Place, Culture, and the Social Amplification of Risk

Risk Analysis, Vol. 26, No. 2, 2006 DOI: 10.1111/j.1539-6924.2006.00749.x

Place, Culture, and the Social Amplification of Risk

Jeffrey R. Masuda1∗ and Theresa Garvin2

This article investigates the role of culture in the social production of risks and risk commu-

nication surrounding industrial development in a region located at a rural-urban interface. A

case study examined a public consultation that was undertaken to inform local residents about

an eco-industrial development proposal being planned near Edmonton, Alberta, Canada. The

research employed the social amplification of risk framework (SARF) to examine the relation-

ships among culture, place, and socially constructed risk. A total of 44 in-depth, semi-structured

interviews were carried out with 33 landowners (farmers, acreage owners), public officials (mu-

nicipal politicians, administrators), journalists, and industry representatives. Analysis revealed

that risk communication occurred in relation to situated experiences of place that were based

on conflicting cultural worldviews. The research shows that place is a useful component of the

SARF, providing a spatial explanation for why some people amplify, and others attenuate,

risks in locally contentious environmental debates.

KEY WORDS: Culture; geography; place; social amplification of risk

1. INTRODUCTION

Technologies that have contributed to economicprosperity and advancement of human health andquality of life have also brought a myriad of new risksto people’s everyday lives. Risks now concern soci-ety in forms as diverse as pollution or accidents fromnoxious industrial facilities, electromagnetic radiationfrom overhead power lines and cell phones, and expo-sure to infectious disease agents such as BSE and WestNile Virus. Addressing concerns about these risks in-volves using mechanisms to keep the public appraisedabout the dangers that technologies pose, and for-mulating subsequent strategies for risk management.Risk communication strategies often play out at the

1 School of Geography and Earth Sciences, McMaster University,

Hamilton, ON, Canada.2 Community, Health & Environment Research Centre, and De-

partment of Earth & Atmospheric Sciences, University of

Alberta, Edmonton, AB, Canada.∗ Address correspondence to Jeff Masuda, School of Geogra-

phy and Earth Sciences, McMaster University, Hamilton, ON,

Canada; tel: 905-525-9140 ext. 24394; [email protected].

local scale where particular groups and institutionsare both responsible for, and affected by, “risky” ac-tivities. Communication among these groups involvesgovernment authorities, industry, and the public. Riskcommunication can often result in conflict about whatare considered to be appropriate decision outcomesand acceptable levels of risk. In this way, risks arenow viewed as social constructions—what constitutesdanger depends on “who is talking to whom” (Cutter,1993; Lupton, 1999). But “who talks” is often deter-mined by politicized processes that privilege certainforms of “risk talk” over others. Risk communicationthen becomes a juxtaposition of contested ways ofmaking sense of the world. Different “sense-making”is inherently cultural, as each group seeks to advo-cate a view of risk that conforms to its way of see-ing the world. Clarifying the role of culture in riskcommunication has become a main concern in re-cent efforts to bridge risk perception research andsocial context (Pidgeon et al., 2003). This article in-vestigates the role of culture in risk communicationthrough an empirical application of the social ampli-fication of risk framework (SARF) (Kasperson et al.,

437 0272-4332/06/0100-0437$22.00/1 C© 2006 Society for Risk Analysis

438 Masuda and Garvin

1988). The research draws on qualitative data col-lected from a case study of an industrial developmentplan in Alberta, Canada and is informed by socialtheories of culture and place (Cresswell, 1996, 2004;Shields, 1991). The results show how culture is enacted“in-place” to influence the social construction of risksas they are acted out through contested place-makingdiscourses in a social conflict.

2. BACKGROUND

Three decades of intense theoretical and method-ological debate have led to approaches that investi-gate how social context is critical to understandingrisk (Rosa, 1998; Short, 1987; Wilkinson, 2001; Wynne,1992). Social context is particularly important becausethe distribution of risk has become recognized as akey contributor to many social conflicts (Beck, 1992;Giddens, 1991). Where risk perception research tradi-tionally viewed individuals as atomized units uncon-nected to a social system, we now understand risk asembedded in social context. In other words, risks aresituated within the social experiences and interactionsof individuals, groups, and institutions (Scherer &Cho, 2003). These new approaches have been incorpo-rated into both psychometric (Slovic, 1987, 2000) andcultural theories (Dake, 1992; Douglas & Wildavsky,1982) and are helping to integrate formerly disparatetechnical and socially-oriented approaches to risk.In particular, SARF, developed by Kasperson et al.(1988), has been a central research agenda in bridgingthe gap between risk perception research and socialcontext.

Since first proposed in 1988, SARF has been usedto explain how social context might influence commu-nication about risk events (Frewer et al., 2002; Pidgeonet al., 2003). In brief, according to SARF, psychologi-cal, social, and institutional factors influence risk per-ceptions and behavior through a network of sociallymediated communication channels. These communi-cation channels can be either formal, such as the me-dia, public relations campaigns, and community meet-ings, or informal, such as word-of-mouth interactionwithin social networks. Of these, the media has beenseen to be influential on risk perceptions and there-fore has received particular attention (Flynn et al.,2001). But in local risk debates, many informal sourcesof risk perceptions, including individual citizens or ac-tivist groups and institutions, can act as crucial ampli-fication or attenuation stations. These operate withincommunication channels embedded in everyday life,

receiving and sending risk signals that in turn influ-ence the risk perceptions of others.

The original work on SARF argued that riskshave meaning only to the extent that they are a re-flection of how people interact within a social con-text (Kasperson et al., 1988). Since then, phenomeno-logical studies have shown that risk not only reflectssocial context, but that culture accounts for differ-ences in the ways that risks are communicated, leadingto unique outcomes from place to place (Kasperson,1992; Pidgeon et al., 2003). As a result, subsequentiterations of SARF have incorporated culture as asupervariable through which risk amplification or at-tenuation occurs. However, to date much of the workon culture and SARF associates culture broadly asnational differences, or as a synonym for ethnicity.Most recently, case-study approaches have begun tosupport a more nuanced inclusion of culture in SARF(Pidgeon et al., 2003). For example, Ortwin Renn usesa comparative case study of solid waste incineratorsiting proposals in Germany to show how risk ampli-fication and attenuation are deliberate strategies usedto influence risk debates (Renn, 2003). He shows thatsuch strategies may be partly mediated by the culturalcontext in which local actors are situated:

All actors participating in the communication process

transform each message in accordance with their previ-

ous understanding of the issue, their application of val-

ues, worldviews, and personal or organizational norms,

as well as their own strategic intentions and goals. (2003,

p. 377)

Case studies such as Renn’s provide evidencefor the integration of situated cultural variables intomainstream risk theory, pointing to ways in whichculture can be operationalized in the local socialconstruction of risk. Further work in this area canilluminate dimensions of culture that make some indi-viduals and groups amplify and others attenuate risksin a given social context. While considerable work hasexamined the “who” aspect of this question (variablessuch as gender, occupation, age, and so on), as geogra-phers we suggest that the answers may also be foundby focusing on “where.”

Recent developments within the “new culturalgeography” have led to a definition of culture asthe social and material processes and outcomes ofcontested meanings attached to place (Martin, 2003;Massey, 2004). Critical scholars suggest that culture isbest understood contextually, as found in the taken-for-granted practices of everyday life (Cresswell,2004; Mitchell, 1995) as they occur in-place. Place

Place, Culture, and the Social Amplification of Risk 439

attachment becomes important in both reinforcingand reflecting the social construction of risk in thelocal environment. Place attachments are associatedwith family, work, leisure, and other interactions ofdaily life (Buttimer, 1993). Tensions can manifest inpower struggles when a particular place has differentmeanings to people with different social and culturalaffiliations (Martin, 2003). Critical views of place andculture have focused attention on such friction andassociated power struggles, including geographies ofresistance in social conflict (Cresswell, 1996; Pile &Keith, 1997). Therefore, place can be seen as centralto the cultural basis by which people select and inter-pret risks.

3. ALBERTA’S INDUSTRIAL HEARTLANDAND THE GEOPOLITICAL CONTEXT

This research is based on a case study of a regionalindustrial park proposed by local governments nearEdmonton, Alberta, Canada and revolves around apublic consultation program that took place between1999 and 2001. The region includes three predomi-nantly rural municipalities (Strathcona, Sturgeon, andLamont Counties) and the small industrial city ofFort Saskatchewan (population 13,000). The rural ar-eas include a mix of semi-rural farming and nonfarm-

Fig. 1. Map of the study area. The dark

line demarcates the proposed AIH

boundary, located approximately 35 km

northeast of Edmonton, Alberta, Canada.

