INTRODUCTION
Species and ecosystems under constant pressure from a rapidly growing human population (Lindsey et al. Reference Lindsey, Alexander, du Toit and Mills2004) have led to increased need for effective planning in nature conservation in order to preserve vulnerable species and their habitats, as well as society's willingness to preserve these species and habitats (Hampicke Reference Hampicke1994; Lindsey et al. Reference Lindsey, Alexander, du Toit and Mills2005; Romanach et al. Reference Romanach, Lindsey and Woodroffe2007). To find acceptable solutions for species management in areas with human-wildlife conflicts, multidisciplinary socioecological approaches are required (Shivik Reference Shivik2006; McShane et al. Reference McShane, Hirsch, Trung, Songorwa, Kinzig, Monteferri, Mutekanga, Thang, Dammert, Pulgar-Vidal, Welch-Devine, Brosius, Coppolillo and O'Connor2011; Linnell & Boitani Reference Linnell and Boitani2012). In this paper, we examine whether the presence of carnivores (wolves Canis lupus, bears Ursus arctos, lynx Lynx lynx and wolverine Gulo gulo) is correlated with people's environmental value orientation (ranging from ecocentric to anthropocentric) and how environmental value orientation in turn relates to the acceptance of carnivores.
Environmental value orientation is the relatively stable expression of how people evaluate the environment, and is rooted in individual basic values, having both emotional and cognitive (knowledge) components (Fransson & Garling Reference Fransson and Garling1999; Schultz & Zelezny Reference Schultz and Zelezny1999; Milfont & Duckitt Reference Milfont and Duckitt2010; Heberlein Reference Heberlein2012 b). Here, we used the new environmental paradigm (NEP), which is a conceptualized measure of people's environmental attitudes, beliefs and values, to measure people's environmental value orientation (Dunlap et al. Reference Dunlap, Vanliere, Mertig and Jones2000; Ardahan Reference Ardahan2012). Previous studies (Kaltenborn & Bjerke Reference Kaltenborn and Bjerke2002; Rauwald & Moore Reference Rauwald and Moore2002; Hunter & Rinner Reference Hunter and Rinner2004; Berenguer et al. Reference Berenguer, Corraliza and Martin2005; Kaltenborn et al. Reference Kaltenborn, Andersen, Nellemann, Bjerke and Thrane2008; Luo & Deng Reference Luo and Deng2008; Ardahan Reference Ardahan2012) and conceptual and psychometric testing (Stern et al. Reference Stern, Dietz and Guagnano1995; Pierce et al. Reference Pierce, Sarason, Sarason, Solky-Butzel and Nagle1997) have shown that high NEP scores are correlated with pro-environmental values (Dunlap et al. Reference Dunlap, Vanliere, Mertig and Jones2000). Low NEP scores are associated with anthropocentrism, where humans view nature solely as a source of food and water. A person with a high anthropocentric score is likely to be sceptical to the likelihood of a future ecological crisis or severe climate changes (Gerhard Reference Gerhard2004). In contrast, high NEP scores are associated with ecocentrism, where nature is seen to have intrinsic value regardless of human utilitarian needs, and that humans are an integral part of the environment (Kortenkamp & Moore Reference Kortenkamp and Moore2001; Hunter & Rinner Reference Hunter and Rinner2004). However, we should view the concept of anthropocentrism to ecocentrism not as a dichotomy, but as a gradient.
Recovery of large carnivore populations is a major challenge in many parts of the world (Noss et al. Reference Noss, Quigley, Hornocker, Merrill and Paquet1996; Linnell et al. Reference Linnell, Broseth, Odden and Nilsen2010). Situations where the human-wildlife conflict depends on large carnivores are well suited to investigating society's environmental value orientation because carnivore presence varies spatially (for example between communities or countries) and attitudes towards large carnivores are generally extreme. As conflicts mount due to the recovery of carnivores, wildlife managers have to consider a range of stakeholders’ interests including carnivore conservation, predation on livestock, predation of game species, killing of hunting dogs, and the fear of human injuries and fatalities (Naughton-Treves et al. Reference Naughton-Treves, Grossberg and Treves2003; Manning et al. Reference Manning, Gordon and Ripple2009; Treves et al. Reference Treves, Jurewicz, Naughton-Treves and Wilcove2009; Barnowe-Meyer et al. Reference Barnowe-Meyer, White, Davis, Smith, Crabtree and Byers2010; Marucco & McIntire Reference Marucco and McIntire2010).
