GUILT: AN EFFECTIVE MOTIVATOR FOR PROENVIRONMENTAL BEHAVIOUR CHANGE?

by
Bedford, T., Collingwood, P., Darnton, A., Evans, D., Gatersleben, B., Abrahamse, W. and Jackson, T.

RESOLVE Working Paper 07-11

The Research Group on Lifestyles, Values and Environment (RESOLVE) is a novel and exciting collaboration located entirely within the University of Surrey, involving four internationally acclaimed departments: the Centre for Environmental Strategy, the Surrey Energy Economics Centre, the Environmental Psychology Research Group and the Department of Sociology. Sponsored by the UKs Economic and Social Research Council (ESRC) as part of the Research Councils Energy Programme, RESOLVE aims to unravel the complex links between lifestyles, values and the environment. In particular, the group will provide robust, evidence-based advice to policy-makers in the UK and elsewhere who are seeking to understand and to influence the behaviours and practices of energy consumers. The working papers in this series reflect the outputs, findings and recommendations emerging from a truly inter-disciplinary research programme arranged around six thematic research strands: Carbon Footprinting: developing the tools to find out which bits of peoples lifestyles and practices generate how much energy consumption (and carbon emissions). Psychology of Energy Behaviours: concentrating on the social psychological influences on energy-related behaviours, including the role of identity, and testing interventions aimed at change. Sociology of Lifestyles: focusing on the sociological aspects of lifestyles and the possibilities of lifestyle change, exploring the role of values and the creation and maintenance of meaning. Household change over time: working with individual households to understand how they respond to the demands of climate change and negotiate new, low-carbon lifestyles and practices. Lifestyle Scenarios: exploring the potential for reducing the energy consumption (and carbon emissions) associated with a variety of lifestyle scenarios over the next two to three decades. Energy/Carbon Governance: reviewing the implications of a low carbon society for governance, and investigating, in particular, the role of community in stimulating long-term lifestyle change.

For further information about our research programme or the RESOLVE Working Paper series please visit our web site http://www.surrey.ac.uk/resolve

GUILT: AN EFFECTIVE MOTIVATOR FOR PROENVIRONMENTAL BEHAVIOUR CHANGE?

by
Bedford, T., Collingwood, P., Darnton, A., Evans, D., Gatersleben, B., Abrahamse, W. and Jackson, T.

RESOLVE Working Paper 07-11

Research Group on Lifestyles, Values and the Environment Centre for Environmental Strategy (D3) University of Surrey Guildford, GU2 7XH, UK http://www.surrey.ac.uk/resolve/

Contact details: Tracey Bedford: email t.bedford@surrey.ac.uk Tel: 00 44 (0)1483 682181, Fax: 00 44 (0)1483 686671

Acknowledgements This research was funded by Department of Environment, Food and Rural Affairs. The support of the Economic and Social Research Council (ESRC) is gratefully acknowledged. This work is part of the interdisciplinary research programme of RESOLVE - the ESRC Research Group on Lifestyles, Values and the Environment.

ISSN 1755-7259

Abstract
Guilt is a form of negative affect aroused by a violation of personal or internalised social values or norms. It is closely related personal norms and operates by reducing self-esteem when an individual violates those norms. As policy makers look to find new ways to motivate and support pro-environmental behaviour, those working with communication and social marketing strategies have become interested in the role of guilt as a possible affect which could be utilised in campaigns. Two pieces of work were carried out examining the potential role of guilt as a motivator for pro-environmental behaviour change. The first was a rapid literature review of the psychological literatures pertaining to guilt and pro-environmental behaviour. The second was a small empirical study examining, amongst a range of other factors relating to pro-environmental behaviour, guilt in everyday decision making and responses to guilt messaging. The literature review covers a broad set of issues of interest to the research commissioning body, Defra. The empirical findings are more tightly focused on the differences between the forms of guilt identified within the sample and their implications, and types of reparation which limit proenvironmental behaviour. The findings from the empirical work suggest that it is useful to recognise the difference between trait, state and moral standards guilt when encouraging proenvironmental behaviour. We have suggested some key mechanisms that individuals use to alleviate guilt. We have argued that these are primarily comparative in nature, allowing the individual to feel that they have compensated for their impacts if they believe they have done their bit in comparison to others. However, throughout this paper, we argue that guilt is unlikely to achieve little unless effective alternative actions and changed structures and systems of provision enable the individual to act.

Key Words: Guilt, pro-environmental behaviour

1. Introduction
Guilt is a form of negative affect aroused by a violation of personal or internalised social values or norms. It is closely related personal norms and operates by reducing self-esteem when an individual violates those norms. As policy makers look to find new ways to motivate and support pro-environmental behaviour, those working with communication and social marketing strategies have become interested in the role of guilt as a possible affect which could be utilised in campaigns. Defra commissioned research into a range of potential motivators for pro-environmental behaviour including identity, social norms and agency. This paper focuses on the results in relation to guilt. Two pieces of work were carried out examining the potential role of guilt as a motivator for pro-environmental behaviour change. The first was a rapid literature review of the psychological literatures pertaining to guilt and pro-environmental behaviour. The second was a small empirical study examining, amongst a range of other factors relating to pro-environmental behaviour, guilt in everyday decision making and responses to guilt messaging. The qualitative research employed in depth interviews with thirty five participants and ten focus groups. All participants in the research were recruited on the basis that they expressed some proenvironmental values. This paper brings together both of these elements of work. The paper will begin with a general overview of the theorisation of guilt before focusing on particular questions about how guilt operates and can be alleviated. Additionally it will briefly review evidence around the successful use of guilt in behaviour change campaigns. The second part of the paper details findings about the differences in guilt responses shown by the respondents from the interviews and focus groups. It examines findings in relation to trait, state and moral standard guilt and suggests that existential guilt plays a role in making some people question the fairness of privileged consumer societies. Finally the paper documents processes of reparation and guilt alleviation, before discussing the potential use of guilt in proenvironmental behaviour.