It encompasses an area of over 190 km2.

ing residents as well as an assortment of petrochemicaland other industrial facilities. The purpose of the pub-lic consultation program was to inform local residentsof the proposal to amend zoning of a 194-km2 areato create the Alberta’s Industrial Heartland (AIH;Fig. 1). The proposal was intended to attract invest-ment in the region by large-scale petrochemical indus-try. Its aim was to build on an existing industrial infras-tructure developed over the past half-century. Despiteefforts to build community support for the AIH pro-posal, widespread opposition mobilized against theauthorities responsible for the AIH proposal. A grass-roots movement called the Heartland Citizen’s Coali-tion (HCC) utilized the consultations, local newspa-pers, and other means to air its opposition. Over time,the planning process devolved into a risk conflict be-tween local authorities and concerned landowners.Since then, steady community opposition to the AIHhas persisted and the HCC continues to try to resistplans to move ahead.

In many ways, the study area is a microcosm ofeconomic and cultural restructuring taking place inother rural locales (Abraham, 1996; Cloutier, 1996;Epp & Whitson, 2001; Gorton et al., 1998). It pro-vides an excellent example of how risks are playingout in places undergoing social and cultural change asthey move away from agricultural land uses toward


industrial and other activities. The imposition of newforms of risk on local populations has led to risk con-flict between industrial proponents and citizens whohave concerns around health, safety, and quality of life(Baxter et al., 1999a, 1999b; Wakefield et al., 2001).

In rural communities around the world, local gov-ernments have been attracted to industrial develop-ment as a means to extend local tax bases in orderto subsidize services and support the regional econ-omy (Dekel, 1994). People with cultural backgroundsbased in agriculture have traditionally seen such de-velopment as a social and economic opportunity—ameans to sustain their way of life through off-farm em-ployment. They support industry as long as it does notimpinge on their farming operations (Dekel, 1994).In contrast, more recent nonfarming rural residentshave arrived in search of a cultural “countryside” andare motivated by the aesthetic features of the locallandscape (Mitchell, 2004). These newer arrivals in-creasingly view industry as incompatible with ruralliving, and see industrial development as a threat tonature and the local quality of life (Baxter et al., 1999a,1999b). As the diversity of groups in these placesincreases, tensions between “industrial” and mixed“residential” cultures become more than simple “foror against” issues: rather, they are multifaceted andoften contradictory (Sullivan, 1994).

At the center of these multifaceted viewpointsare risks that are believed to accompany the arrival offurther industrial development. Importantly for thiscase study, perceptions of risks are influenced by pastand present experiences with local industry. For exam-ple, in August 2001, an underground storage caverncontaining ethane exploded at a petrochemical facil-ity located in the study area, sending up a plume ofblack smoke that could be seen from over 50 km away(Sturgeon Creek Post, 2001). The explosion and sub-sequent fire were caused by a structural failure in apipeline elbow joint, and burned for a total of ninedays. The incomplete combustion of ethane causedthe release of carbon soot, resulting in some residentsreporting respiratory effects. The visibility of the ac-cident, as well as delays by local authorities in pro-viding information to the public, inflamed public con-cerns about the uncertainty of industrial hazards in theregion and reinforced doubt held by some residentsabout the preparedness of local emergency responsesystems. Amid the public outcry in 2001, local author-ities tried to provide reassurance that the cavern firewas a low-risk event. While the smoke appeared to bethreatening, health and safety experts claimed thatthere would be no lasting effects on human health

(Sturgeon Creek Post, 2001). Despite these measures,many residents of the region remained concernedabout health.

The timing of the cavern fire brought industrialrisks into clear public view. The fire was a tangibleevent that conflicted with sentiments of a clean, safe,and natural environment valued by residents. The firethus served as a lightning rod for growing public cri-tique of the AIH proposal. In policy analytic terms,the cavern fire represented a tipping point for subse-quent conflict (Roe, 1994), putting risks to residentsat the center of ensuing public debate associated withindustrial development. Most importantly, the publicresponse to the fire revealed an underlying disjunc-ture in views about appropriate development betweenlocal government, industry, and residents.

4. METHODS

This research used qualitative methods to un-cover respondents’ retrospective views about risk andthe AIH proposal (Hay, 2000; Lincoln & Guba, 1985).The advantage of using qualitative methods in risk re-search is that it facilitates the reconstruction of situ-ated and negotiated realities of individuals and theirexperiences of events and processes related to riskdebates. Culture can be located in attitudes, beliefs,values, and social interactions that influence the artic-ulation of these risks (Santos & Chess, 2003). Qualita-tive information is useful in exploring unknown areasof risk theory and developing new ideas about so-cial processes. Through thick description, qualitativeresearch can be complementary to traditional quan-titative risk research by providing experiential sub-stantiation or offering insight into possible new di-rections of theories, models, and frameworks (in thiscase, SARF) (Esterberg, 2002).

Face-to-face interviews gathered in-depth infor-mation to understand the social construction of en-vironmental risk perceptions from a cross-sectionof residents3 (farmers, acreage owners, subdivisionresidents) and nonresident stakeholders, includinglocal politicians, administrators, journalists, and in-dustry representatives (Fontana & Frey, 1994). Com-bined snowball and random sampling (Creswell, 2003;Esterberg, 2002) gained the perspectives of key indi-viduals and ensured representation from the largercommunity. First, snowball sampling identified keyinformants who were known to be engaged in and

3 Residents were defined as owning homes and/or property in or

near the region defined as the AIH.


knowledgeable about the AIH proposal and publicconsultations (n = 10 resident; 10 nonresident inter-views). Additional residents recruited by random se-lection were drawn from municipal land ownershipmaps (n = 10 interviews; response rate = 70%). Thiskind of random sample mitigated sampling biases in-herent in purposive techniques by seeking out “non-respondents” or silent voices within the community(Neuman, 2000).

Between January and July 2003, a total of 33 peo-ple participated in 30 initial interviews involving onerespondent, and three interviews involving two re-

Table I. Interview Respondents

Round 1 Round 2

Interview Interview Group Interview Length of

(n = 33) (n = 14) (n = 7) Pseudonym Sex Age Occupation Residence

Resident sample

S001b,c • • Richard M 42 Farmer 10+R002b,d • Soren M 37 Farmer 5 – 10

S003 Fred M 67 Retired subdivision resident 10+R004b • • Greg M 50 Farmer 50

R004a,b,d • Jackie F – Homemaker –

R005b • • Gerry M 58 Farmer 10+R005a,b,d • Beth F – Farmer 10+R006b Nick M 82 Retired farmer 78

S007b Joseph M – Nonresident landowner –

R008b George M – Farmer –

R008a,b,d Berniece F Farmer –

R009b Karl M – Farmer –

R010 Sam M 61 Trades/farmer 45

S011 • Leah F 47 Homemaker, acreage owner 15

S012 • • Chuck M – Farmer –

R013 Corinna F 53 Nonresident landowner 53

R015b Carol F 69 Farmer 69

S016b • Erik M 74 Retired acreage owner 43

S017 • Janet F 61 Retired acreage owner 6

S018 Michael M 55 Business owner 45

S029 • Rachel F 45 Subdivision resident 18

Nonresident sample

S014 Kevin M 44 Newspaper representative 34

S019 • Dana F 42 Municipal civil servant N/A

S020 Cory M 24 Elected official N/A

S021 Darryl M 57 Municipal civil servant N/A

S022 • Brenda F 55 Municipal civil servant N/A

S023 Gus M 61 Elected official 61

S024 • Elmer M 59 Elected official N/A

S025 • Brent M 67 Elected official 67

S026 Ben M 61 Elected official N/A

S027 • Jeremy M 43 Industry representative N/A

S028 • • Bob M 49 Industrial planner N/A

S030 Wilma F 45 Municipal civil servant N/A

aHusband and wife interviews—complete demographic information was not collected for the second respondent.bRespondents recruited via random selection from land ownership maps.c“S” denotes snowball sample.d“R” denotes random sample.

spondents. In the latter case, although interviews werescheduled with one household member, the respon-dent’s partner joined the interview. An additional 14follow-up interviews conducted later in 2003 and onegroup interview in early 2004 (group size = 7) con-firmed preliminary results and probed further intoprominent themes that had emerged in the first inter-views. Table I reports demographic information forthe resident and nonresident subgroups of respon-dents. More male than female respondents were in-terviewed (22 versus 11, respectively). Respondentsranged in age from 24 to 82 years, with the majority in


their mid 50s and 60s. Occupations included agricul-ture, homemaking, private-sector industry, small busi-ness, and government. Most of the resident subgrouphad lived in the region for over 10 years, with some asmany as 50 years or more. Finally, family income ofrespondents ranged from Cdn $30,000 to $60,000+.