In Scandinavia, the public has periodically opposed the protection of large carnivores (Bjerke & Kaltenborn Reference Bjerke and Kaltenborn1999; Bjerke et al. Reference Bjerke, Vitterso and Kaltenborn2000; Sjölander-Lindqvist Reference Sjölander-Lindqvist2007) followed by extensive illegal hunting of all large carnivore species (Andren et al. Reference Andren, Linnell, Liberg, Andersen, Danell, Karlsson, Odden, Moa, Ahlqvist, Kvam, Franzen and Segerstrom2006; Persson et al. Reference Persson, Ericsson and Segerstrom2009; Liberg Reference Liberg, Guillaume, Wabakken, Pedersen, Hobbs and Sand2012). In particular, farmers and hunters felt they were excluded from participation and involvement in carnivore management by central authorities (Ericsson & Heberlein Reference Ericsson and Heberlein2003; Skogen Reference Skogen2003). In other parts of the world, Lindsey et al. (Reference Lindsey, Alexander, Frank, Mathieson and Romanach2006) showed how local participation by trophy hunters in Africa contributed to wildlife conservation, while hunters in Wisconsin did not support wolf conservation in their area (Treves & Martin Reference Treves and Martin2011; Bruskotter & Fulton Reference Bruskotter and Fulton2012). In Scandinavia, hunters generally do not support the conservation of large carnivores (Ericsson & Heberlein Reference Ericsson and Heberlein2003; Roskaft et al. Reference Roskaft, Handel, Bjerke and Kaltenborn2007). These hunters see themselves as important wildlife stewards, and express the opinion that local knowledge is more important than scientific knowledge for setting hunting quotas for large carnivores. Thus, Scandinavian hunters may not support protection of carnivores, but they are still concerned with animal welfare and the role of carnivores in the ecosystem (Kaltenborn et al. Reference Kaltenborn, Andersen and Linnell2013).
The recovery of carnivore populations worldwide has been accompanied by an increase in studies that investigate the human and institutional responses to these recoveries (Bisi et al. Reference Bisi, Kurki, Svensberg and Liukkonen2007; Roskaft et al. Reference Roskaft, Handel, Bjerke and Kaltenborn2007; Heberlein et al. Reference Heberlein and Ericsson2008; Gangaas et al. Reference Gangaas, Kaltenborn and Andreassen2013). However, few studies have examined relationships between basic life values such as general environmental value orientation and attitudes towards large carnivores (Bjerke et al. Reference Bjerke, Odegardstuen and Kaltenborn1998 a). Our study addresses these gaps. Specifically, we hypothesize that: (1) the presence of carnivores predicts people's environmental value orientation, and (2) people's value orientation in turn predicts attitudes towards carnivores. More specifically, we expect that people with more ecocentric value orientation will have higher acceptance for carnivores than people with an anthropocentric value orientation. However, due to the ‘not in my back yard’ (NIMBY) effect (Ericsson et al. Reference Ericsson, Bostedt and Kindberg2008; Lidskog & Elander Reference Lidskog and Elander1992), we predict that even ecocentric respondents, who might be positive to protection of large carnivores, do not accept carnivores in their home municipality.
