityto step outside this practical engagement with theenvironment as a set of affordances, to view it ab-stractly as ÔnatureÕ, and thus conÞgure it as a sys-tem of neutral objects. However, he does not be-lieve this is the way the environment is perceivedin everyday life: Òno more than other animals canhuman beings

live

in a permanently suspendedcondition of contemplative detachmentÓ (Ingold,1992, p. 44; emphasis in original). Life is lived inengagement, not disengagement, with the environ-ment.

Some central implications of conceiving envi-ronmental perception as direct recognition of a setof affordances are as follows. First, perception isnot a series of discrete sensory events, but rather a



8

continuous process, the continuous outcome ofwhich is a new state of the perceiver. Second, to saythat individuals share an environment is to say thatthey live in a world of shared environmental af-fordances, the perception/recognition of whichthey also share. Third, Òthe process of perceptionis also a process of action: we perceive the worldas, and because, we act in itÓ (Ingold, 1992, p. 45).Fourth, Òthe structures and meanings that we Þndin the world are

already there

in the information weextract in the act of perception; their source lies inthe objects we perceive, they are not added on bythe perceiverÓ (Ingold, 1992, p. 46; emphasis inoriginal).

Which affordances, of the many offered by anenvironment, are the ones individuals perceive, de-pends upon the activities those individuals are en-gaged in and the

effectivities

they have. The termÒeffectivityÓ Òdenotes the action capabilities of theagentÑwhat he or she is practically equipped todoÓÑand is the reciprocal of ÒaffordancesÓ, whichÒare properties of the real environment as directlyperceived by an agent in the context of practical ac-tionÓ (Ingold, 1992, p. 46). Thus Òthe range of af-fordances of an object will be constrained by the ef-fectivities of the subject, and conversely, the effec-tivities of the subject will be constrained by the af-fordances of the objects encounteredÓ (Ingold,1992, p. 46). What this implies is that effectivitiesand affordances are constitutive of one another, andthat this relationship is the basis for what Ingoldcalls the mutual constitution of persons and en-vironment.

In conceptualizing effectivities (which are char-acteristics of the perceiver) as constitutive of af-fordances, Ingold seems to be leaving room for cul-tural and social organization to play an importantpart in the perception of environmental objects.This is not, however, IngoldÕs position. He statesquite categorically that Òwe

discover

meaningfulobjects in the environment by moving about in itand extracting invariants from the continuallychanging optic arrayÓ (Ingold, 1992, p. 47; empha-sis in original). We

do not generate

them culturallythrough language. Nor do we need language forperception to be shared:

The awareness of living in a common worldÑthe communion of experience that lies at theheart of socialityÑdoes not depend on thetranslation of percepts, initially constructedby subjects from sensory data private to them-selves, into the terms of an objective system of

collective representations encoded in lan-guage and validated by verbal agreement... .Sociality is rather given from the start,

prior

tothe objectiÞcation of experience in culturalcategories, in the direct perceptual involve-ment of fellow subjects immersed in joint ac-tion in the same environmentÓ.

(Ingold, 1992, p. 47; emphasis inoriginal)

Indeed, Ingold suggests that cultural categoriesÑclassiÞcation systemsÑare quite unnecessary forthe perception and practical use of environmentalaffordances. They are only necessary for knowinga ÒnatureÓ abstracted from everyday practical life.Rather, Òit is by their action in the world [and notby their classiÞcation of it] that people know it, andcome to perceive what it affordsÓ (Ingold, 1992, p.48). Therefore, language and symbolic thoughtÑcultureÑare necessary not to know the world, butonly to make others aware of that knowledge, toshare it: Òthe cultural construction of the environ-ment is not so much a

prelude

to practical actionas an (optional)

epilogue

Ó (Ingold, 1992, p. 52; em-phazis in original).

IngoldÕs central proposition that human beingsperceive and experience the environment as a set ofaffordances, the recognition of whichÑand indeedthe constitution of whichÑis constrained by the ef-fectivities and action contexts of the perceiving in-dividuals, is a strong foundation from which tobuild a conceptualization of ecologically-ground-ed senses of place. Clearly, however, his assertionthat individuals perceive environmental affordanc-es directlyÑpre-culturally, so to speakÑis incom-mensurate with the discursive emphasis of our the-sis so far. The main failing of IngoldÕs conceptual-ization is that he stresses the direct relationship be-tween human agents and their environment whileignoring a similarly ÒdirectÓ relationship betweenhuman beings and the societies of which they arenecessarily a part, and within which they necessar-ily encounter the environment. IngoldÕs persons areenvironmentally constituted, but apparently not so-cially or culturally constituted. We wish to acceptthe notion of ÒdirectÓ perception of environmentalaffordances, but with the important qualiÞcationthat human perceivers are always already sociallyand culturally constituted (just as in their social in-teractions they are always already environmentallyconstituted). Human beingsÕ perceptions of theirecological environment are in no way originary/pre-cultural acts, because their action contexts and



9

effectivities are necessarily socially and culturallyimplicated. There is always a relationship betweenmaterial perception of the material environmentand social communication about that environment,between in another lexicon, signiÞer and signiÞed,whatever the sign system (cf. Rappoport, 1979;Gottdiener, 1983). Our insistence that perception issocially and culturally grounded is quite differentfrom the cognitivist approach which conceives en-vironmental perceptions as meaningless and Òex-ternalÓ until organised through a discrete cognitiveprocess, and, indeed, seems commensurate withwhat Ingold says initially about the relationship be-tween affordances and effectivities.

In cognitivist models of perception, human be-ingsÕ knowledge of their environment is supposed-ly based initially on abstract thought about that en-vironment. According to a theory of direct percep-tion, environmental knowledge is grounded in im-mediate practical action in that environment.Individuals come to know their environment asthey begin to recognise what it affords for theirpractical purposes; we never just perceive the en-vironment, we always perceive it

as

somethingwhich facilitates or confounds our purposes. In-goldÕs mistake, as we see it, is in assuming that thisperception/knowledge is unmediated. Social andcultural mediation of environmental knowledgeoccurs in at least three ways, all of which derivefrom IngoldÕs own use of the term ÒeffectivitiesÓ,as Òthe action capabilities of the agentÑwhat he orshe is practically equipped to doÓ (Ingold, 1992, p.46). First, individualsÕ effectivities, so deÞned, areconstrained by their practical purposes, which arethemselves socially constructed. Even somethingas innocent as a walk in a forest is a socially con-structed practical purpose which differs from innu-merable other equally socially constructed practi-cal purposes for venturing into the woods. It maynot be necessary to identify by name the path wewalk along, or to represent it to ourselves as one ina typology of similar objects, including sidewalks,streets, canals, etc. But the recognition that it af-fords an appropriate route for our walk is inßu-enced by the pre-existing social conventions of go-ing for a walk, including childhood admonish-ments not to disturb the vegetation, to stay on thepath to avoid getting lost, and so on. Second, itseems unlikely that many of us ever encounter anenvironmental object that has not already been cul-turally inscribed and socially positioned for us, al-though something approaching this may occurwhen known environments are disrupted, and we

see them anew (Lee, 1976; Brown and Perkins,1992); for example, in the aftermath of war(Hewitt, 1994). Thus our effectivities are constitut-ed in part as what our social context allows us torecognize as the affordances of an environment forour practical purposes. For example, how many ofus can survey our world without being aware, at themost fundamental level, of the social context of pri-vate property? There are many environmental af-fordances that human beings do not recognizeÑin-deed, for all practical purposes these affordancesdo not existÑbecause we do not have legitimatesocial access to them. Third, human beingsÕ prac-tical endeavours are often collective; the tasks weset out to accomplish in the environment must oftencoordinate with larger group projects, and are fre-quently set by others. In that case, communicationof environmental knowledge precedes direct per-ception, and the representations and classiÞcationsimplicit in language shape the effectivities that arebrought to the individuals/environment interac-tion, as well as the direct (but mediated) percep-tions that result from that encounter. It seems thatin arguing for a pre-cultural/pre-linguistic knowl-edge of environment Ingold is trying to describehow human beings would interact with the envi-ronment if we did not already exist socially and cul-turally. But we are always already socially and cul-turally constituted, and that constitution shapes theeffectivities we bring to our continuous encounterswith environmental objects. This somewhat alteredconception of effectivities does not make environ-mental perception any less direct than Ingold sup-poses, but merely less isolated from other spheresof everyday life.

Having established the social and cultural con-stitution of effectivity we can now link it with ourearlier discussion of instrumental and communica-tive action to suggest that environmental effectivi-ties comprise two main types: communicative andinstrumental. If effectivities are what agents areÒpractically equipped to doÓ (Ingold, 1992, p. 46),then communicative effectivities are those capabil-ities that allow agents to conceive ecological ob-jects in speciÞc enabling and constraining ways,and instrumental effectivities are those technicalcapabilities that facilitate speciÞc instrumentaluses of the environment (the difference here is be-tween, for example, being able to conceive a pieceof mountainside as a socially and symbolicallyavailable site for cultivationÑto be able to associ-ate it with our own project of cultivationÑand hav-ing the technicalÑand socialÑcapacity to terrace,



10

irrigate, cultivate it, etc.). The two types of effec-tivities relate to the practical and technical know-ledge constitutive interests identiÞed by Haber-mas. The relationship between these types of effec-tivity is parallel to the relationship between com-municative and instrumental action, and betweenpractical and technical knowledge constitutive in-terests, in that instrumental effectivities are contex-tualised by communicative effectivities; whatagents can do with the environment technically de-pends on what they can conceive it to be. Bothtypes of effectivity are contexts for knowing, thatwhich individuals bring to an encounter with theirecological setting, but both are also the products ofprevious environmental encountersÑby the sub-ject and othersÑso that individualsÕ effectivitiesare constituted in and through place.