The sample size was sufficient to achieve a rig-orous exploration of theoretical concepts (Lincoln& Guba, 1985). Following accepted qualitative re-search practice, sampling continued only until theissues talked about by respondents were frequentlyrepeated (termed data saturation; Hay, 2000). Inter-views were held either in respondents’ homes or theiroffices. Interview length ranged from 29 to 120 min-utes, averaging approximately 45 minutes in Round 1,and from 55 to 120 minutes, averaging 74 minutes inRound 2 interviews. Interviews followed a semistruc-tured format (Hay, 2000) to ensure consistent cover-age of topics across all respondents. Topics related torespondents’ knowledge and experiences of commu-nity, place, risks and benefits of industry, and futureexpectations about the AIH proposal. The interviewguide included probes to encourage more indepth dia-logue along those lines of conversation. Beyond this,the semi-structured interview format does not nec-essarily adhere to strict protocols or question paths.Rather, the interviewer is free to modify the ques-tions being asked, take the conversation down tan-gents, and discuss seemingly unrelated subject mat-ter, depending on the desire of the respondent (Hay,2000). The benefit of this type of inductive researchmethod is that it allows unexpected and unanticipatedconcepts and themes to emerge (Hay, 2000; Lincoln& Guba, 1985).

Interviews were taped, transcribed verbatim, andimported into a qualitative data analysis softwarepackage (NVivoTM). As interviews proceeded, induc-tive analysis occurred with an initial round of free cod-ing of data to establish preliminary categories of sim-ilar units of text (called nodes). These nodes evolvedinto more concrete emergent themes with furtherdata collection (Miles & Huberman, 1994). Potentialthreats to rigor were minimized using accepted strate-gic measures for ensuring excellence in qualitativemethods (Baxter & Eyles, 1997). In a manner thatparallels practices that provide validity and reliabil-ity to quantitative research, qualitative researchersrely on several measures to enhance credibility (howclosely results represent authentic experiences of re-spondents), transferability (to what extent results canbe used in other contexts), dependability (how sta-ble results are consistent over time and space), and

confirmability (to what extent results are protectedagainst influence of biases of the researcher) (Baxter& Eyles, 1997).

A common strategy traditionally employed forqualitative data analysis involves the independentconfirmation of coding frameworks by two or moreindependent researchers (termed “interrater relia-bility”). While there has been debate around inter-rater reliability, recent critiques have resulted in agrowing number of qualitative practitioners choos-ing not to employ such techniques (see Armstronget al., 1997 and Morse, 1997 for in-depth critiquesof this method). Critics are concerned that interraterreliability fails to accomplish what it sets out to doon both logistical and epistemological grounds (Arm-strong et al., 1997; Morse, 1997). This study, instead,employed a set of data quality control measures in-cluding: (1) multimethod sampling (described above)to ensure that perspectives of both active and inactivecommunity members were obtained; (2) pilot testingof interview guides so that questions were appropriateand easily understandable; (3) member checking ofinterview transcripts to confirm accuracy of data withrespondents; (4) follow-up interviews and prolonged(approximately one year) immersion in the fieldworkto establish rapport and confirm interpretations; and(5) a community advisory committee, which checkedthe results against all four criteria for qualitative rigor.Finally, respondents’ confidentiality was safeguardedin the reporting of project results.4

5. RESULTS

The empirical findings are organized around foursocial constructs of risk perceptions related to theAIH proposal. The four social constructs linked theemergent themes from the data set of Round 1 in-terviews5 and represented risks to: (1) life, (2) home,(3) prosperity, and (4) community. Table II presentsthe 11 emergent themes by relevance to each of thefour social constructs. Frequency counts of the dataindicate those themes that were more prominent top-

4 Funding for this project was provided by the Social Science & Hu-

manity Research Council of Canada (SSHRCC) and the Cana-

dian Agricultural Rural Communities Initiative (CARCI). To

meet funding agency and research requirements, this research

passed ethics review and respondents’ identities have been pro-

tected through the use of pseudonyms in all data analysis and

published research results.5 Round 2 and focus-group interviews were used to confirm existing

themes and are not employed in the quantitative reporting in this

article.


Table II. Emergent Themes from Round 1 Interviews (n = 30)

Social Emergent #Interviews #Mentions in

Construct Theme Definition Occurred All Interviews

Life Danger • Industry transforms the region as a “safe” place to a “risky” place. 23 100

Health • Health is threatened by pollution associated with industry. 18 67

Safety • Risks are managed by improvements in industrial practices. 15 27

Total=194 mentions

Home Rural idyll • Industry upsets “country living”—a pace of life removed from

urban problems and in greater connection with “nature.”

18 51

Heritage • Industry threatens traditional, familial roots with the land and the

agricultural community.

11 27

Geography • Attributes of the physical landscape provide incentives for further

industrial development.

15 22

Employment • Employment benefits outweigh risks. 14 20

Total=124 mentions

Prosperity Stigma • Farming practices and property values threatened by the stigma

associated with industrial presence.

16 38

Economy • Potential gains in property values from industrial development

speculation. Industry has spin-off benefits to local businesses and

politicians.

18 46

Total=84 mentions

Community Marginalization • Increased regional focus on industry leads to marginalization of the

local community.

8 11

Philanthropy • Industry supports the wider population through its philanthropic

and economic activities.

13 24

Total=35 mentions

ics during the interviews. First, the “#Interviews Oc-curred” column provides insight into its pervasivenessin local discussions about risks. That more respon-dents talked about a particular risk theme suggeststhat it was an important part of the debate aroundthe AIH proposal. Second, the “#Mentions in AllInterviews” column points to areas of risk that atleast some respondents talked about multiple timesin the Round 1 interviews. That a certain respondentmore frequently discussed particular risk perceptionsmeans that some aspects of risk were especially im-portant to him/her. It should be cautioned, however,that while prevalence is often a good indicator of theimportance of a particular theme, it does not necessar-ily discount less frequently mentioned themes (Baxter& Eyles, 1997). Rather, each theme carries unique in-sights into views about both individual and collectiverisk perceptions. A theme with few mentions may pro-vide valuable perception into a social phenomenonthat is not readily recognizable or commonly knownin the wider social context of the risk debate.

The frequency counts also facilitated comparisonof risk perceptions across the four social constructs.While the data show that risks to life are the most of-ten discussed social construct (194 mentions), risks

to other aspects of everyday life were also impor-tant to respondents. These included risks to home(124 mentions), prosperity (84 mentions), and com-munity (35 mentions), which collectively accountedfor more than half of the themes mentioned (243 outof 437 mentions). The breadth of risk perceptionsacross social constructs indicates that respondentsrecognized the broader dimensions of risks stemmingfrom the AIH proposal. This broader view means thatthe AIH proposal was not only seen as a threat justto life, but also a threat to “ways of life.”

5.1. Social Constructs and Themes

5.1.1. Life

The first social construct centered on the impactsof intensified industrial activity to life. On one side,the prospect of industrial accidents evoked concernsabout the potential threats to the lives of respondentsand their families, and on the other side, assurancesabout the effectiveness of regulations to mitigate haz-ards. A total of 194 mentions among three themes(“danger,” “health,” “safety”) made risks to life themost widely talked about social construct in the re-


sults. The most prominent theme (“danger”) centeredon the threat of living in close proximity to industrialfacilities (23 interviews; 100 mentions). Dangers wereviewed in a spatialized sense: proximity to industrytransformed the region from a “safe” place to liveinto a “risky” place. Respondents linked personal ex-periences with chemicals, noises, fires, and explosionswith fears about increased hazards that would accom-pany further industrial development. Among these,the cavern fire figured prominently for Fred as an ex-ample of the inability of emergency response systemsto manage risks:

[Emergency response] wasn’t very well handled, the

EMS people in the counties, in the city of Fort

Saskatchewan did a very poor job, notifying us about

what was going on . . . the RCMP came and told us to

get out of here, there could be real problems. I said, well

I live right over there. She said, “well if I was you, I’d

go to Edmonton for the day.” (Fred, retired subdivision

resident; 003)6

In a related theme (“health”), respondents recog-nized the long-term health effects of industrial activ-ities, especially among people who lived downwindfrom airborne industrial emissions (18 interviews; 67mentions). Health concerns ranged from increasedcases of asthma and other respiratory illness, to cancerand uncertain neurological disorders associated withpollution. For evidence of these effects, respondentslike Richard pointed to visible indicators on the locallandscape such as smog, residues, rusting metal, andanimal health:

We really don’t know how bad the pollution is. I mean

there’s talk of increased cases of asthma and you hear

lots of horror stories about specific examples of people

with respiratory problems so bad they can’t really live

in that area. And not specifically [human health, but]

cattle are getting sick and aborting. (Richard, retired

farmer; 001)

In contrast to the threats to life represented by thedanger and health themes, other respondents empha-sized a third theme (“safety”) focused on the successof industrial operations in preventing accidents andensuring public safety (15 interviews; 27 mentions).Respondents such as Ben chose to emphasize the viewthat industry has done a good job in employing tech-nologies and regulations to manage risks:

And I can say over the past 35 years, industries have

improved. Their technologies have improved, their pro-

cesses have improved, they’ve become more environ-

mentally conscious. (Ben, elected official; 026)

6 Identifying information for quotations includes the respondent’s

pseudonym, description, and interview number.