METHODS
Study area
Agriculture traditions
The sample of respondents who replied to our questionnaire covered all the Scandinavian Peninsula consisting of Norway and Sweden (see below). These two neighbouring countries, with a long common history, have several characteristic differences with regard to traditions, agriculture and environmental policies. In terms of domestic animals vulnerable to predation by large carnivores, the semi-domestic free-ranging reindeer (Rangifer spp.) dominates the northern part of Scandinavia, while free-ranging domestic sheep (Ovis spp.) are found all over Norway, but not in Sweden. Moreover, sheep graze freely and unattended in uncultivated land in Norway; Norwegian agriculture is dependent on this uncultivated land. The Swedish human population currently stands at 9.6 million people (26 people km−2), while there are 5.1 million people in Norway (18 people km−2) (see Statistics Norway at www.ssb.no and Statistics Sweden at www.scb.se), and 84 % of the human population is clustered around cities in both countries (see www.globalis.no). The resulting rural population density is 4.2 people km−2 in Sweden and 2.9 people km−2 in Norway. The proportion of total land area used for agricultural purposes is 6.4 % in Sweden, and 2.7 % in Norway (see www.globalis.no). Agriculture represents approximately 3.2 % of total employment in Sweden and 2.5 % in Norway.
As Norway is not a European Union (EU) member, the country has maintained a district policy where governmental subsidies are available to support farming, animal husbandry and human settlements throughout the country (Sorensen Reference Sorensen2003; Otterlei & Sande Reference Otterlei and Sande2010). Norwegian agriculture is among the most highly subsidized in Europe (see www.oecd.org). When Sweden became member of the EU, their subsidies increased as they had no tradition of subsidized agriculture. However, Swedish subsidies are almost half of what the Norwegian government pays, and farmers in Sweden are mainly compensated for the loss of income by differences in food prices. Therefore, subsidies in Sweden are not directed towards food production or animal husbandry, as they are in Norway (see www.jordbruksverket.se).
Forestry is another important industry in both countries. The state and state-owned companies own 7 % of productive forestland in Norway, and 17 % in Sweden. Private forest corporations and family enterprises own 77 % of the forest in Norway, and 56 % of the forest in Sweden (Barklund Reference Barklund2009). In 2010, the forest industry contributed 0.6 % of gross domestic product (GDP) in Norway (see www.ssb.no) and 2.4 % GDP in Sweden (see www.skogstyrelsen.se).
Management and politics
Carnivore conservation in Scandinavia shifted from eradication and state bounties in early 1900 (Linnell et al. Reference Linnell, Broseth, Odden and Nilsen2010) to total protection, followed by current year-to-year regulations based on hunting quotas and adaptive management (Ericsson et al. Reference Ericsson, Heberlein, Karlsson, Bjarvall and Lundvall2004; Bull et al. Reference Bull, Nilsen, Mysterud and Milner-Gulland2009). In Sweden, the brown bear received protection in 1913 and the lynx in 1927 (see www.artfakta.se). In Norway, regulations were adopted in the 1970s to protect all carnivores (Swenson et al. Reference Swenson, Wabakken, Sandegren, Bjarvall, Franzen and Soderberg1995). Today, both Norwegian and Swedish management allow harvest of all carnivores, either by quotas or hunting licences. Swedish membership of the EU is accompanied by additional guidelines for carnivore management, including minimum carnivore population sizes that have to be reached before carnivore populations can be controlled by hunting. The quotas or licences are set annually depending on population estimates based on monitoring programmes adapted to each carnivore species, including scat analyses, snow tracking, radio-collared individuals or DNA analyses (Hedmark & Ellegren Reference Hedmark and Ellegren2007; Linnell et al. Reference Linnell, Odden, Andren, Liberg, Andersen, Moa, Kvam, Broseth, Segerstrom, Ahlqvist, Schmidt, Jedrzejewski and Okarma2007). The Norwegian management goals for carnivore population abundance are much lower than in Sweden (www.rovviltportalen.no; www.naturvardsverket.se).