Affordances, by deÞnition, are Òinherent poten-tials of [environmental] objects themselvesÓ thatÒrender it apt for the project of a subjectÓ (Ingold,1992, p. 42). It follows that subjectsÕ projects orÒpractical purposesÓ can be either instrumental

and/or

communicative. What the environment af-fords for persons also falls into two broad catego-ries: those affordances that relate to environmentalknowledge geared to technical projects, and thoserelated to knowledge in support of practicalprojects (associated with constituting a stable lifeworld). The everyday projects or purposes throughwhich we encounter the environment may be in-strumental or communicative

from the start

; con-trary to IngoldÕs conceptualization, communica-tive projects are not an ÒepilogueÓ to environmen-tal perception. If environmental affordances areÒuse valuesÓ as Ingold suggests, then the uses ofenvironmental objects must be recognised as bothtechnical and symbolic. In other words, the ecolog-ical environment yields both symbolic and instru-mental resources (cf. Butz, 1996).

We would like to suggest that ecological dimen-sions of senses of place emerge from accumulatedsets of perceived/known ecological affordances.Ecological senses of place are the knowledges of aplaceÕs ecological characteristics that yield mean-ings which make persons identify with the place.Recognized affordances, whether symbolic or in-strumental, are perceptions that yield such mean-ing because they are generated out of the interplaybetween the characteristics of a speciÞc place-grounded environment and the effectivities of theperceiver. They are not attributes of ecology alone,but rather products of human encounter with an ec-ological setting. In that way, ecological senses of

place are best understood as contingent outcomesof the relationship between effectivities and af-fordances, and as such may be sharply demarcatedor blurred depending on social context; or perhapsas disclosures of what exists between the charac-teristics of human communities and the ecologicalenvironment they occupy. Several points emergefrom conceptualizing ecological senses of place inthis way, which resonate with the sets of commentsmade earlier about the relationships among com-munity and social and ideological aspects of place.

First, peopleÕs effectivities can be understood aslife world elements which, like all aspects of lifeworld, are shaped both by subjectsÕ communica-tions with others and their own instrumental inter-action with the environment. What we take

to

anenvironmental encounter is a product both of whatwe have learned from others and what we have ex-perienced directly in previous encounters with thatecological setting. What we take

from

an environ-mental encounter is the knowledge of certain envi-ronmental affordances, which immediately changethe effectivities we will bring to future encounterswith that (and other similar) ecological settings.The life world-based ÒproductsÓ of this iterative re-lationship between effectivities and affordancesare senses of place which (to the extent that afford-ances are attributes of an actual ecological environ-ment) are themselves grounded in a particular eco-logical setting, and which arise out of a processwhich is grounded in that ecological setting. Ifsenses of place are emplaced aspects of life world,then ecological senses of place are ecologicallyemplaced aspects of life world. In other words, itis the relationship between ecological setting andlife world which comprise the core of ecologicalsenses of place.

Second, ecological affordances, to the extentthat their perception is shared by members of aspeech community (and the instrumental require-ments of social life makes them shared), become abasis for commonality in the life worlds of partic-ipants, which help make some of their validityclaims recognizable and tangible to one another. Insome instances, then, shared ecological senses ofplace can beneÞt efforts to achieve intersubjectiveunderstanding among members of a speech com-munity. Third, the relationship between ecologicalsetting and social interaction also works in the op-posite direction, in that participants in discoursecreate and reproduce effectivities (the potential torecognize affordances) as they attempt to under-stand one another. Ecological aspects of the life



11

worldÑecological senses of placeÑare as muchsocial and mental constructs as they are productsof a physical ecological setting, but they are nev-ertheless strongly rooted in that physical setting,whether or not such rooting is consciously recog-nized. Fourth, ecological senses of place are them-selves constitutive of physical ecological settings.One product of the interplay of effectivities and af-fordances is instrumental action geared toward theutilization and manipulation of ecological objects,and the production of new environments. Again,the relationship is two-way, for the symbolic com-ponent of ecological settingsÑecological sensesof placeÑare materially constituted out of the ex-periences of real people in real ecological settings.

Fifth, ecological senses of place, like senses ofplace in general, are never purely individual orpurely collective. They are never purely individualbecause effectivities are developed and affordanc-es recognized within a social context. Ecologicalplace meanings may be highly private, but are nev-ertheless (contrary to IngoldÕs conceptualization ofenvironmental knowledge) socially implicated.They are never purely collective, because each in-dividual encounters different ecological settingsfor rather different practical purposes, and in thecontext of somewhat different effectivities. Fol-lowing from that, a personÕs ecological senses ofplace are unlikely to be stable; the continuous mu-tually constitutive interplay between effectivitiesand affordances in the context of ongoing ecolog-ical encounters means that ecological senses ofplace are always works in progress, always Òbe-comingÓ. Nor are they likely to be unitary, becauseindividuals encounter an ecological setting frommultiple subject positions, and in the context ofmultiple, perhaps conßicting, practical purposes.

Sixth, and Þnally, following from IngoldÕs treat-ment of direct perception, we have conceptualizedecological senses of place as deriving from per-sonsÕ practically grounded encounters with spe-ciÞc ecological objects. Thus, they are particulartypes of attitudes toward particular types of places.Not all senses of place have an ecological com-ponent, and in no way can we suppose that all sens-es of place are grounded ecologically, or in materialenvironments. It may be that our strongest sensesof place are of places which have no physical en-vironmental grounding (for example, imaginedplaces or ÔcyberplaceÕ), and these can hardly be de-scribed as ecologically grounded. An obvious cor-ollary is that the strongest and most resilient eco-logical senses of place are likely to emerge in

groups whose interaction with a place is rooted innumerous and ongoing ecological encounters, con-textualized by a variety of everyday practical pur-poses, in a social setting characterised by sustainedcommunicative action regarding the symbolic andinstrumental use value of the ecological character-istics of the place.

Having outlined our conceptualization of sensesof place as constituted socially, ideologically andecologically, we are now in a position to apply thisinterpretive framework to two very different casestudies: Shimshal, in mountainous northern Paki-stan, and Towcester, in the English Midlands. Farfrom a uni-directional application of abstract the-ory, the outlines of the interpretive framework de-scribed above were originally developed as an at-tempt to understand empirical data collected inTowcester (Eyles, 1985). The subsequent applica-tion of that interpretive framework to circum-stances in Shimshal inspired its further develop-ment and partial reworking, and especially height-ened attention to ecological constituents of sensesof place (Butz, 1993). Based on his research inTowcester, Eyles (1985, p. 66) suggested that Òec-ological structure

per se

may only be of limited val-ue in conceptualizing sense of place and its deriva-tionsÓ, while Butz (1993, p. 520) insisted that ÒinShimshal at least, ecological context (the naturalenvironment

per se

) relates directly with individu-alsÓ shared identity and shared membership of acommunity... the social and ideological compo-nents of place and community are also closely in-tegrated with ecologyÓ. The conceptualization out-lined above is thus already a product of an initialround of reßection on the empirical circumstancesin Towcester and Shimshal, which we now wish toreapply to those contexts.

The Shimshal case foregrounds the ecologicalcomponents of sense of place. It appears that eco-logical senses of place are central to most Shim-shalisÕ conceptions of, and attachment to, theircommunity, and that they relate in signiÞcant waysto other place-oriented elements of ShimshalisÕlife worlds. The material and social contingenciesthat accentuate ecological senses of place in Shim-shal are largely absent in Towcester. Rather, sensesof place are organized around social relationshipsin which spatial proximity is important, but whichare only weakly mediated by a shared set of em-placed ecological effectivities, affordances andpractices. In Towcester the social (instrumentaland communicative action) and ideological (lifeworld) are less ecologically implicated than in



12

Shimshal, for reasons that will become apparent inthe case studies.

It may be worth summarizing, at this point, themethodologies employed in the two empiricalstudies, especially as we emphasize below that thevalue of our conceptualization emerges in partfrom its grounding in two very dissimilar socialand spatial contexts

and

two somewhat divergentmethodological approaches. The Towcester surveytook Þve months to complete in the summer of1982. It was set up as a structured survey, so the re-sults would have some generalizability. Some 168residents participated in the door-to-door survey,so although structured with scales and closed ques-tion responses, the interviews were face-to-face.There were also a series of open-ended questions,which produced a great deal of information andwhich were useful in constructing the ideal types.While some rapport was established, this was by nomeans a nuanced encounter. The roles of Òinter-viewerÓ and ÒintervieweeÓ were based on a modelof professional competence and brief Òinterfer-enceÓ in peopleÕs lives (Moser and Kalton, 1976;Eyles, 1985). The information on which the Shim-shal study is based was collected during sevenmonths of ethnographic research in Shimshal dur-ing the summers of 1988 and 1989, as part of an ef-fort to evaluate the inßuence of agency develop-ment initiatives on community-level decision mak-ing (Butz, 1993). Triangulation among ethno-graphic methods (observation, participation,conversation) at different sites (village, trails, pas-tures), with special attention to the selective per-ceptions of different groups, gender and the loca-tional context of behaviour, contributed to the de-velopment of a nuanced case study. This was facil-itated by participation in formalized community-level discussions of the portering issue described inthe case study, and by daily and intimate commu-nication with one of the leading participants inthose discussions. Unlike the Towcester study,community members did not allow a systematicsurvey. As we discuss below, these differences inÞeld methods have implications for the senses ofplace we found, and for the subsequent conceptu-alization.

Senses of place in Towcester revisited

In this relatively afßuent small town in MidlandEngland, one of the most urbanized societies in theworld, ecological setting does not seem important.Indeed, in the research carried out in the early to

mid-1980s, the environmental sense of placeÑtheimportance of place in its own right, with social,familial and traditional meanings being relativelyunimportantÑwas held by 1% of the sample ofpeople interviewed (Eyles, 1985). ÒThe country-side was not a stage for acting out roles or lifestyleor way of life. Nor was it a commodity to be used,but something to be lived in itself. That living wasdone with others, but place was more than a back-drop to social or economic activitiesÓ (Eyles, 1985,p. 126). Yet on re-analysis, this seems a very narrowdeÞnition of ecological setting. If this setting ofTowcester provides environmental affordances interms of properties of the environment directly per-ceived by an agent in the context of practical action,then these may exist without being used. The factthat they are not used in such an environment pointsto the indirect relation between individuals and anenvironment in which technology shapes individ-ual consciousness and human relations to other ob-jects. Our instrumental views of space and placeÑhighlighted in the Towcester survey (17%)Ñarerelevant to the seeming unimportance of ecologicalsetting, seen as dominated by our rational ap-proaches to planning, architecture and urban forms(cf. Relph, 1976). In this respect, the environmentoffers for some few affordances communicatively,and therefore, technically. If affordances are notÒseenÓ and ÒtransformedÓ into effectivities, thenthe ecological will remain relatively insigniÞcant.