Assurances about safety focused on monitoring, re-search, industrial operations, and hazard manage-ment that “normalized” visible signs of industry withappeals to the effectiveness of emergency response.Brenda recalled her belief that local authorities hadthe cavern fire under control as she went about herroutine daily activities:

I remember the morning [of the fire] we were going to

church and I saw this huge black plume, so we went

for a ride. Of course the RCMP had the roads blocked.

They said it’s a fire but it’s not out of control. It’s under

control. OK. Nobody was injured. Everything was fine.

Trucks are there. Fine. I went home and made breakfast.

(Brenda, municipal civil servant; 022)

5.1.2. Home

The second social construct centered on theimpact of industrial intensification on the home-space of the rural countryside. A total of 124 men-tions among four themes (“rural idyll,” “heritage,”“geography,” “employment”) reflected how industrymight threaten or benefit experiences of home. Inthe first theme (“rural idyll”), respondents relayedconcerns about the negative image that industrialencroachment imposed on the aesthetic features ofthe countryside (18 interviews; 51 mentions). Accord-ing to this theme, industrial nuisances such as traffic,noise, and smells disrupted the peaceful serenity of liv-ing the country life. This theme was especially impor-tant to respondents like Janet who lived on acreagesand in rural subdivisions:

And we bought the property with the idea that we

would be involved with nature, away from noise and

pollution. We’re very much environmentalists our-

selves, and very involved in bird watching and camp-

ing and canoeing. We really like the outdoors, so we

hoped this would be a special place for us to retire in,

but it hasn’t turned out that way. (Janet, retired acreage

owner; 017)

For the farming community, attachment to homewent well beyond the aesthetic pleasures of countryliving. The second theme (“heritage”) emphasized theattachment to a lifestyle that was associated with thelong-term tenure of the region’s agricultural commu-nity (11 interviews; 27 mentions). Many farmers in theregion had roots dating back to 19th-century home-steaders, with lands passed down from generation togeneration. To Greg, the AIH proposal representedthe “last straw” of a continuing economic transforma-tion associated with the replacement of agriculture


with other economic activities over the past severaldecades:

I’ve lost my motivation. And that’s the hard part. It

almost brings tears to my eyes that some place that my

grandparents worked so hard for, and passed on from

generation to the next generation, that I’m just going

to abandon it. (Greg, farmer; 004)

As under the life construct, not everyone agreedthat industrial intensification was inherently threat-ening. In contrast to the views of the rural idyll andheritage, some respondents saw the area as a “home”to industry—in particular because of local geographiccharacteristics and potential employment opportuni-ties for industrial workers. In fact, half of respondentsacknowledged that features of the physical and sociallandscape (“geography”) were considered to be an as-set not only to the region, but the province as a whole(15 respondents; 24 mentions). From this perspective,industrial development “made sense” based on theregion’s advantageous physical, economic, and socialresources. The availability of low-cost petrochemicalfeedstock, land, water, existing industrial infrastruc-ture, an educated workforce, and a pro-business polit-ical climate made the region an attractive, and hencelogical, place to invest. The region is located at a lat-itude with a relatively cold climate and has large saltcaverns in the subsurface, making it ideal for the stor-age of a variety of petrochemical products (AIHA,2002). Infrastructure (e.g., rail, pipelines) has devel-oped alongside industry and provides incentives forfurther industrial development. For Brent, these at-tributes, in conjunction with the adoption of an eco-industrial planning philosophy (AIHA, 2002; Ayres &Ayres, 1996; Frosch, 1992), seemed to justify furtherindustrial development:

I think there’s only one [issue], and that’s pack it [with

industry]. We have the infrastructure set up in this west

end of the county. And the infrastructure I’m referring

to is the water, the gas, the railways, and power. We

have all that in place for it. (Brent, elected official; 025)

One selling point for AIH promoters was the idea thatthe AIH would create a real estate market for currentresidents. They argued that increased employmentopportunities would draw people to the area, creat-ing a housing demand and pushing up home prices.This belief held that residents no longer comfortablewith nearby industry would sell their homes to futureemployees who would want to live close to their newplace of work (AIHA, 2002). However, when askedabout this strategy, Erik was skeptical that anyonewould want to invest in his property:

Interviewer: I have read that they say that one possible

solution is that with all this industrial expansion there’ll

be people that work for industry who will want to buy

these houses to be close to industry. Do you think that’s

a realistic solution?

Erik (acreage owner; 016): Well, would you put money

into a house that sits on land that any time can become

industrial land?

The fourth theme (“employment”) centered on re-ports of the long-term dependency of farmers on workin local plants. These respondents were optimistic thatthe AIH would be a “land of opportunity” (14 inter-views; 20 mentions) for residents of the region. Indus-try would allow farmers to “stay home” rather than beforced to find off-farm work outside the region. Ac-cording to Nick, the community should understandand accept the consequences of the local industrialeconomy because the benefits outweighed any poten-tial risks:

Well, you can’t have everything. You got to put up with

some of those things if you want to benefit. I worked [in

industry] for 10 years, and they treated me pretty good,

and I got no complaints about them. [Industry] does

the best they can, and the communities benefit from it

considerably. So what more do you want? (Nick, retired

farmer; 006)

5.1.3. Prosperity

The third social construct provided an economicperspective of risk and the AIH proposal. Most re-spondents agreed that the AIH, if implemented,would result in a gradual shift from agricultural andresidential to industrial land use over the course ofseveral years or decades. A total of 84 mentions intwo contrasting themes (“stigma,” “economy”) dealtwith economic risks and benefits of the AIH’s pro-posed industrial growth strategy. In the first theme(“stigma”), respondents recognized the potential forpollution to affect agricultural productivity and prop-erty values. This led to a sense of worry about personaleconomic well-being (16 interviews; 38 mentions). Forfarmers, this worry focused on anxiety about the eco-nomic viability of agriculture in general. For example,Gerry was concerned that airborne emissions wouldthreaten his ability to retain organic certification forfarming operations:

If a farmer can be a certified organic farmer, and then

somebody’ll come along and say, “Okay, I want to test

your land for these chemicals,” and there may be traces

of [pollutants]. There’s nothing that you can do, because

the land that we live in has this pollution everywhere.

(Gerry, farmer; 005)


Similarly, industrial stigma associated with both pol-lution and the reconfiguration of land from residen-tial or agricultural to industrial led to concerns aboutproperty values. For many properties, the real estatemarket depended on the aesthetic features of the nat-ural setting of the river valley in which many residen-tial properties were located. Since industry impingedon the natural setting of these properties, respondentslike Erik felt that resale potential may be constrainedto an industrial market, where it may take years tofind a buyer:

What we are afraid of now is that when the time comes

that we have to move, that suddenly we find we have

no buyer for our land. I must say it has become worth-

less. If the time comes that we have to move the only

alternative we have now is to get off the land, board the

house up or dismantle it, and hope that at some future

day the land will be sold to industry. I don’t think it will

be sold in the next few years, it probably won’t be sold

in a decade or two. (Erik, retired acreage owner; 016)

In contrast to concerns about stigma, otherrespondents chose to emphasize the regional ad-vantages that might accompany the AIH proposal(“economy,” 18 interviews; 46 mentions). Spin-offbenefits for local businesses provided a strong incen-tive to foster industrial investment in the region. Re-spondents like Cory also recognized the political cap-ital to be gained from a robust local economy:

If you’re working at the Tim Horton’s for six bucks

an hour, and the more people we have drive through

the city, the more hours I’m going to get and the more

money they can pay me. And to go right to the other

end, if you’re the City Manager of Fort Saskatchewan,

the more industry I can attract here and the bigger I

can grow the city, the more money I’m going to make

and the better it’s going to look on my resume. (Cory,

elected official; 020)

5.1.4. Community

The final social construct identified by respon-dents related to the idea of community sustainability.A total of 35 mentions in two themes (“marginaliza-tion,” “philanthropy”) offered contrasting views ofthe future community impacts of industrial intensi-fication. Importantly, both themes carry a differentdefinition of community. These different definitionscoincided with whether respondents viewed industryas a risk or benefit. In the first theme (“marginaliza-tion”), respondents emphasized the local residentialcommunities located within the AIH boundaries. Re-spondents thought the AIH proposal had forgottenabout the “little person,” leaving local community

members on the periphery of decision making (8 inter-views; 11 mentions). For Gerry, authorities’ emphasison the economic aspects of industry at the expense ofcommunity concerns fostered resentment:

[Authorities] do everything for the mighty buck, there’s

no doubt about it. I don’t think they concern them-

selves with doing things for the benefit of the whole

community. They locate where it is the best for them,

and without a concern for the people that it affects.