Carnivores
After total absence in the late 1960s, wolves recolonized Scandinavia in 1978 (Wabakken et al. Reference Wabakken, Sand, Liberg and Bjarvall2001) and the population today is approximately 330 individuals (Wabakken et al. Reference Wabakken, Kojola, Maartmann, Strømseth, Flagstad, Åkesson and Zetterberg2013). The brown bear population was as low as 130 individuals in the 1920–1930s, but had increased to more than 2500 individuals in 1995 (Swenson et al. Reference Swenson, Wabakken, Sandegren, Bjarvall, Franzen and Soderberg1995) and was c. 3000–3500 individuals in 2012 (www.rovdata.no; www.naturvardsverket.se). The wolverine and lynx populations are c. 1000 and 2000 individuals, respectively (Norwegian Directorate for Nature Management, unpublished data 2010; Swedish Environmental Protection Agency, unpublished data 2010). Sweden has the major share of large carnivores, but none of these species are evenly distributed across the peninsula (Appendix 1, Fig. S1, see supplementary material at Journals.cambridge.org/ENC).
Survey
Environmental attitudes
The original new environmental paradigm (NEP) consisted of a 12-item questionnaire (Dunlap & Vanliere Reference Dunlap and Vanliere1978), but a smaller number of questions can produce sufficient inter-item reliability and validity (Bjerke & Kaltenborn Reference Bjerke and Kaltenborn1999; Dunlap Reference Dunlap2008; Kaltenborn et al. Reference Kaltenborn, Andersen, Nellemann, Bjerke and Thrane2008). Here we used a seven-item version with a five interval Likert-type scale: ‘highly disagree’, ‘disagree’, ‘neither agree nor disagree’, ‘agree’ and ‘highly agree’. The seven NEP questions measure different aspects of environmental beliefs, which reflect how people think about environmental changes and how they view their role in the environment (Table 1). The NEP score for each respondent was estimated as the mean of the seven answers, where highly disagree was given the value 1, and highly agree the value 5 (Dunlap et al. Reference Dunlap, Vanliere, Mertig and Jones2000).
Table 1 The questions used in the questionnaire to estimate the new environmental paradigm (NEP) answered in a five interval Likert-type scale; ‘highly disagree’, ‘disagree’, ‘neither agree nor disagree’, ‘agree’, and ‘highly agree’. The NEP-value for each respondent was estimated as the mean of the seven answers where ‘highly disagree’ was given the value 1 and ‘highly agree’ the value 5. Significant differences between Norway and Sweden are indicated: *p = 0.001 and **p < 0.001.
Attitudes towards carnivores
All respondents were asked if they found the carnivore situation in their own country to be one of the following: ‘Too many carnivores’, ‘Suitable amount of carnivores’ or ‘Too few carnivores’, and whether they could accept illegal hunting of wolf/lynx/bear/wolverine. In addition, the respondents had to choose one of three statements: (1) ‘I cannot under any circumstances accept carnivores in my country’; (2) ‘I can accept that carnivores establish freely’; or (3) ‘I can accept that carnivores establish in my country under certain circumstances (e.g. zoning, removal of carnivores killing sheep)’. Those that could not accept carnivores under any circumstances (question 1) were additionally asked if they were willing to conduct illegal hunting to prevent carnivore establishment (Likert scale; Appendix 1, Fig. S2, see supplementary material as Journals.cambridge.org/ENC). Those that accepted carnivores became established freely (question 3) were additionally asked if they could live in an area inhabited by wolf/lynx/bear/wolverine (Likert scale).
Sampling design
We used a geographically stratified sampling where four or five individuals in each municipality in the two countries responded to the questionnaire (434 and 290 municipalities in Norway and Sweden, respectively). This sampling ensured that all locations (rural or urban) were presented and weighted equally in the analyses, enabling correlation of human attitudes described from the questionnaire with variation in carnivore densities.
A data collection company (www.norstat.no) conducted the survey by telephone, and based its sample on existing public registers. The interviewer followed a strict protocol dictated by standard research ethics, including presenting the purpose of the study, and stating that participation was entirely voluntary, how long the interview would take, and how the results would be used. There were 2521 respondents (1507 in Norway and 1014 in Sweden), a response rate of c. 15 %, which is common for telephone surveys. However, as the sample represents a very small proportion of people living in high density areas such as cities and suburban areas, it cannot be used as a measure of the mean opinion of people living in a specific region (for example a county) or country.