Thus effectivitiesÑthe action capabilities ofagentsÑare provided primarily in the social andideological components of community and place.As we have argued above, the social provides thebasic material for everyday life in a community,while the ideological pertains to a matrix of sym-bols that can engender a sense of belonging andidentity. For people in Towcester, a sense of placeÞrst and foremost predicates and is predicated ona set of social and cultural (symbolic) relations.These essentially consist of sets of social interac-tions provided in shared meanings, values and in-terests. The social then is the broad brush way ofconÞguring senses of place in Towcester. The so-cial demonstrates the integral interrelations of in-strumental action (geared to technical or instru-mental outcomes) and communicative action(geared toward reaching intersubjective under-standing through negotiating shared meanings).The Þrst analysis of senses of place in Towcesteremphasized their role in the routinization of livea-bility of everyday life. In other words, it tried to an-swer the question of how sense of place relates to,



13

shapes and is shaped by place-in-the-world or in-dividual-in-social-context. It thus established cat-egories of senses of place (Table 1).

A ÔsocialÕ sense of place is one dominated by theimportance attached to social ties and interaction.Place has little meaning without reference to theseties and interactions. This does not mean that thesocial networks are ÒplacelessÓ. They do not occuras activities divorced from their locational (or ec-ological) context. Towcester is regarded as the cen-tre of the local networks. It is the location wherefamily, neighbours and friends are to be found. Tobe sure, other friends are found in different loca-tions and are visited at their homes, but these Òdis-tantÓ friends also come to Towcester. So while so-cial ties predominate, they occur at particularplaces which are, in their turn, regarded as impor-tant because of the social activities which occur atthese places. This apparent tautology dissolves.Place has social signiÞcance and social ties haveplace signiÞcance.

The Òapathetic-acquiescentÓ category may beregarded as having no sense of place at all. Thesense was labelled ÒapatheticÓ because the re-sponses of individuals who were so deÞned dem-onstrated little interest in or commitment to any-thing, let alone place. It was argued that responsessuch as Òall rightÓ, Ònot muchÓ, and Ònothing real-lyÓ demonstrate more than a lack of interest in be-ing questioned once they are placed in ÒcontextÓ.Life seems to possess few affordances, and effec-tivities are not isolated from the living of life itself.It is also labelled ÒacquiescentÓ because apparentapathy may disguise a feeling of powerlessness, ofthe inability to shape the course of events whichform an individualÕs life.

An ÒinstrumentalÓ sense of place is deÞned asone which sees place as a means to an end. Theplace is signiÞcant according to what it does ordoes not provide in terms of goods, services andformal opportunities. The word ÒformalÓ is usedbecause the instrumental category reßects the serv-ice and employment functions of a place ratherthan its sociability. In the case of Towcester, thissense of place is overwhelmingly negative (i.e.ÒThe place stinksÓ, ÒitÕs boringÓ, or ÒthereÕs no[

sic

] shopsÓ). It is what the town does

not

providethat emerges from individualsÕ accounts. The townis seen as a non-provider of most goods and serv-ices and as allowing only limited opportunities toobtain such goods and services. Its location, its rel-ative isolation and inaccessibility may also be seenas a barrier to procuring these desirable goods andopportunities.

A ÒnostalgicÓ sense of place is one dominated byfeelings towards the place at some time other thanthe present. It therefore involves looking back.Feelings about the place are based on the past andin particular these feelings are shaped by speciÞcevents that occurred in Towcester in the past andwhich colour and shape the individualÕs current ap-preciation of place. In this respect, nostalgia is bothpositive and negative. In the sense of longing forsomething, or more usually somebody no longerattainable it is negative because it Þlls many peoplewith remorse, regret and sadness. But it is also pos-itive, in that the act of remembering the sharedtimes of the past often results in contentment anda kind of happiness. Memory is of course a selec-tive device, but it may enable people to come toterms with present lives which they may Þnd, to adegree, unsatisfactory. Bereavement, divorce andloneliness in the present may be overcome in themind by remembering a happy marriage, enjoyablecourtship, or simply good times. But such thoughtsand feelings were related not only to people butalso to place. At the most conscious level, Towces-ter was liked for a remembered courtship or dis-liked because it was where a loved one had died forexample, ÔIt was better when [X] was aliveÕ). Lessexplicitly, it was the place where the personal pasthad happened and was signiÞcant for that.

The remaining six senses of place have fewerthan 20 respondents per category. It may be thatthey represent sub-categories of the four majorones. They are, however, regarded as sufÞcientlydescriptive to categorize separately. The Òcom-modityÓ sense of place is dominated by a search forsome ÒidealÓ place in which to live. ÒIdealÓ is used

Table 1. Senses of place in Towcester.

Category No. of individuals %

Social 31 19Apathetic-acquiescent 29 18Instrumental l28 17Nostalgic 22 14Commodity 12 7Platform/stage 11 7Family 10 6Way of life 10 6Roots 7 4Environmental 2 1

Total 162 99

Source:

Eyles, 1985, p. 122



14

in the sense of having some preconception of whata place should provide in terms of a quiet, safe en-vironment, facilities or types of people. Further,such a sense of place is held by the comparativelymobile, usually in professional, managerial or in-termediate non-manual occupations, for whom theactual place of residence is relatively unimportant.A Ògood placeÓ, however deÞned, is part of the life-style of this group and like any other possession itcan be traded. Place becomes a commodity notonly in the sense of being buyable and sellable butalso usable or ÒconsumableÓ. What emerges is thata place is used for a time and during that time mayor may not be highly regarded (for example, ÒWewonÕt be here longÓ, ÒIÕve tried to get involvedÓ,Òmy husbandÕll be moving soonÓ). After a speciÞctime, it may be discarded for another place. Thissense of place has, therefore, built-in obsolescencewith respect to any speciÞc place. Further, while at-tachment to a speciÞc place may be low (and thelack of involvement of this group in Towcester lifesuggests that this is currently the case), importanceof place remains high. We can speculate that sucha group may value those places with high levels ofenvironmental affordances, with ecological setting(type of landscape, recreational opportunities) be-ing highly prized.

Similar in many respects to the ÒcommodityÓsense of place is the ÒplatformÓ or ÒstageÓ category.It refers to those who see where they live as a stageor platform on which to act out their lives. It mayrefer to some ÒidealÓ picture of place but this senseof place is not as commodiÞed as that of the previ-ous category. They are more likely than those withÒcommodityÓ sense of place to search for and Þndlasting attachment to place and people. They searchfor people like themselves with whom they createstable, patterned social relationships, for example,commenting, ÒweÕve tried to mix in with the neigh-boursÓ. They come to see themselves as ÒTowces-ter peopleÓ, although more often than not their def-initions of Towcester are limited to their subdivi-sion estate or street. Place may symbolize their at-tachment to particular people and activities,although it is important to note that it is the inter-action in a particular place rather than the place it-self that remains dominant.

The ÒfamilyÓ sense of place is deÞned in termsof immediate family connections, often nuclear butsometimes extended. Feelings about place areshaped, therefore, by the nature of family relation-ships. Family life and how a particular place affectsfamily life are seen as central life concerns. For the

majority in this small category, life revolves aroundthe nuclear family, its dwelling, and its happinessand satisfactions. In Towcester at least, little else isregarded as important. Other elements of lifeÑwork, shopping facilities, neighbours and so onÑare signiÞcant only insofar as they impinge on fam-ily life (for example, ÒThereÕs little for my kids todoÓ, Òmy wife would have nothing if it wasnÕt forthe neighboursÓ). Place is a refuge insofar as it iswhere family is located.

It may be argued that the Òway of lifeÓ sense ofplace has in part already been described. The socialdimension is important in the constitution of thissense of place. But it is derived from more than so-cial activities. PeopleÕs whole way of life wasbound up with Towcester. They were ÒlocalitiesÓ interms of jobs, friends and associational life. Theyfelt that they belonged. Closely related to this senseof place is one based on ÒrootsÓ. Again, the placerepresents something important in its own right andthis phenomenon, whether it is social life, lifestyleor sentiment, is strengthened by being based on orrooted in the past. This rootedness usually takes theform of family ties in the town and/or district, so asense of belonging seen in terms of continuity, ortradition, is added to the familiarity which comesfrom basing much of oneÕs life in a speciÞc place.This group belongs to the place without reallythinking about it or articulating their belonging.They simply feel, indeed are, Òat homeÓ (for exam-ple, ÒI love living hereÓ, ÒIÕve been here years, itmust be something in the waterÓ). The importanceof place in its own right may also be seen in the tworespondents with an ÒenvironmentalÓ sense ofplace. However the place was not seen as importantfor its social, familial or traditional meanings butas an aesthetic experience (for example, ÔI feel intune with the countrysideÓ). The countryside wasnot a stage for acting out roles or lifestyles or wayof life. Nor was it a commodity to be used, butsomething to be lived in itself. That living was donewith others, but place was more than a backdrop tosocial or economic activities.