(Gerry, farmer; 005)

In contrast, the second theme under the commu-nity social construct (“philanthropy”) defined com-munity in broader terms. It emphasized a wider com-munity encompassing the entire regional population(13 interviews; 24 mentions). Respondents like Danasuggested that the regional “community” had beenstrengthened by the multitude of charitable acts doneby industry over the years. As such, to complain aboutany risks that accompany industry is to “bite the handthat feeds you”:

[Industry] has done, as far as I’m concerned, an excep-

tional job supporting the community. Putting dollars

back into the community. In community projects, and

recreation projects, and tons of money. Computers to

the schools. They’ve spent a lot of money. There’s been

a lot of money back into the community. (Dana, munic-

ipal civil servant; 019)

Together, the four social constructs show that riskperceptions shape several layers of life and developthrough complex and contested interpretations of thepositive and negative impacts of the AIH proposal.Because of the diversity in respondents’ views, the is-sues of concern cannot be interpreted as simply “for”or “against” the AIH proposal, but must be seen inthe light of a complex, multifaceted juxtaposition ofperspectives.

5.2. The Social Amplification of Risk in the AIH

Further analysis of the 11 emergent themes aboveprovides insight into the relationship among risk per-ception, culture, and the social amplification of risk.Table III groups the themes according to whether theyrepresent high or low risk perceptions within the do-mains of the four social constructs. Six of the themesreflected high risk (danger, health, rural idyll, her-itage, stigma, marginalization; 294 mentions). Risksto life included industrial accidents that could harmpeople living near plants. Risks to home included un-sightly industrial activity that could upset “countryliving.” Risks to prosperity included the impacts of


Table III. The 11 Emergent Themes Grouped by High and Low

Risk Perceptions Across the Four Social Constructs

Risk Perception Social Construct Theme

HIGH Life Danger294 mentions Health

Home Rural idyllHeritage

Prosperity StigmaCommunity Marginalization

LOW Life Safety139 mentions Home Geography

EmploymentProsperity EconomyCommunity Philanthropy

pollution and rezoning on property values. Finally,risks to community marginalized residents living nearindustry for the benefit of the wider population.

In contrast, five of the themes represent low riskperceptions (safety, geography, employment, econ-omy, philanthropy; 139 mentions). These themes em-phasized respondents’ views that industrial develop-ment was not only a low risk, but also a potential ben-efit to the region. Risks to life were minimized withassurances about the safety of industrial operations.Risks to home were discounted by a rationale thatindustry would create a “land of opportunity” wherepeople who wanted to leave could easily benefit fromthe influx of workers who would want to live closeto work. Risks to prosperity were countered by eco-nomic benefits of industry in the form of employmentopportunities and spinoffs to local businesses. Finally,risks to community were dismissed by pointing to thephilanthropic contributions industry makes to localsports, culture, and quality of life.

5.2.1. Risk Amplification and Attenuation

The results above suggest that risk perceptionswere not isolated within the minds of individuals, butmanifested as threats to shared “ways of life” thatincluded people’s sense of belonging and well-beingin the community at large. In this way, risks weredisseminated through social networks via communi-cation channels (e.g., public meetings, newspapers,informal interactions) so that individual risk percep-tions ultimately fed into a community sense of “risk-iness” (Driedger et al., 2002). Social constructions oflife, home, prosperity, and community framed dis-cussions of industrial development, pitting concerns

about risks against assurances about the benefits ofindustry. In the interviews, respondents identified spe-cific ideas or events, using either high or low riskperceptions, to contribute to risk amplification or at-tenuation. For example, respondents communicatedcontrasting risk perceptions when considering the po-tential for a major industrial disaster. On one side,some like Cory had faith that the emergency responseefforts were sufficiently equipped to mitigate any cir-cumstance, no matter how remote:

There’s so many safety precautions and checks and bal-

ances in there that the chances of having something

that catastrophic happen . . . definitely there’s a greater

chance of getting hit by a car or struck by lightning or

whatever. (Cory, elected official; 020)

On the other side, respondents like Fred recountedstories based on past experiences of similar types ofdisaster and the potential for such events to occuragain:

On one occasion, a propane tank truck caught fire on

the road [in the AIH]. I don’t know if you’ve heard of

a “BLEVE”, a BLEVE is a boiling liquid expanding

vapour explosion. [A BLEVE] happened in 1983 down

in Illinois. A propane tank car caught fire. When the

car exploded, the tank car flew 3330 feet. The burn area

from the fire was a thousand feet. We’ve got these types

of cars and products 30 metres from our house, and I

don’t need more of it. (Fred, rural subdivision resident;

003)

Importantly, all of the discourse around risk anddecision making, including that of authorities, cen-tered on aspects of risk that were not based on ab-solute or quantitative risk evaluations. Rather, theywere generated by underlying community or institu-tionally mediated worldviews that were in favor of oragainst the AIH proposal. In a circular and reinforc-ing manner, worldviews influenced a particular riskperception and that perception served to reinforceand legitimate the preexisting worldview. In this way,the same risks were amplified as “riskier” by some re-spondents and attenuated as “less risky” by others inaccordance with preexisting perspectives.

5.2.2. Who Amplifies, Who Attenuates?

The next step in the analysis sought to confirmpatterns of residents’ and nonresidents’ risk percep-tions. Figs. 2 and 3 present total frequency counts ofhigh and low risk perceptions charted for residentsand nonresidents, respectively.7 In both figures, high

7 Respondents were separated based on the groupings in Table I.


∗These interviews contain themes from both husband and wife respondents.

Fig. 2. Risk perceptions of resident

interviews (n = 18). High risk perception

themes (dark gray) versus low risk

perception themes (light gray). Sample

means are reported for respondents listed

as risk amplifiers (amp; n = 14). F =Farmer; A = Acreage owner; S = Rural

subdivision; N = Nonresident landowner.

∗Note that Respondent #014 did not have any risk themes coded in the transcript.

Fig. 3. Risk perceptions of nonresident

interviews (n = 12). Low risk perception

themes (light gray) versus high risk

perception themes (dark gray). Sample

means are reported for respondents listed

as risk attenuators (att; n = 7). E =Elected official; P = Public servant; I =Industry; M = Media.

risk perceptions appear as positive frequency counts(dark gray bars) and low risk perceptions as negativefrequency counts (light gray bars). Not unexpectedly,the results show a clear difference between residentsand nonresidents.

Fig. 2 lists each resident from highest to low-est frequency count of high risk perception men-tions/interview. For example, Fred (rural subdivisionresident; 003) mentioned 41 incidents of high risk per-ceptions and zero of low risk perceptions. Within this

group, 14 of 18 residents/couples8 were categorizedas “risk amplifiers,” reporting substantially more highthan low overall risk perceptions. Among the 14 riskamplifiers in Fig. 2, high risk perceptions were re-ported an average of 18 times per interview, versusonly three mentions of low risk perceptions. Impor-tantly, Fig. 2 also reveals that acreage owners (listed

8 Because the three interviews of couples were coded together, they

are grouped together in this analysis.


as “A”) and residents of subdivisions (listed as “S”)almost consistently reported high risk perceptionsmore frequently than farmers (listed as “F”). The onlyexceptions included Greg (#004) and Gerry (#005),both of whom were not primarily farmers by occupa-tion. The difference between farmers and nonfarmersmight be explained by the closer interdependency be-tween farmers and industry and suggests that countryresidents saw industry as a greater threat than didfarmers. This disjuncture is consistent with literaturedocumenting the transformation of traditionally agri-cultural regions via the influx of incomers yearningto escape from “urban” problems (pollution, noise,traffic, crime) in a countryside dominated by industri-alized agriculture. These newcomers are recognizedas being less tolerant of intrusive activities such as in-dustry (Baxter et al., 1999a, 1999b; Kaplan & Austin,2004; Mitchell, 2004). The finding also supports theidea that risk perceptions were not based solely onthe location in which people lived, but on more com-plex cultural experiences related to how people wereattached to place.