Statistical procedures
Predicting NEP
To describe the variation in the NEP scores we used a linear mixed model (Zuur et al. Reference Zuur, Ieno, Walker, Saveliev and Smith2009), where county was considered a random effect and the following predictor variables as fixed effects: number of carnivores living in the respondent's county, number of free-ranging sheep, whether the respondent thought they lived in a county with strong hunting traditions (Likert scale 1–5), if the respondent had a negative experience with carnivores such as loss of sheep or dog, home country, and covariates describing the respondent (such as age, gender, education level and income). The covariates can be important predictors of attitudes, but we do not discuss them further. Education level was used as a continuous variable from secondary school, high school, university undergraduate and/or graduate degree (MSc, PhD). Carnivore densities are based on data from national wildlife databases (see www.dirnat.no and www.naturvardsverket.se). The most parsimonious model was selected using a step-wise procedure and AIC (Burnham & Anderson Reference Burnham and Anderson1998). First we selected the two most parsimonious models including only the carnivore predictors, and only the demographic covariates. The selected carnivore and covariate models were then combined in a full model with hunting traditions, negative experience with carnivores and country. We then used a backwards selection procedure based on AIC values. Finally, we entered each of the removed variables, one at a time, to see if they could improve the selected model. From all models, we selected that with the lowest AIC as our final most parsimonious model. All analyses were done using statistical procedures available in R 2.13.1 (http://cran.r-project.org/). GIS tools were used to visualize the NEP scores on maps with ArcMap 10 (ESRI).
Attitudes towards carnivores
To analyse how the respondents’ NEP scores were associated with attitudes towards carnivores, we categorized individual scores as anthropocentric when NEP < 2.8, neutral when 2.8 ≤ NEP < 4.0, and ecocentric when NEP ≥ 4.0, these threshold values being based on the distribution of the NEP scores.
RESULTS
New environmental paradigm (NEP)
The most parsimonious model explaining the individual mean NEP (Fig. 1) included country, age of respondent, gender and education level (AIC = 4894.14; Tables 2 and 3), while the second best model also included whether the respondent had any negative experience with carnivores (ΔAIC = 3.45). Neither income (ΔAIC > 25.85) nor carnivore densities (ΔAIC > 26.31) improved the model.
Figure 1 The distribution of the mean NEP scores from the seven NEP questions in the survey. The black bars to the right represent the respondents characterized as ‘ecocentric’, the black bars to the left represent the respondents characterized as ‘anthropocentric’, and the white bars the ‘neutral’ respondents.
Swedes reported higher ecocentric values than Norwegians in all seven NEP questions (Table 1; Fig. 2), resulting in higher mean NEP scores (± SE) in Sweden (3.65 ± 0.02; Table 2) than Norway (3.38 ± 0.02; Table 3). However, Norwegian mean NEP scores were also skewed towards an ecocentric orientation. Males were more ecocentric than females, education level was positively correlated with NEP scores (namely a higher degree of ecocentrism), while age was negatively correlated with NEP (namely a higher degree of anthropocentrism; Table 3).
Table 2 AIC values for the seven models with lowest AIC, and results for some of the other models used to explain mean NEP scores. The predictor ‘neg experience’ indicates the respondent has had a negative experience with carnivores, such as the loss of sheep or dogs; ‘sheep’ are the number of free-ranging sheep in the county of the respondent. Education level was used as a continuous predictor, as this provided models with ΔAIC ≥ 14.01 lower than models where education level was a categorical variable with four levels.
Table 3 Parameter estimates from the most parsimonious model predicting the mean NEP value of individuals.
Figure 2 Map showing mean NEP scores for each county in Norway and Sweden. The darker the colour of the county the more ecocentric the mean value.