These senses of place were categorized as idealtypes, Òone-sided attenuations of reality... a meansof selection of the facts and a mechanism for spec-ifying their signiÞcanceÓ (Hirst, 1976, pp. 58Ð9).Senses of place could be negative or positive. Butwhat was categorized was a dominant sense ofplace because the purpose of the analysis was todiscover the range in variation in senses of place,given the characteristics of the population andplace under investigation. In some respects, it was



15

an attempt to isolate distinct speech sub-communi-ties. But if we argue, with Habermas (1984), thatinstrumental and communicative action occur co-incidentally and that the social and ideological im-plicate one another, we can now view the senses ofplace as contingentÑdependent on a particularconÞguration of circumstances, values, activitiesand actions. Indeed, participation in several speechsub-communities (place-bound or not) makes itlikely that senses of place will often be tentative,with this tentativeness being an alternative expla-nation of what was categorized as apathetic-acqui-escent: a feeling of being unsure of or over-whelmed by all that was happening in particularcommunities in oneÕs place in the world. Further,contingency may also be seen as being dependentupon the negotiation or struggle over meanings(both instrumental and communicative) which inturn shape the meaning(s) of place-in-the-world it-self (and the signiÞcance or apparent non-signiÞ-cance of ecological setting within it). In otherwords, what is important in that negotiation at aparticular time and in a particular place? Is it whatan individual does (instrumental)? Or, as seems tobe so important to the Towcester sample, where anindividual is located socially (communicative)?Put differently, what symbols or social relations de-Þne characteristics of an individual in a place? Is itthe signiÞcance of the past (nostalgic), an ideal ofwhat a community should be like and should pro-vide (commodity, platform) or a shared life world(social, way of life, roots)? A re-interrogation ofthe Towcester data with the conceptual frameworkof this paper would suggest it could be any. It alldepends, yet individuals have a sense of why theyare where they are of these, communicatively andtechnically, socially, and ideologically, and interms of the signiÞcance of the ecological; in termsof affordances for their life worlds.

But, further, saying Òit dependsÓ does not meanthat anything goes and we can have any sense ofplace we wish or desire. Sense of place depends in-strumentally in a Habermasian sense on materialcircumstances. It also depends on the value attach-ments of a particular social order. In the relativelyafßuent world of small-town England, those valuesemphasize the rural and the landÑand anti-urban-ism (Nisbett, 1966; Glass, 1968; Williams, 1969;Newby, 1977)Ñdependent on the once-removedrelation of the urban dweller from his or her envi-ronment. Affordances in Towcester are technolog-ically mediated within a frame of reference whichstill largely emphasizes the domination of nature

for the betterment of humankind. Seeing the worldin mechanistic terms (Buttimer, 1993) and humanhistory as progress through the control and domes-tication of nature allows for the setting aside of theecological. It also points to the centralization of thesocial and the ideologicalÑthe world as a sociallyconstructed place. The senses of place derivedfrom Towcester stem from such a world view, onethat may relegate the ecological further as the so-cial is threatened by technological change as weenter the new millennium. Similarly, the worldview (and material interests) dominant in Shimshalprovides the logic for their derived senses of place.

Senses of place in Shimshal, Pakistan

In his interpretation of senses of place in Towces-ter, Eyles identiÞes and describes ten ideal types of(negative or positive) Òattitudes towards a place orplaces that are regarded as the most important phe-nomenaÓ in distinguishing an individualÕs sense ofplace (1985, p. 123). He notes that although Òit isnot suggested that the relative importance of thesenses of place can or will be replicated else-whereÓ, the Òsense of place categories them-selves... may be of wider signiÞcanceÓ (1985, p.123). ButzÕs (1993) subsequent ethnographic Þeld-work in ShimshalÑa community with social, po-litical, economic and ecological characteristicsvastly different from TowcesterÑshows that Ey-lesÕs typology is indeed of wider signiÞcance.Some of the same factors connect residents of thetwo communities to their places, although, as ex-pected, the relative importance of the various typesvaries between Towcester and Shimshal. However,that overlap does not necessarily signify a deepersimilarity either in the ways those senses of placeare constituted in Towcester or Shimshal, or in theirimplications for behaviour in the two communities.Indeed, we wish to demonstrate that the social, ide-ological and ecological aspects of community aresufÞciently dissimilar in the two communitiesÑand integrate sufÞciently differentlyÑthat thesame types of sense of place are constituted quitedifferently in Shimshal than in Towcester. As weshall see, the Shimshal case allows for an explicitrecognition of the signiÞcance of contingency insenses of place; of the interconnections amongsenses of places; and of the pivotal role of ecolog-ical context. Our discussion of the constitution ofsenses of place in Shimshal begins with a brief con-textualizing sketch of the community, followed bysome attention to an issue that has been preoccu-



16

pying Shimshalis recently: a perceived need to reg-ulate portering employment in the community.

Shimshal is an indigenous mountain communitylocated high in the Karakoram Himalaya in Paki-stanÕs northern areas. Its 1,300 inhabitants are allIsmaili Muslim, Wakhi-speaking members of land-holding, farming and herding households. Exceptfor a few craft workers and teachers, all permanentresidents labour on (mainly) subsistence agricul-tural activities in the household compounds, irri-gated terraces and pastures, or for cash, carryingloads for visitors trekking to the permanent settle-mentÑthree daysÕ walk from the nearest roadÑorto wilderness attractions further upslope. EachShimshali lineage can trace its ancestry via one ofthree sub-clans through several centuries of contin-uous occupation of the place called Shimshal, andeventually to Mamu Shah and Khodija, the com-munityÕs founders. Except for household com-pounds and irrigated terraces, which are underhousehold control and passed from generation togeneration according to agnatic lineage, all otherspaces and places in Shimshal are the collectiveproperty of the community, although the use ofsome of these (for example, improved pastures andirrigation channels) is organized along maximallineages and sub-clan lines. We see immediatelythat Shimshalis experience a more direct and moreinstrumentally signiÞcant relationship with theirecological setting than do inhabitants of Towcester.

The communityÕs socio-economic homogeneityand emphasis on collective organisationÑboth ofwhich are remarkable, even for northern Paki-stanÑwould suggest that inhabitantsÕ senses ofplace are likely to be largely shared. Nevertheless,apart from myriad individual differences, at leastfour sets of group attributes are important for dif-ferentiating the way adult Shimshalis experiencethe places in their community. First, members oflarge and wealthy householdsÑthose with a sur-plus of some combination of cash, land, animals ormembershipÑexperience ShimshalÕs social, eco-logical and ideological qualities as more enablingthan members of smaller and poorer households.The range and depth of wealthy householdsÕ actioncapabilities allow their members to recognize andutilize affordances unavailable to poorer villagers.Second, perhaps a quarter of Shimshali men workor study down-country, and reside only seasonallyin Shimshal. These young men, with their greaterexperience of the outside world, high levels of for-mal education and lack of familiarity withÑor di-rect reliance onÑindigenous ecological/agricul-

tural practices, have differently conÞgured sensesof place from those Shimshalis who have alwayslived in the community all year round. These mendo not necessarily have shallower, less ecologicallyoriented, or less tradition-based senses of placethan others. However, their practical projectsÑboth communicative and instrumentalÑand the ef-fectivities they bring to these projects are thussomewhat different from those of other Shimshalis.SpeciÞcally, most seasonally resident Shimshalisare involved in portering, many are advocates ofWestern-style environmental activism, and severalhave developed an active interest in ShimshalÕs richhistory and mythology. Third, some general differ-ences exist among the generations. In particular,older Shimshalis retain immediate material andsymbolic connections to a time when social andeconomic organization was quite different fromthat practised today: for example, when the com-munity was controlled by the feudal kingdom ofHunza; when clan and lineage organization, androyally sanctioned prerogative, were more impor-tant determinants of everyday life, and communitywas less so; when ShimshalÕs highest and most val-uable pastures were controlled by China, and werethus inaccessible to Shimshalis for a time; when thecommunity depended almost solely on subsistenceagriculture; and when most Shimshalis never lefttheir valley. Symbolic remnants of these circum-stances inform the life worlds of their juniors. But,in general, the place experiences of younger Shim-shalis are more directly inßuenced by contempo-rary circumstances: for example, a growing de-pendence on waged labour and seasonal labour mi-gration; an increasingly cash-oriented agriculturaleconomy; greater interaction with outside visitors;higher levels of formal education; and governmentand development agency intervention into most ar-eas of community life. Given the pace of social,economic and political change in Shimshal, it is notsurprising that both older and younger Shimshalisbring different effectivities to their understandingof community and place, utilize different af-fordances in their instrumental and communicativeaction, and thus reproduce a different place and dif-ferent senses of place.

Fourth, while the data from Shimshal tell usmainly about menÕs understandings of place, it isclear that women and men experience differentplaces, and that women experience their com-munity differently from men. Throughout the com-munity women enjoy less autonomy than men.What autonomy they have is greatest in the places



17

they spend the most timeÑhousehold compoundsand high pasturesÑwhile menÕs control is greatestin the Þelds, village public spaces, and on the trail.In addition, women have less experience of the out-side, participate less in cash endeavours, have low-er levels of formal education, and utilize importedtechnology less. On the other hand, it seems thatmany women experience a more direct relationshipwith the communityÕs mythological history andtraditional lore, and its many place referents, thando most men. It is apparent that the places whichconstitute Shimshal offer different environmentalaffordances to women than to men, and women aresocialized to bring different effectivities and prac-tical purposes to their interactions with the placesthey encounter in their daily lives.

The sets of differences suggested above begin tolend some diversity to ShimshalÕs apparent homo-geneity. It is important to recognize, however, thatthe distinctions between these categories are not al-ways clearly drawn, and that considerable variationof experience and life world occurs within eachcategory. Despite the existence of the four mainsets of dichotomies outlined above, and their mul-tiplicity of combinations, contradictions and nego-tiations in any individual Shimshali subject, it isapparent that some senses of place are largelyshared, at least among Shimshali men.