Fig. 3 lists nonresidents from highest to lowestfrequency of low risk perception mentions/interview.Within this group, 7 out of 12 nonresidents were cate-gorized as “risk attenuators,” reporting more low thanhigh risk perceptions. Among the seven risk atten-uators in Fig. 3, low risk perceptions were reportedan average of nine times per interview, versus onlytwo mentions of high risk perceptions. Notably, theseven risk attenuators each had vested political inter-ests related to industry, occupying positions in localgovernments that were responsible for carrying outthe AIH proposal. Among these, three respondentsstood out as “strongly” risk attenuating (Respondents#026, #020, #028). These three officials were all moreclosely affiliated with industry either as a result of po-litical affiliations or direct involvement in the planningof the AIH proposal.

Notably, the average frequency count of high riskperceptions among the 14 risk amplifiers in Fig. 2(x = 18) doubles the average frequency count oflow risk perceptions among the 7 risk attenuatorsin Fig. 3 (x = 9). The disparity in the strength ofrisk perceptions suggests that residents on averagewere stronger in their reports of risks than nonresi-dents. With the exception of the three “strong” riskattenuators, nonresidents tended to talk less aboutrisk perceptions from a perspective that recognized,and sometimes sympathized with, the concerns ofresidents.

Furthermore, the results support the idea thatwithin-group trends of residents and nonresidents canbe interpreted as resulting from, and contributing to,shared risk perceptions. In SARF, individuals, groups,and institutions can act as amplification or attenua-tion stations by sharing risk perceptions within socialnetworks. Whether individuals will amplify or atten-uate risks, then, depends on cultural worldviews thatare influenced by the social network in which theyare situated. Risk amplifiers mainly included residentswhose worldview was based on place attachments tofarms and acreages, as well as personal livelihoodin the countryside. Residents’ worldviews were per-ceived to be under threat from the prospect of indus-trial encroachment that would accompany the AIHproposal. Amplification stations could therefore in-clude a group of residents communicating with eachother through newspapers or informal social networksto spread word about their concerns and actions. Forexample, the HCC formed out of informal networksbuilt among concerned residents who rallied neigh-bors who lived within or nearby the AIH-designatedboundaries. Members of the HCC such as Rachel re-ported that they communicated regularly with neigh-bors by telephone, in groups, and at public meetingsabout their concerns:

I felt some responsibility to inform the community. I

mean, it was important. I felt that the neighbors should

know. So that’s what I spent most of my time on, is going

out to the public. Trying to inform people. I made a lot

of phone calls. (Rachel, rural subdivision resident; 029)

In contrast, risk attenuators included mainly non-residents whose worldviews were influenced by polit-ical or economic motivations related to their jobs orpositions in government. Nonresidents’ worldviewswere at a scale that encompassed the wider con-stituency of their respective municipalities or busi-nesses. According to this worldview, industrial de-velopment would be good for the regional economy,serving the interests of all citizens of the four munic-ipalities. Attenuation stations could include groupsof officials and industrial representatives who cametogether in committees whose purpose was to moveforward on the AIH proposal. For example, the AIHsteering committee (of which several of the nonresi-dent respondents of this study were members) actedas an attenuation station by directing the AIH Asso-ciation to promote the benefits of the AIH proposalthrough public relations materials, newspaper adver-tisements, and during public meetings. According to


Brent, this committee was responsible for directingthe coordinated development and promotion of theAIH proposal to serve the interests of like-mindedorganizations:

[The AIH Directors] were involved, the government

was involved, the federal government was involved, the

provincial government, they were all involved in that.

They were at our meetings. We’re happy that’s the pro-

cess we used. And we give them the information. They

had an opportunity to review it, and come back to us,

and tell us what they think it was. And that’s the route

we took. (Brent, elected official; 025)

Lastly, it is important to note that several respon-dents reported risk perceptions that contradicted theresident or nonresident group to which they were as-signed in this study. Among residents, these includedRespondents #006, #015 (equal high and low risk per-ceptions), and #012 (more low risk perceptions). Twoof these respondents were retired farmers and one wasa spouse of a now deceased farmer, all of whom re-ported having depended on industry at some point intheir lives to help sustain the family farm income. Thisdependency may have influenced views of the role ofindustry, leading to lower risk perceptions. In the in-terviews, respondents like Gus believed that farmersgenerally supported the AIH proposal as they wereconvinced that the greater good of the local economyoutweighed the concerns of other residents:

In the agricultural community they have been more

ready to accept industry and not question the environ-

ment in which they live. As readily as people who live

in multi-developed subdivisions who have come out of

Edmonton. (Gus, elected official; 023)

Exceptions in the nonresident group included Re-spondents #022, #027 (equal high and low risk percep-tions), #019 and #024 (more high risk perceptions),and #014 (neither high nor low risk perceptions).These respondents included former officials and in-dustry representatives who were not directly involvedin the planning of the AIH proposal or who had tiesto the local communities through other political orsocial avenues. In the interviews, these respondentscame across as nonaligned in the debate. Like Jeremy,their risk perceptions arose in relation to being sym-pathetic to community concerns:

From my standpoint, when I drive down this road and

I go all the way to Fort Saskatchewan with [name of

company] there on the roadside, knowing what I know

about the industry, and seeing all the things coming up

from the plant, I’ll be honest, I wonder sometimes, my-

self. So why should we expect any more sympathy and

understanding from the public when they have even

less knowledge than ourselves? (Jeremy, industrial rep-

resentative; 027)

Together, these exceptions suggest that concernabout risk is more complicated than “just” being a res-ident or nonresident. The overall trend suggests thatrisk amplification and attenuation depends on expe-riences related to cultural worldviews that are under-pinned not only by one’s own place attachments, butthose of others as well. Whether respondents ampli-fied or attenuated risks depended on whether theysaw the region generally as a place to live, or as aplace for economic progress.

In sum, the results suggest that risks are based onbeliefs about what constitutes a threat not only to lifebut also to home, prosperity, and community. As such,risks such as those surrounding the AIH proposal can-not be quantified or simply reduced to a probability ofinjury or death, but must be understood in the contextof all aspects of everyday living. Furthermore, risksare situated at the intersection of competing viewsabout whether the region is suitable or not suitablefor industry.

6. DISCUSSION

This research set out to improve our under-standing of culture within the social amplification ofrisk framework (SARF). Contemporary understand-ings of culture see it as a complex political processthrough which meaning is constructed and negotiated(Duncan & Ley, 1993). For example, critical culturalgeography identifies culture as politics by anothername (Mitchell, 2000). By this definition, culture iscomplex and contested, enacted in the places of ev-eryday life. The AIH case study provided the routeto access risk meanings through the everyday livedexperiences of people participating in risk communi-cation within the context of a public consultation pro-gram. Qualitative analysis showed that risks formedthrough four social constructs that encompassed as-pects of risks related to life, home, prosperity, andcommunity. Further investigation revealed that riskperceptions were shaped by place attachments andrelated to worldviews. Respondents interpreted risksthrough the context of a “safe place” or a “place fordevelopment,” depending on whether they communi-cated from the perspective of the local resident com-munity or the wider economic well-being of the re-gion. The centrality of place in framing respondents’discussions about risk therefore provides a way to op-erationalize culture in SARF.


6.1. Place and Risk Theory

The relationship between place and culture hasimplications for the spatial articulation of risk theory(Cutter et al., 2001, 2003), which emphasizes the situ-ated experiences of risk in people’s everyday lives. Re-cent advances in social theory of space and place un-derstand that space is not just a passive container, butan active dimension of social relations (Gregory, 1994;Lefebvre, 1991; Soja, 1996). Space is a social agent thatacts together with culture to create places that are “ac-tualized and endowed with meaning” (Shields, 1991,p. 64). Places are thus constructed and defended ac-cording to contested understandings of what does ordoes not “belong” (Cresswell, 1996). In the case ofthe AIH, risk did not belong “in-place” to residents’whose worldviews differed from broader municipalgoals of development held by government officials(Phillips, 1998).

The importance of place depended in part on thescale of worldviews in relation to the risks and bene-fits of the AIH proposal. Risk amplifiers saw the AIHproposal in the context of an “insider’s” place attach-ment that was based on lived experiences—industrialencroachment was seen as a threat to the “lived-in”places in which people carry out their lives. This grouptended to include those people with strong emotionalattachments to home and local community as a “safeplace.” When faced with what they viewed as an inap-propriate intrusion on home life and personal liveli-hood, residents came together to form the HCC. TheHCC provided a means to express opposition to theAIH proposal through the collective communicationof high risk perceptions and an attempt to deter publicacceptance of industrial intensification.

In contrast, risk attenuators saw the AIH pro-posal in the context of an “outsider’s” place attach-ment that was based on a commitment to economicprogress. Respondents who were outsiders to the lo-cal community9 viewed industrial development as anopportunity to create an “industrialized” place thatwould be a boon to the entire region. This grouptended to include authorities who had a vested polit-ical attachment to serving the wider community as a“place for development.” An eco-industrial park cen-tered on petrochemical manufacturing would createemployment opportunities and business spinoffs forthe entire region. In their effort to promote the AIH tothe general public, these people established formal in-stitutions (e.g., the Association, Steering Committee).