2. Theorisations of guilt
Guilt is generally understood to be one of the emotions felt in response to the violation of personal norms (Christensen et al, 2004). The feeling of guilt can lead to a negative assessment of the self, and therefore is believed to motivate via a reduction in self-esteem (Burnett, 1994). Thogersen (2006; 2007a) suggests we are most likely to feel guilt and shame in response to subjective norms or a partially internalised social norm, rather than from personal norms determined by our own cognitive understanding of moral obligation; the latter, he suggests, motivate through inducing a positive self-concept. However, most theorists would suggest that guilt will be felt in response to both personal and internalised social norms. Additionally, Christensen et al (2004) suggest violations of injunctive social norms may create guilt.

However, violating a descriptive norm is of less consequence to the self, and therefore may be productive of nothing more than surprise1. Shame is classified as an avoidance emotion: that is one where the motivation produced by the emotion is to hide away. Shame tends to arise in a relationship between the individual and a group; and we will work hard to avoid feeling shame (Jackson, 2006) Guilt on the other hand is an approach emotion: one which motivates the individual to make reparation or find a way to deny responsibility (Burnett and Lunsford, 1995). Because of this, there is the possibility that guilt may be a useful emotion for motivating action. Whilst it is usual for academics to focus on state and trait guilt, Kugler and Jones (1992) separate guilt into the state, trait and moral standard guilt. Trait guilt endures over long periods of time. It can include the predisposition to feel guilty; State guilt is behaviour specific and felt in response to breaching norms; Moral standards guilt is related to more unifying values or codes of conduct, such as being frugal or concern about climate impacts. It is not behaviour specific, but encompasses a range of behaviours covered by the moral standard.

2.1 Guilt and identity Several theorists have suggested a link between guilt and identity. Schmader and Lickel (2006) suggest that the difference between shame and guilt is based in identity. They argue that people feel ashamed about who they are, but guilty about what they do. In practice it has been difficult to separate feelings of guilt from feelings of shame. For Schmader and Lickel, this difficulty can be explained by the close relationship between the two concepts. In short, bad people do bad things, therefore I feel bad about myself if I do something bad. This corresponds with studies of the implication of guilt feelings, which suggest that feeling guilty leads to a lowering of self-esteem (Burnett, 1994). Since self-esteem is one of the primary aspects of selfidentity according to Identity Process Theory (Breakwell, 1986), it can be seen that guilt feelings have a negative impact on self-concept, leading to self-regulation of behaviour. Whilst guilt is related to self-identity, shame is more closely related to social identity. Shame is a socially embedded set of emotions (including embarrassment and humiliation) and one of the consequences of not acting in accordance with social norms (Scheff, 2000). It is dependent upon social surveillance resulting from the negative assessment of an individual by a social group. From this it can be seen that

Two different types of social norms have been identified, each with different functions (Cialdini et al, 1991). The first is a descriptive norm, that is a belief about what most people do; the second is an injunctive norm, that is a belief about what others expect the individual to do.

we would anticipate both guilt and shame to be formative in the relationship between pro-environmental behaviour and identities. 2.2 Perceived social acceptability of a behaviour and levels of guilt Guilt is theorised as being felt as the negative affect from violating ones internal standards and norms. It is the self-punishment incurred by not following through personal norms. To the extent that personal norms can be understood as internalised social norms, then there will also be a relationship between social expectations of behaviour and guilt (Thogersen, 2006). However, for pro-environmental behaviour it is likely that there is a gap between whether people ought to practice a behaviour and its social acceptability, understood as the difference between injunctive norms and descriptive norms. Kaiser and Schimoda (1999) suggest that shame is caused by the violation of descriptive social norm, whereas guilt is caused by violating a personal injunctive norms. From this perspective, a non-environmental social norm with regards to a behaviour could clash with a pro-environmental personal norm. For example, it is not the done thing to refuse to eat a meal a friend has cooked you, even if the fish was not sustainably sourced and the vegetables were not local and in season. You may end up feeling guilty at eating the meal but at least avoid the shame of refusing someones hospitality. This ambiguity around the most appropriate behaviour may release the individual from a feeling of guilt. In short, all behaviour is context and situation dependent, and the more complex models of consumer action acknowledge that a range of different personal, social and structural factors determine the desirability of a behaviour and therefore intention to act. Moreover, the success of appeals to guilt or empathy is likely to depend on whether a behaviour is viewed as a social issue or a social cause (Bagozzi & Moore, 1994). Where a pro-environmental behaviour is a social cause, that is one that is accepted by society, then emotional appeals will be more fruitful. However, for environmental behaviours classified as social issues, that is one which society finds controversial (and it is likely flying and driving fall within this category due to a lack of alternatives), then emotional appeals are likely to have less success. Not all theorists see guilt as being based on internalised social norms (see the section below). Instead it can be conceived of as being based on empathy (Berndsen, 2007) or a felt moral obligation that has been cognitively constructed, rather than simply socialised (Schwartz, 1977, Stern, 2000). In this case, it would be possible to feel guilty about not undertaking a pro-environmental behaviour regardless of the social acceptability of the behaviour. It is clear that some individuals and social groups have strong personal and subjective/group norms in relation to a particular proenvironmental behaviour even in the absence of a social norm. This inevitably leaves the individual having to negotiate between moral obligation and different social expectations, or even between guilt for inaction or shame for action. Which norm is salient at the moment of decision would be likely to affect the level of guilt felt and therefore the intention to act.