In the last decade, porteringÑcarrying loads forpayÑhas become the most important form of cashemployment for resident Shimshali men. Apartfrom its considerable monetary rewards, porteringhas several other attractions: it is ßexible enoughto accommodate the seasonal demands of subsist-ence agriculture and other community obligations;it is at least potentially accessible to every able-bodied male regardless of education, wealth, line-age or social standing; and it is one of the few typesof manual wage labour that attracts some prestigeapart from the Þnancial gains associated with it.The fact that there is limited demand for Shimshaliporters has precipitated considerable debateamong villagers regarding the regulation of porter-ing opportunities and activities. This debate, whichhas been ongoing for at least a decade, reached aclimax in 1988 and 1989, when two positions to-ward portering regulation solidiÞed in oppositionto one another. In the summer of 1989 the commu-nity made a tentative decision in favour of one po-sition. Since then the debate has considerably sub-sided, but has not disappeared entirely. Our pur-pose is not to rehearse the details of this debate overportering regulation, its discursive foundations, or

its tentative resolution (cf. Butz, 1995), but ratherto discuss how senses of place are represented in,constituted by, and constitutive of, the communi-cative process of determining the instrumentalitiesof portering in Shimshal.

The two positions that emerged in the porteringdebate paralleled two long-standing ideologicalformations, and indeed relied on them for their re-spective legitimacy. Advocates of what Butz(1995) describes as an ÒauthoritativeÓ position to-ward portering argued for the perpetuation of theprerogative of a few authoritative leadersÑthosewith ties to the former royal administration or cur-rent central governmentÑto regulate the economicopportunities of villagers, and especially their in-teraction with outsiders. These men thus claimedthe formal right to select porters, regulate food pur-chases, act as paid guides, establish porter wages,mediate all personal relationships between Shim-shalis and visitors, and authorise routes and itiner-aries, without the explicit and ongoing interventionof the council of household heads. The Òdiscur-siveÓ position (cf. Butz, 1995) represented the mostrecent incarnation of long-standing efforts to estab-lish what proponents perceived to be a more equi-table process for regulating economic activities,and especially for selecting porters, by removingtraditional privilege from the realm of portering,and opening porter regulation to a consensual proc-ess aided by a set of technical guidelines that wouldensure the relatively equal access of all householdsto portering opportunities, regardless of size,wealth or lineage. Thus, what emerged in Shimshalwas a debate over portering which was interwovenwith a larger process of communicative action con-cerning the contemporary practical signiÞcance oftwo deeply rooted ideological formationsÑprivi-lege and equityÑboth of which saturate Shimsha-lisÕ daily lives and life worlds.

It is clear that the ideological contexts whichShimshalis brought to the portering issue weregrounded in their daily experience of, and theirspeciÞc social and material situation in Shimshalas a place-based community. How individualShimshalis aligned themselves behind these ideo-logical formations relied largely on their relativeabilities to beneÞt from equity or privilege. Thisdepended, in the Þrst instance, on kinship connec-tions. Other things being equal, those householdswith the closest kinship connections to Òauthorita-tiveÓ leaders had the most to gain from authorita-tive regulation of portering, and a privilege-baseddecision-making structure more generally. House-



18

hold wealth, size and composition were also im-portant. Wealthy extended households with lots ofmembers felt they would beneÞt, or at least avoidsuffering, from the tactics of ÒauthoritativeÓ lead-ers, mainly because they have diversiÞed incomes,and enough members to exploit portering opportu-nities when they arise, but without relying on them.At the opposite end of ShimshalÕs socio-economicspectrum are several dozen poor elementaryhouseholds; those with exceptionally small landand animal holdings and few adult members.These are the households which most needed an in-ternal labour market to survive, but which had thefewest resources to actively seek porteringÑorotherÑemployment. Thus, their members relymost on benevolent ÒauthoritativeÓ leadership, andsuffer most from leadership that does not recognizetheir need. Dependent as they are on the goodwillof village elites, the very poor tended not to advo-cate a more discursive procedure for regulatingportering opportunities, but rather to advocateÒprivilege with a conscienceÓ, in the form of au-thoritative leadership that would direct some op-portunities their way. Between these two extremesare the majority of Shimshali households, whoseresources are sufÞcient to maintain a comfortablesubsistence, and who wish to participate in themarket economy, but whose membership is toosmall to permit labour migration. These house-holds sought equal opportunities to participate inportering when their members have the time, andreasoned that a consensual regulatory processwould provide that opportunities. Within house-holds, there was a greater tendency for youngermen, and those with high levels of formal educa-tion and experience of the world outside Shimshal,to favour the ÒdiscursiveÓ position.

To some extent, what we see here are Shimshalisresponding to their different sets of ecological ef-fectivities and associated affordances. The envi-ronment affords a great deal to big, rich households(i.e. large amounts of land, livestock, trees) andthese households can muster instrumental andcommunicative effectivities to utilize them. Con-versely, the environment affords very little to thesmallest and poorest households, whose effectivi-ties are limited by their access to the benevolenceof others. Heads of the remaining households seeportering as affording them a range of ecologicallybased opportunities that their householdsÕ instru-mental effectivities could use, but realize that theircommunicative effectivities are too limited to com-pete for those affordances with privileged house-

holds, unless an equitable system of regulation isput in place.

Clearly, different social, economic and ecologi-cal circumstancesÑdifferent action capabilitieswithin the social and spatial characteristics of theirsurroundingsÑinßuenced the life world character-istics, or senses of Shimshal as a place, householdmembers brought to the debate. In the end, howev-er, a consensus was reached, which seems to haveemerged from a shared sense of place which tran-scended, and to some extent united, the ideologicalpositions, and variable action capabilities, dis-cussed so far. Part of the argument of the discursivegroup was that Shimshal was Þrst and foremost afarming community, whose material and moralsustenance relied on maintaining the subsistencebase of each household, and whose future as a com-munity depended on complementing subsistenceagriculture, Þrst with market-oriented agriculture,and second with other economic opportunities.They argued further that their regulation schemewould strengthen and complement the other col-lective endeavours (for example, completing arough road into the community, building bettertrails to pastures, selectively breeding animals, col-lective marketing of produce) necessary to realizethose visions. Finally, they connected the compet-ing ideological streams by claiming that the pastauthoritative leaders whose memories were mostrespected in Shimshal were those who used theirinßuence to unite the community in collective ac-tion for the beneÞt of all households. And Shim-shalis, to a household, eventually agreed that theydid indeed live in a farming community, that theyrelated to it as farmers, and that effectivities shouldbe developed that would allow them to utilize andenhance their ecological affordances in ways thatwould sustain that identity.

It seems that a shared sense of place, groundedecologically, facilitated efforts to achieve some de-gree of intersubjective understanding among thoseinvolved in the portering issue. By reafÞrming theirshared identity as farmers, Shimshalis unitedaround a particular ecological sense of place, a par-ticular set of conventions surrounding the techni-cally and normatively appropriate way to interactas Shimshalis with the ecological characteristics oftheir place. The implications of the resulting con-sensus travelled in two directions: backwards tolife world, and forwards to instrumental action.First, and without any immediate need for instru-mental changes, the senses of place of Shimshaliswere realigned, if only minutely. The debate and its



19

outcome foregrounded, valorized and further nat-uralized background convictions and place sym-bols associated with both collectivity and equity,and farming, and discredited those associated withelite privilege. The whole issue added a chapter tocertain strains of ShimshalÕs place history, but alsoreconstituted the signiÞcance of previous chapters;senses of place are dynamic and contingent. Sec-ond, the outcome of communicative action regard-ing the portering issue has had instrumental impli-cations. Socially and materially, Shimshal imme-diately became a different place, or at least imme-diately had the potential to become a differentplace. Portering regulation itself changed less thanmany Shimshalis had hoped. Within a year, two au-thoritative leaders had resumed

ad hoc

day-to-daycontrol of much of the portering, although with sig-niÞcant input and monitoring from the council ofhousehold heads, especially to ensure that porter-ing activities complemented other collective activ-ities. More generally, the debate sparked renewedinterest in collective activity, resulting in a spate ofroad- and trail-building activity, pasture and live-stock improvement, and schemes to facilitate mar-ket-oriented production. A realignment of lifeworld foregrounded a somewhat altered set of prac-tical purposes. A shift in social organization creat-ed a somewhat different set of environmental ef-fectivities. The result was a greater commitment to,and capability for, expensive and/or labour inten-sive infrastructural and agricultural projects. To theextent that these alterations created/revealed newecological affordances, Shimshal became an in-strumentally (materially as well as socially) differ-ent place.

The discussion above reveals something of therelationships among social, ideological and eco-logical components of community, and illuminateshow senses of place constitute and are constitutedthrough communicative and instrumental action.Focused as it was on a speciÞc issue, it does not al-low us to identify a deÞnitive typology of senses ofplace in Shimshal, or to examine EylesÕs accentu-ations that may pertain to Shimshal. A different ex-ample might have illuminated somewhat differentsenses of place. At least four of the types identiÞedfor Towcester emerged as salient within the exam-ple we provided. They constitute categories of feel-ing or attitude which link Shimshalis through placeto their community.

To begin, Shimshalis have an

instrumental

at-tachment to place, in which place is seen as a meansto an end. Place is signiÞcant in what it does or does

not provide in terms of goods, services and em-ployment. The debate over portering emerged inpart over different ideas about the best way to in-crease placeÕs instrumental value. This sense ofplace was highly positive for most Shimshalis, andfundamentally ecological; they conceived theircommunity as affording a large set of ecologicallybased opportunities, but also worried that many ofthese could not be realized instrumentally by mosthouseholds without some collective effort to sus-tain certain effectivities. Traditionally, environ-mental affordances had been limited to subsistenceactivities and hunting. Increasingly, they also in-cluded using the landscape to earn money throughportering and market-oriented agriculture. Theportering debate expressed a formal recognition ofthese new instrumental opportunities, and the de-sire to utilize them in ways that also reproducedprevious instrumental attachments to place. It alsorevealed some variations in different ShimshalisÕinstrumental senses of place. This form of attach-ment to place is likely to relate closely to the effec-tivitiesÑthe action capabilitiesÑof the subject.The instrumental attachments of members of smalland poor households were considerably differentfrom those of wealthy householders. Young menwho had spent much of their youth in schoolsdown-country found relatively little instrumentalvalue in agricultural opportunities, but many op-portunities in portering. Despite these differences,participants in the portering debate evidently co-alesced around a largely shared sense of Shim-shalÕs positive instrumental value.