NEP and attitudes towards carnivores
With regard to the main question about how respondents found the carnivore situation in their own country, Swedes were more positive towards carnivores (χ2 = 190.52, df = 2, n = 2521, p < 0.001). In particular, fewer Swedes thought there were too many carnivores in their country, and acceptance of carnivores increased with increasing (more ecocentric) NEP scores. In Sweden, 187 people found there were ‘too many’ carnivores (mean NEP score ± SE was 3.40 ± 0.03), 664 reported that carnivore numbers were ‘suitable’ (3.64 ± 0.02) and 163 felt there were ‘too few’ carnivores (3.93 ± 0.04). The corresponding numbers from Norway were ‘too many’: 679 (3.20 ± 0.02), ‘suitable’: 659 (3.45 ± 0.02) and ‘too few’: 169 (3.75 ± 0.04). The mean NEP scores differed significantly between Norway and Sweden (p < 0.005). This national difference was still present when comparing the frequency of respondents to various scenarios of living with carnivores: 7 % of the Norwegians and 2 % of the Swedes could not accept carnivores under any circumstances. However, the majority of both Norwegians and Swedes answered that they could accept carnivores under certain conditions (Norway 80 %; Sweden 71 %) and that they could accept carnivores being established freely (Norway 13 %; Sweden 27 %; Table 4).
Table 4 The number of respondents (n) in Norway and Sweden answering that they: (1) do not accept carnivores under any circumstances, (2) accept carnivores are freely established, or (3) accept that carnivores become establish under certain conditions. For each of these questions, we show the percentage of the respondents belonging to the anthropocentric–ecocentric gradient. For those respondents that could not accept carnivores under any circumstances, we also asked if they were willing to conduct illegal hunting themselves to prevent establishment (1.1), and, for those who accepted carnivores became freely established, we also asked if they were personally willing to live in carnivore areas (2.1).
Among the respondents that could not under any circumstances accept carnivores in their own country (100 Norwegians and 21 Swedes), 4% and 14% of Norwegians and Swedes, respectively, reported an ecocentric attitude, while 41 % and 29 % of Norwegians and Swedes, respectively, reported an anthropocentric view. However, these differences were not significant (χ2 = 3.88, df = 2, n = 121, p = 0.144; Table 4). Out of the 100 Norwegians, 36 % were willing to shoot carnivores illegally, while, in Sweden, only 10 % who could not accept carnivores were willing to illegally hunt carnivores (Table 4).
Among those respondents accepting that carnivores become established freely, there was no difference in NEP scores between countries (χ2 = 4.26, df = 2, n = 470, p = 0.119; Table 4). In both countries, about 25% of respondents were unwilling to live with carnivores (Table 4), although respondents were more ecocentric in Sweden than Norway (χ2 = 7.48, df = 2, n = 115, p = 0.024; Table 4).
Most respondents (1207 in Norway and 723 in Sweden) could accept carnivores under certain conditions, such as in protected areas or if carnivores that kill livestock were removed. More Swedes that accepted carnivores under certain conditions had ecocentric NEP scores than Norwegians (χ2 = 89.72, df = 2, n = 1930, p < 0.001; Table 4).
DISCUSSION
NEP scores were negatively correlated with acceptance of illegal hunting and the wish for fewer predators, corroborating previous studies (Bjerke & Kaltenborn Reference Bjerke and Kaltenborn1999). NEP scores were mainly affected by national differences and demographic covariates, and not associated with the presence of carnivores. Ecocentric value orientation was more common among Swedes than Norwegians. Even Swedes that answered in a rather anthropocentric way, such as acceptance of illegal hunting or unwillingness to live in carnivore areas, showed a somewhat more ecocentric orientation than Norwegians. A person may well have a positive view of nature and carnivores in general, and at the same time support behaviour that is not associated with an ecocentric value orientation. The question is how central the attitude(s) is to that person's fundamental belief system. People change their behaviour when personally affected, or they may actually have a different personal norm than the attitude they report (Heberlein Reference Heberlein2012 a). This gap between people's attitudes and how they might act also depends on the extremity of the behaviour, and the correlation between behaviour and general environmental value orientation is often indirect and weak (Manfredo Reference Manfredo2008; Heberlein Reference Heberlein2012 a).