2.3 Guilt: a motivator of pro-environmental behaviour? Guilt is a motivator of pro-environmental behaviour in that the expectation of negative affect is deemed to motivate individuals to act in line with their internalised subjective norms or personal norms, as outlined within the Theories of Reasoned Action and Planned Behaviour (Fishbein and Ajzen, 1980) and the Norm Activation Model (Schwartz, 1977) respectively. We would expect that guilt is, therefore, a motivator for those individuals who have accepted the need to undertake proenvironmental behaviours. However, it is less clear whether guilt could be a primary motivator for those who have not accepted responsibility for environmental impacts. For these people, it is more likely that the rewards and punishments related to behaviours are based in external surveillance and expectations. Hence, rather than guilt as a driver of action, for these people the motivations are extrinsic, such as shame or financial penalties. Since self-regulation is seen to be the most constant and efficient driver of behaviour, it can be assumed that those people who are externally motivated will be less likely to carry through behaviours in different spaces, conditions of ambiguity or in private (Deci and Ryan, 2000; Lindenberg and Steg, 2005). Hence, it is perceived that extrinsic motivators are at their most useful when they lead to the development of intrinsic motivations. Or in other words, if policy were to think about creating external reasons for environmental actions they should continue to facilitate the ability of the individual to internalise responsibility to eventually act through personal choice. For example, when applying tax based incentives to reduce car use, policy should also aim to continue to explain the impact of car use on the environment and how individuals can take responsibility for their impact by reducing their car use.

2.4 Guilt, personal impacts and social expectations As highlighted above, different relationships between guilt and norms have been theorised. For Fishbein and Ajzen (1980), an internalised subjective norm and beliefs about the outcome of a behaviour (i.e. whether it would reduce my impact on the environment) both moderate the likelihood of behavioural intention. In this model, I am influenced by both [internalised] social expectations and a recognition of my impact. In the Norm Activation Model (Schwartz, 1977) action is premised on an awareness of the consequences of the action and an acceptance of responsibility for any impacts, which then feed into a personal norm. In this model, my personal understanding of my impacts can be more important than social expectations. However, since personal norms and group norms tend to be closely linked, and my visible behaviour tends to take place in social contexts, actions will be affected by both social expectations and personal norms. Another debate about the relationship between guilt, impacts and social expectations is formulated around whether guilt is created by the acceptance of responsibility and inconsistencies in my norms of behaviour (for example Higgins self-discrepancy theory) or whether guilt created those feelings of responsibility (Baumeister et al, 1994, Hoffman, 2000). In this second formulation, guilt is interpersonal. An

understanding that ones actions harm others produces guilt directly. The feeling of guilt leads the individual to take responsibility for their actions, rather than the other way round (Berndsen and Manstead, 2007). Feeling empathy for the distress caused by ones actions exists outside of social norms and expectations. Thus, seeing the suffering of a polar bear due to climate change or turtles because of plastic bags creates guilt and leads to an individual taking responsibility for their actions. In either case, the greater the perception that the impact is under the control of the individual, the greater the sense of personal responsibility and the more guilt is felt (Berndsen and Manstead, 2007). Hence, we can assume that understanding personal impacts and having accessible alternatives leads to higher levels of action. Frequently environmental impacts exist at a more abstract level, which undermines the potential for empathy or a sense of our own impact. In this case smart norms need to be created. Smart norms link abstract concepts such as contributing to climate change is bad to a moral injunction to reduce your impact on climate change through a specific behaviour by driving less. Lindenberg and Steg (2007) suggest that environmental smart norms can be strengthened by moralisation, that is by attaching social stigma to the failure to comply with the norm. Under these circumstances, social expectations would strengthen personal responsibility.