The way the portering issue unfolded also illus-trates ShimshalisÕ identiÞcation of their communi-ty with a

way of life

. The two positions towards por-tering, and the alternative ideologies that informedthem, emphasized different notions of how lifeshould be lived in the community. The achievementof the ÒdiscursiveÓ position was to reconcile thesetwo sets of guidelines to a comprehensive vision ofa Shimshali way of life that resonated with otherparticipants in the discussion: Shimshalis are farm-ers and herders. This was more than a statement ofoccupation. It was an afÞrmation of a particular setof relations among ecological affordances and ef-fectivities, and associated social structures andpractices, in which both elite privilege and egali-tarian collectivity played a part. In short, it was astatement of a way of life comprised of ecological,social and ideological components. That Shimsha-lis associate their community with this way of life,and seem mainly to value it, does not mean that all



20

Shimshalis live it the same way. ÒWe are farmersÓmeans something quite different to a junior mem-ber of a small and impoverished household than itdoes to a wealthy lineage elder whose main incomeis an army pension, or to a recently returned collegestudent who does not know how to farm, has no au-tonomous formal voice, and sees no opportunity touse his newly acquired academic training.

Despite extensive variance in the ability of indi-viduals to beneÞt from what they perceived as aShimshali way of life, most resident menÑwiththe exception of a few returning college studentsand army pensionersÑdid feel they belonged. Thisstemmed in part from a positive belie that they didnot belong any where else, but perhaps more fromstrong, positive

roots

senses of place. For Shimsha-lis, place indeed Òrepresents something importantin its own right and this phenomenon, whether it issocial life, lifestyle or sentiment, is strengthened bybeing based on or rooted in the pastÓ (Eyles, 1985,p. 126). As Eyles says, Òa sense of belonging seenin terms of continuity, of tradition, is added to thefamiliarity which comes from basing much ofoneÕs life in a speciÞc placeÓ (p. 126). The fact thateveryone involved in the portering debate had beenborn in Shimshal, had lived most of their lives inShimshal, and were descended from lineageswhose history in Shimshal bridged several centu-ries, informed every aspect of the discussion, andprovided a solid basis for at least a degree of inter-subjectivity. A sense of historical continuity satu-rated the issue: in the ideologies espoused, in theprotocol of discussion, in the place of discussionand the place being discussed, in the ecological af-fordances and associated effectivities at issue. Itwould be difÞcult to overemphasize the extent towhich Shimshalis experience their life worlds andsenses of place as constituted

in situ

, through cen-turies of social and ecological activity in this place.The commitment to maintain a focus on agriculturefacilitated by collective works was not just an ar-gument for instrumental efÞcacy, but also an afÞr-mation of a way of life rooted in the past.

The portering issue also reveals a variant on Ey-lesÕ

family

sense of place, manifest in Shimshal inhousehold and lineage. We emphasised how Shim-shalisÕ interaction with their community membersis strongly mediated by the social and genealogicalposition of their household and lineage within thelarger community fabric, and by individualsÕ posi-tions within the composition of their own house-hold and lineage. Shimshalis relate to one anotherand to their environment as members of house-

holds and lineages. Ecological affordances arestructured by household (ownership of terracedland, animals and trees) and lineage (access to pas-tures). The effectivities Shimshalis bring to ecolog-ical affordances are organized according to house-hold (agricultural production is household based,as is access to portering opportunities). Indeed,households are described by Shimshalis as units ofcollective production and consumption. The indi-viduals who participated in the portering-debatedid so mainly on behalf of their householdsÑandsometimes lineagesÑand discussion revolvedaround how to appropriately structure the opportu-nities and obligations of households, and relation-ships between them, given the increasing impor-tance of portering. Shimshal is meaningful to its in-habitants in part because it provides the social andecological context for their existence as membersof households and lineages.

The portering example does not allow us tomake a convincing case for the importance toShimshalis of the other six ideal types, although itdoes help us to suggest that several of these are

not

likely to be important. Place as

commodity

impliesa potential for disengagement that seems unlikelyfor individuals whose lives have been so complete-ly and so continuouslyÑso ecologicallyÑlived inShimshal. In a more instrumental sense, space hasnot been buyable or sellable (technically or norm-atively), although that may be slowly changing asthe economy becomes more cash-oriented. Since

apathy

is a passive attitude it is not easily identiÞedin ethnographic research. However, apathy towardplace contradicts ShimshalisÕ stated self-percep-tion: a Shimshali who is apathetic about his or herÔplaceÕ is considered a Ôpoor ShimshaliÕ, a labelwhich is more derogatory than the equivalent inTowcester would be. It is hard to imagine how aShimshali could be apathetic toward place withoutbeing apathetic about life in general. It may be,however, that the poorest households merely

ac-quiesce

to the interests of their patrons in the hopeof gaining from Òprivilege with a conscience.Ó

In addition, several of the ideal types whoemerged as important in their own right in Towces-ter seem to collapse into others in Shimshal. As-pects of the portering debate indicate that Shimsha-lis do feel

nostalgic

toward their setting, but thatnostalgia is for a long-term shared history that maybe better expressed within the

roots

sense of place.Similarly,

platform/stage

sense of place, as mani-fested in Shimshal, is bound up within

way of life

.Not even the most manipulative/theatrical Shim-



21

shalis separated themselves from their stage sufÞ-ciently for this ideal type to be relevant to Shim-shal. The

environmental

sense of place did notemerge as important in this discussion, but neitherdid evidence emerge to suggest that it would not beimportant in others. Indeed, in other contexts (cf.Butz, 1996) Shimshalis do claim to experiencetheir environment in aesthetic terms, and apartfrom its instrumental value.

Some of the ideal types developed to explain therange of variation in senses of place in Towcesterare clearly applicable to the small chapter of Shim-shali community life described above. Some thatdo not seem important to the portering issue arelikely to emerge as signiÞcant in Shimshal given adifferent set of issues (for example, environmen-tal), while others seem unlikely to be descriptive ofShimshalisÕ senses of place under any circum-stances (for example, apathetic). What the Shim-shal case provides is a strong sense of the contin-gency, tentativeness and instability of senses ofplace. Earlier in the paper we described senses ofplace as place-based life world elements, and ar-gued that life world both informs and is constitutedthrough communicative and instrumental action.ShimshalisÕ senses of place were reconstitutedthrough their use of them when arguing about por-tering. The nuances of sense of place that emerged,and the way sense of place informed the arguments,were contingent on the speciÞc issue under discus-sion, and the longer-term unfolding of senses ofplace through previous instrumental and commu-nicative action. A different issue, or a different dis-cursive approach to the same issue, would haveforegrounded different senses of place, and left dif-ferent enduring traces in ShimshalisÕ place-basedlife worlds. Contingency implies that any individ-ualÕs senses of place are likely to be tentativeÑsub-ject to challenge. We can infer from the Shimshalcase that tentativeness also emerges from individ-ualsÕ multiple subject positions (i.e. their partici-pation in disparate speech communities). The por-tering issue allowed certain of these subject posi-tions and their concommitant senses of place todominate others for a time, and thus reproduceplace-based life world in a speciÞc way. But com-munity members commit themselves to a speciÞcset of dominant subject positions only tentatively.All this implies that senses of place are only rela-tively stable, not solidly-positioned life world ele-ments which anchor individualsÕ connection totheir habitat (as is often imagined), but rather shift-ing ideologies of place that unfold contingently and

tentatively; never autonomous from previous sens-es of place, or the senses of place of other commu-nity members, but neither are they determined bythem.

Despite the dynamism that the terms contingen-cy, tentativeness and relative instability suggest,the Shimshal case study also indicates the durabil-ity of at least one limited regularity, founded inShimshalisÕ largely shared recognition of a set ofpositively valued ecological affordances, constitut-ing and constituted from a commitment to a spe-ciÞc array of action capabilities or ecological ef-fectivities. As we have demonstrated, the two po-sitions on portering regulation, drawing from quitedifferent interpretations of ShimshalÕs social, ide-ological and material context, shared a conceptionof ShimshalÕs ecological environment as the sourceof life, and the foundation for normatively valued(albeit quite differently conceived) set of social andmaterial practicesÑÓwe are farmers and herdersÓ.Broad-brush regularity and durability are entirelyconsistent with contingency, tentativeness and in-stability; the latter agents of dynamism are limitedby their social, ideological and material context,even as they shape that context, and thus tend notto be radically transformative.

Conclusion

In this paper we have attempted to reconceptualizesense of place using HabermasÕs theory of com-municative action and IngoldÕs reworking of Gib-sonÕs environmental psychology, and to presentcomparative analyses of sense of place in two rad-ically different cultural and geographical settings.The theoretical framework we describe and thecases we present are simultaneous outcomes of aprocess of initial analysis of case studies, engage-ment with theoretical debates, reßection and re-interpretation; a process informed by the juxtapo-sition of disparate empirical contexts, methodolo-gies and theoretical perspectives. In the space thatremains we wish to comment on the contributionsof this process and the juxtapositions it entailed. Inthis, we recognize that we represent the voices ofothers, but feel that because we shared, to some ex-tent, the life worlds of those voicesÑEyles foreight years in Towcester, Butz for seven months inShimshalÑwe are in a unique position to provideinsider/outsider accounts (Powdermaker, 1966;Simmel, 1950).