9 But perhaps insiders to the wider municipality at large.

These institutions subsequently implemented a strat-egy to communicate the benefits of industry and pro-vide assurances about risk management that wouldconvince the local population of the virtues of theAIH proposal.

During the consultations, it became clear thatmany residents disagreed with the pro-industry per-spective. For example, many residents distrusted ex-pert claims about safety and the environmentally con-scious approach of the AIH proposal. Exceptions tothis opposition were either few and far between, orsilent during the public consultations and this study.10

Similarly, supporters of the AIH proposal discountedhigh risk perceptions as scientifically unfounded andbased solely on the vested interests of selfish landown-ers. These contrasting viewpoints were communicatedin public debates where they came face to face in ef-forts to define the region as a place for home or a placefor industry. On the surface, the debates were aroundperceptions of the risks and benefits of the AIH pro-posal. However, at a deeper cultural level, this debaterepresented competing definitions of place.

6.2. Bringing Place to SARF

The original impetus of SARF was to bridge thegap between technical concepts of risk and the so-cial, cultural, and individual response structures thatshape public experiences of risk. Place is an importantpart of culture that bridges that gap. Geographers inparticular have given explicit attention to local socialand cultural geographies in the examination of riskperceptions (Baxter et al., 1999a, 1999b; Bickerstaff& Walker, 2001; Cutter et al., 2003; Garvin, 2001;Luginaah et al., 2002; Wakefield et al., 2001). Casestudies such as this one have provided evidence thatrisks are socially constructed according to a complexarray of localized factors specific to particular places.This research has built on earlier work in cultural the-ory that has sought to understand how differencesin risk perceptions can block effective communica-tion, thus supporting the view of Covello and Johnson(1987):

We can see how people from different organizational

and institutional contexts may talk past each other be-

cause they maintain different culturally conditioned

perceptual filters. These admit concerns relevant to

their day-to-day experience, while blocking those ideas

that are irrelevant or would place obstacles in the way

10 Although the potential for this was mitigated by the random

selection process.


of their daily lives. By making the basis of this disagree-

ment explicit, we might alert policy makers to unfore-

seen problems before they become political crises as

well as suggest some new approaches to their solution.

(1987, p. 21)

Examining risk from a spatialized perspectivehelps to make disagreements explicit. A place-sensitive application of SARF shows how risk andplace operate in local risk conflicts to explain howpeople talk “past” each other through different cul-tural filters. Understanding of cultural worldviews asthey are based in place attachments sheds light on theoperation of culture in risk communication.

This perspective shows how residents, when facedwith a future industrial vision for the landscape, am-plified risks to defend their view of place. Risk ampli-fication occurred on the basis of place attachmentsembedded in family, tradition, and a rural idylliclifestyle. This process created an idealized residen-tial place where industrial intrusion was not welcome.Ironically, subtle differences between agricultural andpostagricultural rural residents were also contentious(Salamon & Toratore, 1994; Sharp & Smith, 2002).Incompatibilities between agricultural practices (e.g.,pesticide and manure use) against the desire of coun-try residents to have a high environmental quality oflife were divisive issues among respondents in thisstudy. Despite such tensions, these two groups ap-peared to have aligned their positions in ways thatmutually amplified the risks of industry.

In contrast to residents, risk attenuators empha-sized the suitability of the region for eco-industrialgrowth and development. With a focus on develop-ment and risk management, the AIH proposal at-tempted to create a sense of place based on industrialinvestment and productivity. The decline of agricul-ture as the dominant economic driver exposed theregion to new development initiatives. The AIH pro-posal set the stage for one iteration of a longstandingdebate between rural landowners and industrialdevelopers that collectively make up the chang-ing rural identity of much of the North Americanwest (Salamon & Toratore, 1994; Singelmann, 1996;Winson, 1997).

7. CONCLUSION

This research has contributed to the socialamplification of risk framework (Kasperson et al.,1988) by showing how risk communication is a cul-tural process that operates in place. The AIH pro-

posal was conceived as a means for local govern-ments to control the direction and benefits of futurepetrochemical industrial development in a specific re-gion in Alberta. However, over time, it became muchmore than that. By creating a forum for public con-sultation, authorities transmitted the inherent dan-gers and uncertainties associated with industrial de-velopment (e.g., pollution, industrial accidents, risksto quality of life) into the public arena. In a com-munity previously sensitized to industrial risk, andwith a distrust of local risk managers, the public con-sultation program effectively provided a venue forrisk debate and controversy. The result was the socialproduction (and amplification) of risk. Underlyingthis controversy were competing “places” groundedin deeply held cultural worldviews of residents andnonresidents.

It is worth reiterating that risk amplification andattenuation are not mutually exclusive. To categorizepeople as exclusively “risk amplifiers” or “risk atten-uators” runs the danger of oversimplifying social pro-cesses. The perspective of place in this study resiststhis temptation by recognizing that the cultural land-scape is complex, and often contradictory. Despitethe necessary simplifications taken for the purpose ofanalysis, it has provided empirically robust evidencefor the validity of case-study approaches in advanc-ing the linkages between risk and place that consti-tute one part of the cultural component of the socialamplification of risk framework.

Further case studies that explicitly incorporateplace may play an important role in further elucidat-ing the cultural “supervariable” in SARF and helpingto strengthen our understanding of the links betweenrisk perceptions and social context. The work pre-sented here identified distinct patterns of risk percep-tions that were mediated by place attachments. Futurestudies can include place attachment as a constitutivevariable among factors that influence the amplifica-tion and attenuation of risk (e.g., gender, age, socioe-conomic status). Such studies should include quanti-tative models that will help to elucidate the mecha-nisms within which place operates in local contexts.According to Rob Shields (1991), “[S]patialisation(sic) . . . plays an important role in social causality andshould be incorporated into social science discussionsof cause and effect” (1991, p. 261). A quantitativelydetermined incorporation of place would be a helpfulcontribution to risk communication research because,to this point, place theory has been largely overlookedas a constitutive variable of the social context ofrisk.


ACKNOWLEDGMENTS

This study was funded exclusively through re-search grants from the Social Sciences and Human-ities Research Council of Canada and the CanadianAgricultural Rural Communities Initiative. We wishto thank two anonymous reviewers and our sponsors,as well as the participants who made this study pos-sible. Copies of study findings and recommendationsare available at http://research.eas.ualberta.ca/carci.

REFERENCES

Abraham, C. (1996). Agriculture and rural employment: Anoverview. Canadian Journal of Agricultural Economics, 44,497–505.

AIHA (Alberta’s Industrial Heartland Association). (2002). Al-berta’s Industrial Heartland Complementary Area StructurePlan Background Report.

Armstrong, D., Gosling, A., Weinman, J., & Marteau, T. (1997).The place of inter-rater reliability in qualitative research: Anempirical study. Sociology, 31(3), 597–606.

Ayres, R. U., & Ayres, L. (1996). Industrial Ecology: Towards Clos-ing the Materials Cycle. London: Edward Elgar.

Baxter, J., & Eyles, J. (1997). Evaluating qualitative “rigour” ininterview analysis. Transcripts of the Institute of British Geog-raphers, 22, 505–525.

Baxter, J. W., Eyles, J. D., & Elliott, S. (1999a). “Something hap-pened”; The relevance of the risk society for describing thesiting process for a municipal landfill. Geografiska Annaler,81B(2), 91–109.

Baxter, J. W., Eyles, J. D., & Elliott, S. J. (1999b). From siting prin-ciples to siting practices: A case study of discord among trust,equity and community participation. Journal of EnvironmentalPlanning and Management, 42(4), 501–525.

Beck, U. (1992). Risk Society: Towards a New Modernity. London:Sage.

Bickerstaff, K., & Walker, G. (2001). Public understandings of airpollution: The “localisation” of environmental risk. Global En-vironmental Change, 11, 133–145.

Buttimer, A. (1993). Geography and the Human Spirit. Baltimore:Johns Hopkins University Press.

Cloutier, S. (1996). Employment in agriculture and closely relatedindustries in rural areas: Structure and changes, 1981–1991. Pa-per presented at the International Symposium: Perspectives onRural Employment held October 11 to 14, 1995, in Coaticook,Quebec.

Covello, V. T., & Johnson, B. B. (1987). The social and culturalconstruction of risk: Issues, methods, and case studies. In V.T. Covello & B. B. Johnson (Eds.), The Social and CulturalConstruction of Risk. Dordrecht: Reidel.

Cresswell, T. (1996). In Place Out of Place: Geography, Ideol-ogy, and Transgression. Minneapolis: University of MinnesotaPress.