2.5 Guilt alleviation strategies Guilt and shame are conceived as two different types of motivations. Guilt is believed to be an approach motivation (that is one for which the individual must do something to alleviate the feeling), whereas shame is an avoidance motivation (one which provokes the individual to hide if they have done something that provokes it). The alleviation of shame can be as simple as avoiding carrying out an action in public. However, the need to relieve the negative affect of norm violation is assumed to prompt an individual into one of four potential courses of action (Burnett and Lunsford, 1994): Doing good deeds Undoing harm to the injured party Self-criticism Self-punishment (through guilt). It is anticipated that three different types of behaviour can be produced to reduce the affects of guilt or cognitive dissonance: compensation, expiation or denial of responsibility. In short, if employing guilt based behaviour change strategies, it would be essential to know under what conditions people would be prompted to undertake the behaviour, carry out a different compensatory behaviour or deny responsibility for the behaviour. The concept of needing to undertake the behaviour to reduce guilt is selfexplanatory. However, we shall look more closely at the other strategies employed to alleviate guilt.

The first strategy is to carry out a different behaviour. This behaviour can have less negative consequences for the self or be more simple to accomplish, such as carbon offsetting or doing more of something one finds rewarding. In our own work, we have found that people internally trade their perceived carbon impacts allowing them to select some behaviours and not others (Collingwood, 2007) or have a clear sense of an understanding that one should do their bit, but can only do so much (Bedford, 1999). This suggests that people are able to offset or trade away their obligations and therefore their guilt about not undertaking an action. This has close links to the Door-in-the-Face concept within persuasion theory (Cialdini and Goldstein, 2004). The DITF technique for selling hypothesises that if an individual is requested to undertake a behaviour then they feel guilty if they say no. If the same person then requests the individual to undertake a smaller behaviour they will feel obligated to agree, even if the second behaviour is still more than they would have agreed to if it had been the original request. From this we can surmise that the willingness with which the public is currently complying with media campaigns to reduce plastic bag use could be partly based on a desire to offset their guilt for larger actions without structural or normalised solutions, such as flying or driving. The final technique for guilt alleviation takes place under conditions of ambiguity, where it is unclear as to whether the behaviour is in line with the social norm or efficacious; and in the face of decisions which would have little personal gain or high personal cost (Biel and Thogersen, 2006). This is the defence mechanism of selfserving denial. Lindenberg and Steg (2007) list a series of different mechanisms that the individual can utilise to shift away from moral obligations and, therefore, neutralise even an activated personal norm (Biel and Thogersen, 2006): 1) Actors can deny the seriousness of the environmental problem by ignoring, minimising or distorting the impacts; 2) Actors can project their contribution as being meaningless compared to some identified larger actor; 3) They can deny their personal ability to deal with the problem, undermining their perceived behavioural control; 4) Or they can deny their responsibility for solving the problem.

2.6 Can guilt based marketing campaigns change behaviour? The potential of guilt as a tool for promoting pro-environmental behaviour using communication tools must be considered in the light of several studies into the impact of guilt appeals in advertising and social marketing. These studies suggest that overtly manipulative guilt appeals create resistance in the audience, rather than guilt (Cote et al, 2005). In particular, the use of existential guilt in social marketing campaigns (that is guilt caused by the discrepancy between personal circumstances and the well-being of others) has been used so frequently that individuals are wise to the manipulation technique, which has led to diminishing returns in terms of motivation content.

10

This is not to say that guilt appeals cannot promote pro-environmental behaviour. Indeed, even where the use of guilt creates negative reactions towards the advert, it can produce positive impacts on behaviour (Cote et al, 2005). However, it is suggested that several factors could improve the effectiveness of the appeal: New and interesting tactics to appeal to guilt; Resonance with consumers own experiences; The inclusion of actions by which the consumer can reduce their feelings of guilt (Cote et al, 2005);

And the use of empathy for non-contentious social causes (Bagozzi & Moore, 1994).

3. The empirical study: methodology

In depth qualitative interviews were conducted with thirty five interviewees spread across seven different locations. The locations were selected to ensure we had a spread of interviewees from across the country, representing urban, suburban and rural lifestyles. The interviews were approximately 1.5 hours long covering the interviewees lifestyle, roles, routines and aspirations; and their reaction to a range of pro-environmental behaviours which could be more or less easily enacted within their everyday lives. Ten focus groups were held, divided between London and Leeds. The focus groups were two hours in length and had been designed around a series of tasks to stimulate discussion about the phenomena being investigated. Rather than a more flexible exploratory discussion, the groups were used to test our findings from the quantitative analysis and interviews, and extend our understanding of the motivations in line with theory where necessary. Unlike the interviews, the focus groups were designed around discussions about environmental attitudes and pro-environmental behaviours. Environmental concepts, such as climate change and carbon footprints were introduced to the participants during the session. Rather than studying the seven headline proenvironmental behaviours, the groups were given the opportunity to introduce their own understanding of environmental actions. Guilt based messages were introduced to the groups to test their initial reactions to communication strategies. Many of the findings arising from the empirical work drew attention to the roles of state, trait and moral guilt. These findings were incidental and were not designed into the methodology. Hence we would like to clarify that the empirical findings within this paper are partial, and require a specialised study of pro-environmental guilt to substantiate and further develop the conclusions before they should be used for policy purposes.