The material from Towcester identiÞes ten idealtypes of attitudes toward place that emerge as im-



22

portant characeristics of individualsÕ senses ofplace. Indeed, the primary purpose of that researchwas to establish ideal types, to categorize and clas-sify, to remove from context in order to reinsertinto life worlds. There is some evidence that mostof these types are also applicable to Shimshal;broadly similar sentiments connect residents of thetwo communities to their places. It is evident, how-ever, that these ideal types describe somethingquite different in Shimshal than in Towcester, interms of their constitution, distribution and impli-cations for behaviour. This is no surprise, given themajor differences in social, ideological and eco-logical contexts between the two communities. Wewish to concentrate here on three main differences.First, despite the fact that an ÒenvironmentalÓsense of place, as deÞned for Towcester, was notimportant to the portering issue, all the senses ofplace that did emerge as important were constitut-ed to a signiÞcant extent in ecological terms. InShimshal, for example, an ÒinstrumentalÓ sense ofplace derives largely from ShimshalisÕ recognitionof a varied set of ecological affordances, whileÒway of lifeÓ describes a fairly literal relationshipbetween ecological affordances and place-basedeffectivities. This was not the case in Towcester,where attachments to place were deÞned mainlyaround social relationships. This is not to say thatthe social is overwhelmed or determined by the ec-ological in Shimshal, but rather to recognize thatalmost all social life has ecological implicationsthere. It makes a difference for the constitution ofsenses of place that (among other things) in Shim-shal most people make their living from the land,work where they live, and have roots in their envi-ronment that bridge many generations; and itmakes a difference that Towcester is a satellite ex-urban community of mainly recent migrants,whose worlds are technologically shaped by forcesand institutions over which they exert little directinßuence.

Second, while individual Towcester residentsÕimportant senses of place overlap somewhat withthose of other individuals, the data do not indicatethat senses of place are shared in any deeper sense.Its residents live largely separate and household-based lives, implying a segregation from most oth-er residents for most purposes. In Shimshal, somesenses of place do seem to be largely collective, forexample, sufÞciently shared to inform the outcomeof communicative action relating to the porteringissue. Shimshal, as a social entity, is structuredaround the primary, face-to-face interaction of in-

dividuals whose lives intersect continuously in aspatially bounded territory. Despite variations inexperiences and subject positionings, all Shimsha-lis belong to households which share common an-cestors, a common set of ways to make a living, acommon history in Shimshal, and a more or lesscommon positioning in terms of the rest of theworld. These commonalities contribute to the po-tential for intersubjectively shared senses of place.Inhabitants of Towcester share much less in termsof history in place, occupation, ancestry, and socialworld. They forge place attachments as membersof secondary social groups, or as members of sub-or extra-community primary groups. The commu-nity itself is a relatively unimportant source of in-tersubjectivity. A result is less cohesive (and per-haps less politically effective) senses of place. Butthe individuality of senses of place in Towcesteralso hints that place attachment there may operatein an atmosphere that is somewhat freer from moralsanction than in Shimshal. Inhabitants of Shim-shal, unlike those of Towcester, are clearly expect-ed to express certain types of place attachments. Inthis regard, it is worth noting that Shimshal is farfrom an Òideal speech communityÓ, where Òeachsubject who is capable of speech and action is al-lowed to participate [fully and equally] in dis-courseÓ (White, 1988, p. 56). Therefore, intersub-jective understandings of place are (as the casestudy indicates) inevitably and systematically dis-torted by power; that senses of place seem more in-tersubjectively shared in Shimshal than in Towces-ter implies nothing about their relative authenticity.

Third, the Towcester case depicts individuals asexhibiting one or more discrete sets of positioningstoward signiÞcant places. Residents ÒpossessÓ, forexample, ÒinstrumentalÓ or ÒrootsÓ or ÒfamilyÓsenses of place, or perhaps several of them, whichare stable external expressions of subjectsÕ posi-tioning of themselves

vis-�-vis

their surroundings.But there is little comprehension that these sensesof place constitute one another, or help constitutethe subjects who exhibit them. In Shimshal, on theother hand, these positionings emerge as more in-tegrated with in one another, and in the constitutionof subjects, less discrete and less stable. For Shim-shalis, what we have called a ÒrootsÓ sense of placeis not clearly separable from an ÒinstrumentalÓ, orÒfamilyÓ or Òway of lifeÓ sense of place. They areall integrated with one another in a way of life thatis saturated with the contingencies of a particularplace at a particular time. As these contingencieschange, the already blurry boundaries among dif-



23

ferent sources of place attachment realign, withconsequences for the reproduction of subjectivity.

While we are conÞdent that each of these differ-ences actually describes dissimilarities in the con-stitution of sense of place in the two communities,we are also aware that actual differences are at leastexaggerated by the methodologies used in the twocase studies. The original survey and interview re-search in Towcester was designed to solicit eachparticipantÕs dominant sense of place in order todiscover the range in variation in senses of place,and ultimately to construct ideal types descriptiveof broad categories of variation. This methodologyinvolves a twofold process of dissectionÑthe sep-aration of individualsÕ attitudes from those of otherindividuals, and the isolation of discrete predomi-nant senses of place within individualsÑand thentheir re-aggregation as ideal types. Small wonder,then, that the place attachments of different Tow-cester inhabitants seem to overlap without beingshared, and that the data do not reveal a contingentand continuous negotiation of sense of place inter-nal to individuals. The Shimshal case study, devel-oped through ethnographic Þeldwork and geared toa particular community-level issue, suffered differ-ent limitations. Rather than attempting to identifythe range of variation in Shimshali senses of place,it sought evidence in a single social event for thetypes developed in Towcester. The result of this fo-cus on senses of place as they are utilized in com-municative action is an emphasis on a few domi-nant and shared discourses of place attachment rel-evant to the issue in question (to the exlusion of al-ternative or oppositional senses of place), on anunderstanding of the way different senses of placecombine and recombine in response to contingentcircumstances (without much attention to their po-tential autonomy), and on how senses of place arecontinuously reconstituted through communica-tive and instrumental action. Two points can bemade which reduce the suspicion that apparent dif-ferences in senses of place are mainly a methodo-logical artefact. First, both authors develop their in-terpretations out of their experience of living in thecommunities, and thus, despite methodologicaldifferences, both bring the insider and outsider per-spectives of participant observation to their inter-pretations. Second, that our comparative analysisindicates considerable overlap in the senses ofplace between the two communities, despite meth-odological dissimilarities, suggests that differenc-es in our interpretations of senses of place are un-likely to be mainly methodologically determined.

There remain two theoretical contributions towhich we would like to refer. First, by reconceptu-alizing senses of place as place-based constituentsof life world we are able to theorize how sense ofplace is constituted through social and material cir-cumstances, and howÑas an ideological effectÑit inßuences the reproduction and transformationof those circumstances. In particular, the relation-ships Habermas posits among communicative ac-tion, instrumental action and life world help clarifythe ways that place, community and senses of placeare integrated, and suggest several points whichhelp move the sense of place concept beyond itsroots in humanistic geography to recentre it withinwell-developed theories of social organization andsociety. First, the communicative efforts of aspeech community necessarily occur somewhere.The social process of communicative action, to theextent that it is emplaced, engenders senses ofplace on a very small-scale, which links the placeof interaction with the form of interaction and lendsthem both signiÞcance. Second, place, to the extentthat it is shared by members of a speech commu-nity, becomes a basis for commonality in the lifeworlds of participants, so that shared senses ofplace facilitate efforts to achieve intersubjectiveunderstanding among members of a speech com-munity. Third, the process of communicative ac-tion ensures that life world is as much a social con-struct as it is a mental one; shared senses of placeare outcomes of communicative action as well asconstituent elements of it. Fourth, not only is theideological component of place socially constitut-ed, but so are its material aspects. Places are con-structed symbolically and materially as products ofcommunicative and instrumental action respec-tively. The symbolic component of place, in itsturn, is materially constituted in a social context;place is the corporeal setting for individual lifeworlds. It follows from these conceptualizationsthat (1) social interaction, place and sense of placeare mutually constitutive, and none of them can beconceived as originary; (2) senses of place are nev-er purely individual or purely collective, but ratherthe product of social interaction mediated throughindividual subjectivities, and (3) an individualÕssense of place is unlikely to be stable or unitary, butrather subject to the vagaries of both social and ma-terial circumstances and subjectivity formation.All this would suggest that conventional notions ofsenses of place as deÞnitive of the relationship be-tween groups of people and their places should giveway to a conceptualization of senses of place as



24

necessarily tentative and contingent, particularisticand at least potentially contradictory.

The second main theoretical contribution of thispaper is to integrate ecological considerations intoa reconstituted theory of sense of place. Eyles(1985) identiÞed an ecological dimension to senseof place, but did not theorize it fully, nor Þnd it tobe important to senses of place in Towcester. Hu-man/ecological relationships clearly were signiÞ-cant for Shimshali senses of place; not overwhelm-ing social relationships, but rather in a relationshipof mutual constitution with them. By drawing fromIngoldÕs (1992) reworking of GibsonÕs (1979) the-ory of ecological perception, we were able to sug-gest that ecological dimensions of senses of placeemerge from accumulated sets of perceived/knownecological affordances. They are the knowledgesof a placeÕs ecological characteristics that yieldmeanings because they are generated out of the in-terplay between the characteristics of a speciÞcplace-grounded environment and the socially con-structed effectivities of the perceiver. These effec-tivities can be understood as life world elementswhich, like all aspects of life world, are shaped bothby subjectsÕ communications with others and theirown instrumental interaction with the environ-ment. If senses of place are emplaced aspects of lifeworld, then ecological senses of place are ecolog-ically emplaced aspects of life world. It is thus therelationships between ecological setting and lifeworld that comprise the core of ecological sensesof place. Not all senses of place have an ecologicalcomponent, and in no way can we suppose that allsenses of place are grounded ecologically. It seems,for example, that senses of place in Towcester, asperhaps in most of the post-industrial West, arelargely independent of an explicitly ecologicalcomponent. On the other hand, the strongest andmost resilient ecological senses of place are likelyto emerge among individuals whose interactionwith a place is rooted in numerous and ongoing ec-ological encounters, contextualized by a variety ofeveryday practical purposes, in a social settingcharacterised by sustained communicative actionregarding the symbolic and instrumental use valueof the ecological characteristics of the place. Shim-shal is exemplary of those characteristics.