As expected, respondents who did not favour carnivores in their country reported more anthropocentric attitudes than those who could accept carnivores becoming freely established in their country. However, 25 % of those respondents claiming that carnivores should be able to become established were unwilling to live in a carnivore area themselves. This is likely an example of the common discrepancy between general attitudes toward a subject or issue and specific behavioural intentions. Environmental value orientation represents an environmental world view; it is possible that individuals with an ecocentric orientation believe that unrestricted carnivore establishment is more equitable than having certain carnivore zoning areas resulting in more burden for inhabitants of these areas.
Even though our geographical stratified sampling does not describe the mean Norwegian or Swede, the differences in environmental value orientation (NEP scores) and carnivore attitudes between Norway and Sweden are noteworthy. Both countries have democratically-elected governments with comparable economy and education levels (Otterlei & Sande Reference Otterlei and Sande2010). They also share the same carnivore populations and the same obligations to large carnivore protection, though their carnivore population abundances differ significantly (Woodroffe et al. Reference Woodroffe, Thirgood and Rabinowitz2005; Otterlei & Sande Reference Otterlei and Sande2010). Moreover, contemporary Scandinavia is thought to be homogenous across a range of social and economic factors, including law enforcement, political stability, political trust, government effectiveness, rule of law, redistribution of goods and benefits, as well as control of corruption and accountability (Svallfors Reference Svallfors1999; Sinani Reference Sinani, Stafsudd, Thomsen, Edling and Randøy2010). In terms of management practices and environmental protection, Norway and Sweden tend to be characterized as ecologically and politically similar (Gulbrandsen Reference Gulbrandsen2005; Gulbrandsen Reference Gulbrandsen2008). On social issues, however, some studies have shown that Swedes tend to be more egalitarian. For instance, Sweden is more positive towards gender equality, while Norwegians tend to be more xenophobic (Knudsen Reference Knudsen1997; Jakobsson & Kotsadam Reference Jakobsson and Kotsadam2010). These differences may indicate important social and national identity distinctions; cultural traits, including fear of the unknown, can be important in understanding how societies deal with growing carnivore populations, and are also highly relevant for similar considerations outside Scandinavia.
In Norway, there is also a strong tradition of community-led governance, where local communities have considerable influence on national laws and decision-making (Otterlei & Sande Reference Otterlei and Sande2010). The protection of carnivores by law in the 1970s was perceived by some other rural inhabitants as being a national government imposed decision, where ‘external forces’ permitted the recolonization of Norway by large carnivores (Skogen & Krange Reference Skogen and Krange2003; Kleiven et al. Reference Kleiven, Bjerke and Kaltenborn2004). This opposition to central authorities may be absent or less important in Sweden, where it is more customary to determine policy at a national level (Otterlei & Sande Reference Otterlei and Sande2010). The Swedish tradition of central government may also enable more stringent conservation of nature than in Norway (Otterlei & Sande Reference Otterlei and Sande2010), exemplified by the much earlier Swedish protection of brown bears and lynx, as compared to Norway (Swenson et al. Reference Swenson, Wabakken, Sandegren, Bjarvall, Franzen and Soderberg1995; see www.artfakta.se). Hence, centralized carnivore management may be considered as a more provocative policy among Norwegians than Swedes, motivating rebellion against central authorities (Muth & Bowe Reference Muth and Bowe1998), and leading to Norwegians being more accepting of and willing to commit illegal hunting (Otterlei & Sande Reference Otterlei and Sande2010; Krange et al. Reference Krange, Tangeland, Sandström and Ericsson2012). For example, rural men with strong hunting traditions resist carnivore re-establishment, especially the wolf, as an act of cultural or social protest (Skogen Reference Skogen2001, Reference Skogen2003; Skogen & Thrane Reference Skogen and Thrane2008).