11

4. The empirical study: findings

The empirical findings for guilt revealed some evidence that guilt played a role in the continuation of pro-environmental behaviour, with less evidence of it being a primary motivator for pro-environmental behaviour. However, there was large variety in the level and type of guilt exhibited by individuals in the sample. The findings included here focus on the role of trait, state and moral guilt; questions of personal or social bases for guilt responses; and methods of making reparation for guilt feelings. 4.1 Trait, state and moral guilt As we have outlined above, the findings from the research have suggested a role for state, trait and moral standards guilt. For the purposes of this research we shall also include existential guilt within trait guilt. Here we shall use existential guilt to denote a sense of guilt about affluent or privileged lifestyles. These have different implications for policy and how guilt can be utilised to change behaviour. We shall examine these in some depth below. 4.1.1 Trait guilt It became clear throughout the interviews that some individuals were more predisposed to general guilt about their behaviour than others. Once made aware of an issue they develop a guilty conscience and feel the need to make reparation. The clearest account of the role of trait guilt predisposition is demonstrated by Paula.

Paula is very guilt driven and feels guilt across a range of different concerns and behaviours.
I do [feel guilty] about wasting electricity because I think its just a thing that takes time. I do about the car, but then I cant see any way around it in my position. The locally grown food, yes I do feel guilty when I don't buy it but then sometimes its like I havent got enough money to be able to do

She recognises that she is perhaps too guilt driven and relates it to her family and religious upbringing.
I was brought up in the church. If anything it has made me a little bit too Certainly when I was growing up, you know if I did something that was a little bit naughty I was thinking oooh, which I dont think was very good. I think it had a bad impact on me, you know when you have your first boyfriend and all that sort of thing. I think it just, in many ways it wasnt good because it made me feel guilty, everything I did it sort of felt a bit, filled me with guilt in terms of, you know, is this classed as not being thoughtful enough or is it classed as a sin.

Because of this trait guilt, Paula tends to have to act on things which prick her conscience. When she becomes aware of an issue then she will add it to her other actions. She says the media has brought environmental issues to her awareness, aided by her husbands interest in them.

12

I mean my husband has made me a little bit more, you know sort of conscious about things as well. Because I think a lot of it is, I mean he got me into the programmes and once I watched the programmes then I had a conscience about it really. Before I suppose I never really thought about it. Then once I do start thinking about it then I start feeling bad about it.
However, Paula was not alone amongst the participants in being a guilt driven person. Indeed another example is a focus group participant who said that she had to act if she felt bad about anything because otherwise the guilt would condemn me. It is less easy to be confident about recognising those with a low predisposition to feelings of guilt in this type of research. However, we can be certain that two of our interviewees claimed to be concerned about the environment and tried to adopt proenvironmental behaviours, yet appeared to be unmotivated by guilt: James: I am quite conscious, I make an effort. Q: So I cant make you feel guilty about anything? James: No, I have no conscience on that. The research was not designed to look at trait guilt, either from personal predisposition or existential anxiety. Indeed, in the literature review we dismissed the exploration of existential guilt because studies have shown that its overuse in advertising had reduced its efficacy (Cotte et al, 2003). Guilt is normally defined as the negative affect aroused by the violation of personal or internalised social values and norms, and viewed as being dependent on the extent to which individuals feel responsible for and in control of their impacts. Existential guilt is a more constant sense of guilt about having a perceived, unfair advantage over others or a privileged lifestyle. It is closely related to empathy through a knowledge that others are in a less fortunate position than ones self. However, for some respondents these issues appeared to be so central to their motivations for pro-environmental behaviour that they inevitably became part of the discussion. Thus we have included them as a key part of our findings. Here we give two examples, Melanie and Amy. Melanies explanations of both her pro-environmental behaviours and those she could not negotiate were caught up in notions of a spoilt and privileged society, where our expectations and wealth had led to us being able to consume in a way which previous generations could never have imagined. This informed her distaste of wasteful behaviours and her personal norms around second hand clothes. Amy similarly talks about notions of being rich and spoilt. Indeed, her feared identity was becoming someone who was too spoilt to have enough thought for others. Whilst she was not the most proactive member of the sample, she was doing her bit partly because she recognised that she was in a relatively privileged position.

13

Having worked in a poorer country she said she had been unable to avoid seeing poverty around her all the time and this had affected her views in life: Amy: I dont know, you just carry on with your life and you just behave like you know is the norm, but when you are faced with [seeing how poor some people are compared to you] and Ive been to stay with friends in Africa, so Ive seen that as well, and it just brings it home to you. So it does make you think a bit more, it does open up your mind a bit, and you know you are doing something you Yes it does change you I think, yes you have more thought for others I guess and you dont take what youve got for granted any more. In addition to the above examples, several of the respondents mentioned how travel in poorer regions had made them think more about their privileged lifestyle: F: Im very, very careful about food wastage because Ive travelled a bit and Ive seen poverty around the world, so Im very conscious about not throwing away food. I freeze it. (London focus group) Unlike other forms of guilt where there is a relationship between your actions and the consequences of these actions for other people or the environment, existential guilt can be seen to motivate simply because you are aware of the disparity between your lifestyle and the lifestyle of others (Schmitt et al, 2000)2. This seems to be particularly powerful for wasteful behaviours. Indeed, existential and equity concerns about wasting food transcended income levels for those who were motivated by it, with both our interviewees with the highest and the lowest incomes mentioning it. For example, a young single mother was particularly motivated through an internalisation of her mothers existential guilt: Q: Do you ever feel guilty about any of those sorts of things? Jenny: I hate throwing away food. [] My mum always hated that dont waste food, there is children starving, when we was younger. Possibly the likelihood of being motivated into action by existential guilt may depend upon to whom you compare yourself. For example, one interviewee feared being too rich and spoilt. Another feared being the type of wealthy person who only cared about their looks. It may be that they are positioning themselves against materialism in line with their perceived self-identity. However, it could be that the level of importance of relative wealth and material possessions to a person affects existential guilt. Certainly Montadas (1989) research points to the possibility that how important the gains from inequality are to a person will affect how likely they are to feel inclined to make reparation for that inequality. Existential guilt then operates as a background recognition of injustice. It could be part of the motivation for many different behaviours which an individual feels are
Although a clear causal relationship between ones own privilege and others disadvantage is most likely to motivate reparative action.
2