David Butz, Department of Geography, Brock Uni-versity, St. Catharines, ON L2S 3A1, CANADA.(905) 688-5550(email: [email protected])

.

John Eyles, Department of Geography, McMasterUniversity, Hamilton, ON L8S 4K1, CANADA.(905) 525-9140.

Notes1 The authors wish to thank the people of Towcester and Shimshal

for their cooperation, and three anonymous reviewers for their helpful suggestions as to the content of this paper.

ReferencesAGNEW, J. (1987): Place and Politics. Allen & Unwin, Boston.AGNEW, J. (1989): ÔThe devaluation of place in social scienceÕ in

AGNEW, J. and DUNCAN, J.S. (eds): The Power of Place:Bringing Together Geographical and Sociological Imaginations.Unwin Hyman, London, pp. 9Ð29.

BROWN, B.B. and PERKINS, D.D. (1992): ÔDisruptions in place at-tachmentÕ, in ALTMAN, I. and LOW, S.M. (eds): Human Behav-ior and Environment, Vol 12: Place Attachment. Plenum, NewYork, pp. 279Ð304.

BUTTIMER, A. (1993): Geography and the Human Spirit. OxfordUniversity Press, New York.

BUTZ, D. (1993): Developing Sustainable Communities: CommunityDevelopment and Modernity in Shimshal, Pakistan. UnpublishedPhD thesis, Department of Geography, McMaster University,Hamilton.

BUTZ, D. (1995): ÔLegitimating porter regulation in an indigenousmountain community in northern PakistanÕ, Environment andPlanning D: Society and Space 13: 381Ð414.

BUTZ, D. (1996): ÔSustaining indigenous communities: symbolicand instrumental dimensions of pastoral resource use in Shim-shal, northern PakistanÕ, The Canadian Geographer 40: 36Ð53.

CALHOUN, C.J. (1980): ÔCommunityÕ, Social History 5: 105Ð29.CLAGGETT, W.W. et al. (1984): ÔNationalization of the American

electorateÕ, American Political Science Review 78: 77Ð91.CLOKE, P. et al. (1991): Approaching Human Geography: An Intro-

duction to Contemporary Theoretical Debates. Guilford Press,London.

COHEN, A. (1982): Belonging. University of Manchester Press,Manchester.

COSGROVE, D. and DANIELS, S. (eds) (1988): The Iconography ofLandscape. Cambridge University Press, Cambridge.

CUBA, L. and HUMMON, D.M. (1993): ÔA place to call homeÕ,Sociological Quarterly 34: 111Ð31.

DARDEL, E. (1952): LÕHomme et la terre. Press Univerzitaires deFrance, Paris.

DUNCAN, J. (1973): ÔLandscape taste as a symbol of group identityÕ,Geographical Review 63: 334Ð55.

DUNCAN, J. and LEY, D. (eds) (1993): Place/Culture/Representa-tion. Routledge, New York.

DUNLEAVEY, P. (1979): ÔThe urban basis of political alignmentÕ,British Journal of Political Science 9: 409Ð43.

DURKHEIM, E. (1938): Rules of Sociological Method. Free Press,New York.

ENTRIKIN, J.N. (1991): The Betweenness of Place. Johns HopkinsUniversity Press, Baltimore.

EYLES, J. (1985): Senses of Place. Silverbrook Press, Warrington.FOUCAULT, M. (1980): Power/Knowledge, Harvester, Brighton.FRASER, N. (1987): ÔWhatÕs critical about critical theory? Haber-

mas and genderÕ, in CORNELL, D. (ed.): Feminism as Critique:Essays on the Politics of Gender in Late-Capitalist Societies. Pol-ity Press, Cambridge, pp. 31Ð56.

GIBSON, J.J. (1979): The Ecological Approach to Visual Perception.Houghton Mifflin, Boston.

GIDDENS, A. (1982): Profiles and Critiques in Social Theory. Uni-versity of California Press, Berkeley.


Geografiska Annaler · 79 B (1997) · 1 25

GLASS, R. (1968): ÔUrban sociology in Great BritainÕ, in PAHL, R.(ed.): Readings in Urban Sociology. Pergamon, Oxford, pp. 47Ð73.

GOTTDIENER, M. (1983) ÔUrban semioticsÕ, in PIPKIN, J.S. et al.(ed.): Remaking the City. SUNY Press, Albany, pp. 101Ð14.

GREGORY, D. and WALFORD, R. (eds) (1989): Horizons in Hu-man Geography. Macmillan, London.

HABERMAS, J. (1984): The Theory of Communicative Action: Rea-son and the Rationalization of Society. Beacon Press, Boston.

HARVEY, D. (1982): The Limits to Capital. Arnold, London.HECLO, H. (1972): ÔPolicy analysisÕ, British Journal of Political

Science 2: 83Ð108.HEWITT, K. (1994): ÔÒWhen the great planes came and made ashes

of our city...Ó: towards an oral geography of the disasters of warÕ,Antipode 26: 1Ð34.

HIRST, P.Q. (1976): Social Evolution and Sociological Categories.Allen & Unwin, London.

HONNETH, A. and JOAS, H. (eds) (1991): Communicative Action:Essays on J�rgen HabermasÕs ÔThe Theory of CommunicativeActionÕ. MIT Press, Cambridge.

HUMMON, D.M. (1992): ÔCommunity attachment: local sentimentand sense of placeÕ, in ALTMAN, I. and LOW, S.M. (eds): Hu-man Behaviour and Environment, Vol 12: Place Attachment. Ple-num, New York, pp. 253Ð78.

INGOLD, T. (1992): ÔCulture and the perception of the environ-mentÕ, in CROLL, E. and PARKIN, D. (eds): Bush Base: ForestFarmÑCulture, Environment and Development. Routledge, NewYork, pp. 39Ð56.

KEITH, M. and PILE, S. (eds) (1993): Place and the Politics of Iden-tity. Routledge, London.

LEE, B.A. et al. (1984): ÔTesting the decline of community thesisÕ,American Journal of Sociology 89: 1116Ð88.

LEE, T.R. (1976): ÔCities in the mindÕ, in HERBERT, D.T andJOHNSTON, R.J. (eds): Social Areas in Cities: Vol 2 SpatialPerspectives on Problems and Policies. Wiley, Chichester, pp.159Ð87.

MEYROWITZ, J. (1985): No Sense of Place. Oxford UniversityPress, Oxford.

MOSER, C. and KALTON, G. (1976): Survey Methods in Social In-vestigation. Heinemann, London.

NEWBY, H. (1977): The Deferential Worker. Athalone, London.NISBET, R. (1966): The Sociological Tradition. Heinemann, Lon-

don.PAHL, R.E. (1970): Patterns of Urban Life. Longman, London.PHILO, C. (ed.) (1991): New Words, New Worlds: Reconceptualising

Social and Cultural Geography. Department of Geography, StDavidÕs University College, Lampeter.

PILE, S. (1993): ÔHuman agency and human geography revisited: acritique of Ònew modelsÓ of the selfÕ, Transactions of the Instituteof British Geographers 18: 122Ð39.

PILE, S. and THRIFT, N. (eds) (1995): Mapping the Subject: Geog-raphies of Cultural Transformation. Routledge, London.

POWDERMAKER, H. (1966): Stranger and Friend. Norton, NewYork.

PRED, A. (1983): ÔStructuration and placeÕ, Journal for the Theoryof Social Behaviour 13: 157Ð86.

RAPPOPORT, A. (1979): The Meaning of the Urban Environment.Pion, London.

RELPH, E. (1970): ÔAn inquiry into the relationship between phe-nomenology and geographyÕ, Canadian Geographer 14: 193Ð201.

RELPH, E. (1976): Place and Placelessness. Pion, London.RELPH, E. (1985): ÔGeographical experiences and being-in-the-

world: the phenomenological origins of geographyÕ, in SEA-MON, D. and MUGERAUR, R. (eds): Dwelling, Place and En-vironment: Towards a Phenomenology of Person and World. Co-lumbia University Press, New York, pp. 15Ð31.

SEAMON, D. (1979): A Geography of the Life World. Croom Helm,London.

SIMMEL, G. (1950): The Sociology of George Simmel. WOLFF, K.(ed.), Free Press, New York.

STACEY, M. (1969): ÔThe myth of community studiesÕ, BritishJournal of Sociology 20: 134Ð47.

STEHR, N. (1994): Knowledge Societies. Sage, Newbury Park..STEWARD, J.A. (1955): Theory of Culture Change: The Methodo-

logy of Multilinear Evolution. University of Illinois Press,Chicago.

THRIFT, N. (1987): ÔNo perfect symmetryÕ, Environment and Plan-ning D: Society and Space 5: 400Ð7.

TILLY, C. (1973): ÒDo communities actÓ, Sociological Inquiry 43:209Ð40.

T�NNIES, F. (1957): Community and Society. Michigan State Uni-versity Press, East Lansing.

TUAN, Y.F. (1976): ÔHumanistic geographyÕ, Annals of the Associ-ation of American Geographers 66: 206Ð76.

TUAN, Y.F. (1977): Space and Place. Arnold, London.WALTER, E.V. (1988): Placeways. University of North Carolina

Press, Chapel Hill.WEBBER, M.M. (1964): ÔUrban places and the non-place urban

realmÕ, in WEBBER, M.M. et al. (eds): Explorations into UrbanStructure. University of Pennsylvania Press, Philadelphia, pp.79Ð153.

WEIL, S. (1955): The Need for Roots. Beacon, Boston.WHITE, S. (1988): The Recent Work of J�rgen Habermas: Reason,

Justice and Modernity. Cambridge University Press, Cambridge.WILLIAMS, R. (1969): Keywords. Fontana, London.WIRTH, L. (1964): On Cities and Social Life. University of Chicago

Press, Chicago.YOUNG, I.M. (1990) Justice and the Politics of Difference. Prince-

ton University Press, Princeton.

reconceptualizing senses of place: social relations ... · 1997: reconceptualizing senses of place:...

Documents