Kaltenborn et al. (Reference Kaltenborn, Bjerke and Strumse1998) used the NEP scale to look at how attitudes toward predators were reflected by environmental value orientation. They grouped their respondents into subgroups as farmers, wildlife managers and research biologists, and found significant differences between farmers’ NEP scores and those of the other two groups (Bjerke & Kaltenborn Reference Bjerke and Kaltenborn1999). Sheep farmers and hunters have traditionally been associated with anthropocentric environmental values (Kaltenborn et al. Reference Kaltenborn, Bjerke and Strumse1998); Vitterso et al. Reference Vitterso, Kaltenborn and Bjerke1998; Bjerke & Kaltenborn Reference Bjerke and Kaltenborn1999). We were unable to confirm this in our study, as rural traditions (such as sheep farming or big game hunting) were not related to the NEP scores. Even the experience of losing sheep or hunting dogs to large carnivores was not reflected in our respondents’ environmental value orientation. However, other studies have shown a higher variation in hunters’ environmental value orientation compared to that of sheep farmers (Bjerke et al. Reference Bjerke, Reitan and Kellert1998 b, Reference Bjerke, Vitterso and Kaltenborn2000; Vitterso et al. Reference Vitterso, Kaltenborn and Bjerke1998; Ericsson & Heberlein Reference Ericsson and Heberlein2003; Heberlein & Ericsson Reference Heberlein and Ericsson2008).
The lack of relationship between NEP scores and experiences, such as loss of dogs or livestock, has similarities to studies from South Africa, where livestock owners from two apparently similar communities experienced loss of cattle to predators, but showed highly divergent attitudes towards predators (Lagendijk & Gusset Reference Lagendijk and Gusset2008). The community with a tradition of hostile attitudes toward lions (Panthera leo) showed lower acceptance of predation and presence of predators compared to the community that did not have this tradition of hunting and killing lions, even though the consequences of having predators in their neighbourhood were the same (Gusset et al. Reference Gusset, Maddock, Gunther, Szykman, Slotow, Walters and Somers2008; Lagendijk & Gusset Reference Lagendijk and Gusset2008). Therefore, cultural differences played an important part in explaining these highly divergent attitudes towards lions (Lagendijk & Gusset Reference Lagendijk and Gusset2008).
Albeit attitudes toward carnivores in general have become more positive in the last century, studies from Finland and Sweden show how attitudes toward wolves turn in a negative direction when carnivores became established close to people (Bisi et al. Reference Bisi, Kurki, Svensberg and Liukkonen2007; Heberlein & Ericsson Reference Heberlein and Ericsson2008). The most positive attitudes toward the wolf may be found among people who have the least experience with them (Williams et al. Reference Williams, Ericsson and Heberlein2002). In contrast, more negative attitudes toward the wolf exist among urban people who had little contact with the countryside or had no experience with wolves (Heberlein & Ericsson Reference Heberlein and Ericsson2008). These examples illustrate the complexity and multi-scale nature of the human-carnivore conflict.
CONCLUSION
Environmental value orientation is generally thought to influence attitudes towards carnivores, and to some extent the behavioural intentions related to carnivores, for instance willingness to participate in hunting, commit illegal hunting or support for conservation (Fransson & Garling Reference Fransson and Garling1999; Kaiser et al. Reference Kaiser, Wolfing and Fuhrer1999; Hunter & Rinner Reference Hunter and Rinner2004; Milfont & Duckitt Reference Milfont and Duckitt2010). Differences in environmental value orientation between Norway and Sweden show that cultural differences may overshadow the presence of, or experience with carnivores, together with a generally greater acceptance of central authorities in Sweden compared to Norway. This difference in attitudes and responses to governance and management actions between Norway and Sweden explains the difficulties in the management of the common carnivore population across the Scandinavian Peninsula. Combined with spatial ecological descriptors of carnivore sustainability and economical descriptors, our spatial approach may form the basis for identifying optimal areas for carnivore establishment. Policy and management practices and concrete human-wildlife conflicts occur at different scales (Ericsson & Heberlein Reference Ericsson and Heberlein2003; Ericsson et al. Reference Ericsson, Bostedt and Kindberg2008), and multi-scale studies are needed to unravel the relationships between local, regional, national and international responses to carnivore conservation and expanding carnivore populations.
ACKNOWLEDGEMENTS
This study is funded by Hedmark University College. We are grateful to Peter Kiffney for the valuable comments on an earlier draft of the paper and for proofreading the English.
Open access