14

within their control. It is viewed as being a sense of privilege or feeling of inequality from being overpaid, and is therefore the opposite of the sense of deprivation and relative underpayment (Schmitt et al, 2000). Hoffman (1990) named it the guilt of affluence. In this it can be seen to correspond closely with Sopers work on citizenconsumers. Soper (2005) has argued that people, particularly more affluent individuals, are critical of the levels of consumption we have in consumer societies and enjoy an alternative sense of reward through acting in ethical ways. A recognition of inequality is not enough to determine behaviour. Instead Schmitt et al (2000) and Montada (1989) argue that several factors may determine whether someone is likely to intend to act or actually act on existential guilt including: Moral outrage about the injustice of discrepancy; A causal link between ones own privilege and others deprivations; Levels of perceived control or sense of hopelessness; Contentment with own privileges and fear about losing those privileges.

Their research was about gender inequalities rather than pro-environmental behaviour. Because of this it is impossible to be certain that the same factors would apply. However, a specialised study of existential guilt, contentment with and attachment to the gains of privilege, and pro-environmental behaviour could prove to be enlightening. Certainly, our study (which did not aim to cover such findings) would suggest a link between affluence, existential concern and a need to undertake small pro-environmental or ethical behaviours to give something back. Whilst the finding can only be indicative, this appeared to be particularly relevant for the more proactive interviewees. For example, one successful and fairly affluent interviewee appeared to have become post materialistic in his values, suggesting he no longer needed to work to support his own needs. He introduced the subject of existential guilt himself, arguing that it concerned him at an international and community level. Furthermore, he was concerned about leaving the environment in good shape for future generations: Michael: I hate to see, you know, sorts of articles and documentaries where we are ruining a part of the world just for our Western, just to service our western culture, our western lifestyle you know what I mean. [] living in a more rural location you get concerned about, you know, anything that might affect those people that are dependent on a life from the land, if you like. Which is not something that I have to think about, but I would be concerned if I didnt, if there was something I knew that I was doing or everyone else was doing that needed to stop because it was affecting those people who have to live their lives like that, and I would do that. Whilst the above quote illustrates the impact of existential guilt on behavioural intention, it is worth noting that this is not to argue that it would inevitably create behaviour change. The notion of balance limits how much many of the less proactive individuals felt obliged to do and structural lock-in makes actions impossible or undesirable:

15

Michael: I dont know why we havent [bought second hand]. I think we just kind of have always felt that we do enough already. We will return to discuss the limits to action later in this paper. However, here we wish to conclude this section on existential guilt by suggesting more research would need to be undertaken to understand the relationship between existential guilt, affluence, values and pro-environmental behaviour before policy makers considered using it. From our limited study it appears likely that it provides part of the motivation for behavioural uptake. However, how much behaviours actually change remains to be seen. For example, is it limited (as it appears to be) by the individuals sense of giving back a little bit and how they perceive their role in proenvironmental behaviour? Or, given that existential guilt is about a whole system of global consumption and economic inequalities, would the individuals inability to remove guilt about the outcome of that system eventually lead to demands for systemic change or to such a sense of hopelessness that the individual would disengage altogether? 4.1.2 State guilt State guilt is specific to a behaviour based on breaching a personal norm or internalised social norm for that behaviour. In both the interviews and focus groups we asked respondents if there were any behaviours they felt guilty about not undertaking. Across the sample there were large differences in the number of behaviours people felt guilty about. The most proactive, and especially those with strong pro-environmental values, listed a wide range of behaviours including flying, car use, water use and home energy use behaviours. Those with a waste focus tended to feel guilty about wasting energy, food and water or buying unnecessary products. By contrast, the least proactive respondents had few feelings of guilt about any behaviours other than recycling: M: Im afraid none of them would make me feel guilty. Q: No? M: I wouldnt lose sleep over them. (London focus group) Recycling was the behaviour most respondents felt guilty about not undertaking, being established as both a personal and social norm for individuals across the sample: Jason: I do feel, the major one, the recycling one I really do feel guilty on because I have always recycled, and my parents have always recycled for as long as I can remember really. So for me now to stop something that Ive been doing for years does make me feel a bit guilty because I know how much goes in that bin. Q: So how do you deal with feeling guilty about it? Jason: We just try and recycle as much as we can really. I mean, as I say the problem is those bins downstairs, theyre full after a couple of days, so you generally find well recycle for the first couple of days after bin collection, and then after that most of the stuff will go in the bin.