Cresswell, T. (2004). Place: A Short Introduction. Oxford:Blackwell.

Creswell, J. (2003). Qualitative, Quantitative, and Mixed MethodsApproaches. Thousand Oaks: Sage.

Cutter, S. L. (1993). Living with Risk. London: Edward Arnold.Cutter, S. L., Boruff, B. J., & Shirley, W. L. (2003). Social vulnerabil-

ity to environmental hazards. Social Science Quarterly, 84(2),242–261.

Cutter, S. L., Hodgson, M. E., & Dow, K. (2001). Subsidized in-equities: The spatial patterning of environmental risks andfederally-assisted housing. Urban Geography, 22(1), 29–53.

Dake, D. (1992). Myths of nature: Culture and the social construc-tion of risk. Journal of Social Issues, 48(4), 21–27.

Dekel, G. P. (1994). The fiscal impact of development: A regionalperspective in Canada. Land Use Policy, 11(2), 128–141.

Douglas, M., & Wildavsky, A. B. (1982). Risk and Culture: An Es-say on the Selection of Technical and Environmental Dangers.Berkeley: University of California Press.

Driedger, S. M., Eyles, J., Elliott, S. D., & Cole, D. C. (2002). Con-structing scientific authorities: Issue framing of chlorinateddisinfection byproducts in public health. Risk Analysis, 22(4),789–802.

Duncan, J., & Ley, D. (Eds). (1993). Place/Culture/Representation.London: Routledge.

Epp, R., & Whitson, D. (2001). Writing Off the Rural West: Global-ization, Governments, and the Transformation of Rural Com-munities. Edmonton: University of Alberta Press.

Esterberg, K. (2002). Qualitative Methods in Social Research. NewYork: McGraw Hill.

Flynn, J., Slovic, P., & Kunreuther, H. (2001). Risk, Media andStigma: Understanding Public Challenges to Modern Scienceand Technology. London: Earthscan.

Fontana, A., & Frey, J. H. (1994). Interviewing: The art of science.In N. Denzin & Y. E. Lincoln (Eds.), Handbook of QualitativeResearch (Ch. 22). Thousand Oaks: Sage.

Frewer, L. J., Miles, S., & Marsh, R. (2002). The media and genet-ically modified foods: Evidence in support of social amplifica-tion of risk. Risk Analysis, 22(4), 701–711.

Frosch, R. A. (1992). Industrial ecology: A philosophical introduc-tion. Proceedings of the National Academy of Sciences, USA,89, 800–803.

Garvin, T. (2001). Analytical paradigms: The epistemological dis-tances between scientists, policy makers, and the public. RiskAnalysis, 21(3), 443–455.

Giddens, A. (1991). The Consequences of Modernity. Cambridge:Polity Press.

Gorton, M., White, J., & Chaston, I. (1998). Contested space: Therural idyll and competing notions of the good society in theUK. In A. Light & J. Smith (Eds.), The Production of PublicSpace: Philosophy and Geography II. Landham: Rowman andLittlefield.

Gregory, D. (1994). Geographical Imaginations. Oxford: Blackwell.Hay, I. (2000). Qualitative Research Methods in Human Geography.

Oxford: Oxford University Press.Kaplan, R., & Austin, M. E. (2004). Out in the country: Sprawl and

the quest for nature nearby. Landscape and Urban Planning,69, 235–243.

Kasperson, R. E. (1992). The social amplification of risk: Progressin developing an integrative framework. In S. Krimsky & D.Golding (Eds.), Social Theories of Risk (Ch. 6). Connecticut:Praeger.

Kasperson, R. E., Renn, O., Slovic, P., Brown, H. S., Emel, J., Goble,R., Kasperson, J. X., & Ratick, S. (1988). The social amplifi-cation of risk: A conceptual framework. Risk Analysis, 8(2),177–187.

Lefebvre, H. (1991). The Production of Space. Oxford: Blackwell.Lincoln, Y., & Guba, E. (1985). Naturalistic Inquiry. Beverly Hills:

Sage.Luginaah, I. N., Taylor, S. M., Elliott, S. J., & Eyles, J. (2002). Com-

munity responses and coping strategies in the vicinity of apetroleum refinery in Oakville, Ontario. Health and Place, 8,177–190.

Lupton, D. (1999). Risk. London: Routledge.Martin, D. G. (2003). “Place-framing” as place-making: Constitut-

ing a neighborhood for organizing and activism. Annals of theAssociation of American Geographers, 93(3), 730–750.

Massey, D. (2004). Geographies of responsibility. Geografiska An-naler, 86B(1), 5–18.

Miles, M. B., & Huberman, A. M. (1994). Qualitative Data Analysis:An Expanded Sourcebook, 2nd ed. Thousand Oaks: Sage.


Mitchell, C. J. A. (2004). Making sense of counterurbanization.Journal of Rural Studies, 20, 15–34.

Mitchell, D. (1995). There’s no such thing as culture: Towards areconceptualization of the idea of culture in geography. Trans-actions of the Institute of British Geographers, 20(1), 102–116.

Mitchell, D. (2000). Cultural Geography: A Critical Introduction.Malden: Blackwell.

Morse, J. M. (1997). Perfectly healthy, but dead; the myth of inter-rater reliability. Qualitative Health Research, 7(4), 445–447.

Neuman, W. L. (2000). Social Research Methods: Qualitative andQuantitative Approaches, 4th ed. Needham Heights: Allyn &Bacon.

Phillips, M. (1998). The restructuring of social imaginations in ruralgeography. Journal of Rural Studies, 14(2), 121–153.

Pidgeon, N., Kasperson, R. E., & Slovic, P. (2003). The Social Am-plification of Risk. Cambridge: Cambridge University Press.

Pile, S., & Keith, M. (1997). Geographies of Resistance. London:Routledge.

Renn. (2003). Social amplification of risk in participation: Two casestudies. In N. Pidgeon, R. E. Kasperson, & P. Slovic (Eds.),The Social Amplification of Risk (pp. 374–401). Cambridge:Cambridge University Press.

Roe, E. (1994). Narrative Policy Analysis. Theory and Practice. Lon-don: Duke University Press.

Rosa, E. A. (1998). Metatheoretical foundations for post-normalrisk. Journal of Risk Research, 1(30), 15–44.

Salamon, S., & Tornatore, J. B. (1994). Territory contested throughproperty in a midwestern post-agricultural community. RuralSociology, 59(4), 636–654.

Santos, S. L., & Chess, C. (2003). Evaluating citizen advisory boards:The importance of theory and respondent-based criteria andpractical implications. Risk Analysis, 23(2), 269–279.

Scherer, C. W., & Cho, H. (2003). A social network contagion theoryof risk perception. Risk Analysis, 23(2), 261–267.

Sharp, J. S., & Smith, M. B. (2002). Social capital and farm-ing at the rural-urban interface: The importance of non-farmer and farmer relations. Agricultural Systems, 76, 913–927.

Shields, R. (1991). Places on the Margin: Alternative Geographiesof Modernity. London: Routledge.

Short, J. F. Jr. (1987). Social dimensions of risk: The need for a soci-ological paradigm and policy research. American Sociologist,12, 167–172.

Singelmann, J. (1996). Will rural areas still matter in the 21st cen-tury? (or) Can rural sociology remain relevant? Rural Sociol-ogy, 61(1), 143–158.

Slovic, P. (1987). Perceptions of risk. Science, 236, 280–285.Slovic, P. (2000). The Perception of Risk. London: Earthscan.Soja, E. W. (1996). Thirdspace: Journeys to Los Angeles and Other

Real-and-Imagined Places. London: Blackwell.Sturgeon Creek Post. (2001). Faulty elbow caused BP fire. Novem-

ber, 14. 6, 5.Sullivan, W. C. III (1994). Perceptions of the rural-urban fringe:

Citizen preferences for natural and developed settings. Land-scape and Urban Planning, 29, 85–101.

Wakefield, S., Elliott, S. J., Cole, D., & Eyles, J. (2001). Environ-mental risk and (re) action: Air quality, health and civic in-volvement in an urban industrial neighbourhood. Health andPlace, 7, 163–177.

Wilkinson, I. (2001). Social theories of risk perception: Atonce indispensable and insufficient. Current Sociology, 49(1),1–22.

Winson, A. (1997). Does class consciousness exist in rural com-munities? The impact of restructuring and plant shutdowns inrural Canada. Rural Sociology, 62(4), 429–453.

Wynne, B. (1992). Risk and social learning: Reification to engage-ment. In S. Krimsky & D. Golding (Eds.), Social Theories ofRisk (ch. 12). Connecticut: Praeger.

Place, Culture, and the Social Amplification of Risk

Documents

Transcript of Place, Culture, and the Social Amplification of Risk