16

Q: Do you feel guilty to some extent? M: I do with recycling. I mean obviously my green bins outside and my normal bins inside. Its easier to put it in the normal rather than walk outside you feel a little bit guilty I suppose. (Leeds focus group) The quote above highlights the tenuous relationship between guilt and behaviour, as the feeling of guilt is not enough to overcome even minor structural inconveniences. Home energy use, particularly leaving appliances on standby, and unnecessary or single occupancy car use were also regularly listed as behaviours people felt guilty about. For some of the less active participants, their personal norms were still in the process of being created in response to campaigns, raised awareness or social pressure. The immediacy and visibility of the specific problem appeared to help support state guilt in prompting feelings that one should change behaviour to address it: Q: What sorts of things might make you feel guilty? F: Well, one thing Ive been really, really conscious of lately is that I always seem to be chucking out packaging in the rubbish. I dont know why I am more aware these days, but I am [] My DVD player goes on standby and I look at that orange light virtually every day and I just feel really guilty as Im looking and I think Oh my god, I must do something about that. (London focus group) Having a large engined car was a source of guilt for the female respondents more than the men in the sample. Rather surprisingly, locally grown seasonal food whilst being a desirable behaviour was not often mentioned as guilt provoking. However, a few individuals felt guilty about not supporting local or small businesses or the impact on their carbon footprint from buying food that had been brought to their local stores over long distances. 4.1.3 Moral standards guilt Analysis of the narratives around pro-environmental behaviours and guilt suggest that some individuals have unifying moral values which motivate behaviours. For example, a few of the most pro-environmental individuals were concerned about their overall impact on climate change through their carbon footprint. Those with strong anti-waste values can be concerned about all wasteful behaviours, either through a dislike of waste or a dislike of wasting money. For these individuals, it appears that guilt is motivated by a failure to adhere to a personal moral standard rather than a behaviour-specific personal norm. Moral standard guilt has implications for how people will respond to different behaviour change initiatives. For example, those with high levels of concern and personal responsibility in relation to climate change tended to feel guilty about flying. However, those without a strong moral standard in relation to climate change were more likely to undertake smaller actions, possibly as internalised social norms, but to have less guilt about flying. It may be that both sets of people continue to fly; however, those with moral standard guilt will need to make reparation through what Collingwood (2007) calls internal carbon trading:

17

Sally: I think we do enough. The thing thats not always good is the fact that we do fly and I do feel bad about that. I dont know if its enough to make me stop, but if you can do your bit in other ways. I do make an effort, put it that way, maybe I make more of an effort because I feel bad about flying. M: I want to go to Australia and I want to do that; I dont need to - I can enjoy where I am. Im making sure my carbon footprint [is kept low]. For the next three years I will do something environmentally friendly so that I can feel justified in going off. (Leeds focus group) Similarly, those with waste values talked in the focus groups about how their central concerns were about waste, rather than climate change. We presented focus groups with two different guilt messages, one an empathetic image about the impacts of climate change, the other a waste based message (with a message about avoiding both wasting money and resources more generally). Whilst responses tended to be mixed in other groups, the waste focused were clear that they preferred the waste message as this talked to their own moral standards. For those who appeared to have thrift or financial frugality as a moral standard, a money saving message was established as being in line with their moral values and standards: M: I dont feel guilty about waste, I feel guilty about the cost of waste. So if you cook too much sort of thing and then youre throwing it away, I feel guilty about the waste of money as opposed to the food. (London focus group) What this suggests is that the format of the awareness raising and guilt messages would be most usefully targeted in the most appropriate format for the target audience. Thus offsetting messages should be targeted at those with moral standards around climate change or carbon footprints. However, for those with strong waste values, carbon messages (and perhaps even carbon footprints) might have more impact if they are couched in terms of how much is being wasted rather than just used. Indeed, further research which established whether using carbon footprints enables the development of moral standards across all groups could help inform the best way forward with guilt based initiatives. It may be that it pushes individuals to think about their entire impact rather than doing a few small actions; or it could fail to resonate with those who have different concerns. As it stands, the focus on climate change can alienate some individuals who oppose the concept: M: The problem is that the environment automatically adds up to global warming. If it was something else I might bother, but it doesnt do it for me. (Leeds focus group)

4. 2 Guilt: a personal or social emotion? The original specification for this project asked for an understanding of whether guilt was based on personal impacts or social expectations. From our findings we would argue that both can be sources of guilt. We would tentatively suggest that the most proactive individuals tended to be concerned about their personal impacts on climate

18

change or levels of waste, but most individuals in the total sample felt guilty about impacts in relation to particular behaviours. Similarly, most members of the sample were affected by social expectations. However, some of the younger interviewees appeared to principally feel guilty in relation to social expectations. Whilst some of these young interviewees did acknowledge feeling guilty about particular behaviours this was often in relation to pressure from friends or family, occasionally appearing to be a transient emotion rather than substantial enough to motivate behaviour. Indeed, one individual suggested that he only tended to feel guilt in response to discussions about the issues: Brad: It is only when I am talking about it that I think about it. The car, yes, sometimes I do feel guilty about using it, I mean I do use it for stupid journeys and things, but it is not in the forefront of my thoughts. This suggests an absence of personal norms and that social expectations have not yet been internalised. Indeed, some of the young participants appeared to be lacking a proactive social group who would provide the social expectations of behaviour, although they suggested that if the process of social comparison highlighted a failure to follow the descriptive norm they would feel guilty: F: I think [seeing other people doing things] would work with me because Im a little bit influenced by other people around me and I think if I saw other people do it, and Im not doing it, then I do tend to feel a bit guilty. And then I will kind of copy them. (London focus group)

4.3 Making reparations Guilt is a distressing emotion (Miceli and Castelfranchi, 1998). Classified as an approach emotion, an individual who is feeling guilt has to make reparation to remove guilt and restore self esteem. For much of the time this reparation is to change behaviour to avoid the negative impacts and feelings. However, from the interviews it appears that feeling guilty about a behaviour does not mean it will be acted upon, even for individuals with trait guilt, such as the example below: Q: Do you ever feel guilty about these sorts of things or are you quite happy with what you are doing? Melanie: No. I threw a couple of plastic containers away in the bin and I was thinking this is so not good. And the other thing is if things were packaged better in cardboard and things. [] but it doesnt stop me buying it, thats the problem, maybe I should make more of a conscious effort not to buy it which is not what I am doing. I am picking up the apples or whatever and putting them in a plastic bag. Habitual behaviours, behaviours which are too difficult or demanding and which have not been facilitated by accessible alternatives and normalised structures and

19

systems of provision, and behaviours which have too much meaning for the individual or important others, may also limit the likelihood of action: Gill: Yes, again, you do feel guilty that you should be doing that little bit more, and you get a burst of trying to be good, and then you slip back into your old habits. A range of devices allows the individual other ways to reduce their guilt or to restore self esteem which removes the need to act on a specific behaviour. These include: 1. Compensation, for example by offsetting impacts through undertaking different (and often easier) behaviours; 2. Denial of responsibility. We will discuss these briefly below. Before that we wish to make the point that from our sample, social comparison appears to be central to the ability to negate guilt. Whether the interviewee was talking about compensating with other behaviours or denying they were the real problem, comparing ones actions favourably with other peoples was often key to ameliorating guilt. The very notion of doing my bit allowed many in the sample to see themselves as doing as much as, if not more than, other people. If everyone is doing their bit for the environment, then there is no need to do more than a few simple actions until others do the same. And whilst those with strong moral standards undoubtedly felt the need to do more than those with personal or social norms for individual behaviours, they could still repair self-esteem by recognising that they were doing more than others. In part this recognition of doing more than others reflects the failure to introduce and normalise the cultures, structures and systems which would allow individuals to participate in pro-environmental behaviours without extreme inconvenience to the self. For many behaviours, undertaking responsibility would be to step outside of normal standards and expectations. Hence a recognition that much proenvironmental behaviour is above and beyond the actions of society in general.

4.3.1 Compensation Compensation can include offsetting guilt through other behaviours, as we have seen in the case of the Positive Greens and their attitude to flying. As we have suggested, possibly because of the conceptualisation of pro-environmental behaviour as doing my bit, people were able to do a limited amount to offset their conscience, including doing less of the problem behaviour. Victoria: I feel guilty if I have a bath. Q: Do you? Why? Victoria: Because again its the water and the heating thats you know, energy efficient and its much better to have a shower. You know, but having a bath is sort of my luxury now. Q: You still have a bath even though it makes you feel guilty? Victoria: I do, but less often and I make it, you know, very smelly and lovely and more of a treat.

20

Debbie: No, I dont feel guilty [about the sports car] to be honest because I only use the car when I have to and I car share with it.

4.3.2 Denial of responsibility The second major mechanism we identified for repairing guilt was to deny responsibility for the impact. Following Lindenberg and Stegs (2007) formulation of self-serving denial, we analysed the transcripts for the following phenomenon: Denying the seriousness of the problem; Suggesting personal impact is insignificant compared to others; Denying the ability to control the problem; Denying responsibility for the problem. Whilst some of the sample undoubtedly denied they were the problem, it is less clear that it was a process of denial so much as a failure to accept personal responsibility for any environmental impacts in the first place. However, amongst those who had accepted some level of responsibility, the main process of denial appeared to be located around suggesting ones personal impact was insignificant compared to others. Kevin: I couldnt be called someone who kind of wilfully abused the resources of the earth. You know I dont think I am using up or wasting stuffSo I would say I think if everybody behaved like me, I suspect, immodestly, that the world would be in a better shape than it is. Helen: I am not going to feel guilty for flying a lot because some countries and people do nothing and dont care at all and are ignorant and dont even think about it. There was some suggestion that particular groups of individuals deny behavioural control, thereby negating the need for guilt. Those interviewees, who were doing their bit without strong personal norms, talked about how they tried to balance environmental needs with the needs of friends and family. So whilst they may feel guilty about their lack of pro-environmental behaviour, they would feel guiltier about having a negative impact on their family. Because of this, people described how they were not fully in control of their own pro-environmental behaviours: Jeff: Guilt that I could be doing more, yes. But some of the decisions I dont feel are mine alone to make in as much as Ive got a partner and having a happy marriage is more important to me than trying to push my ideas on cars. Additionally, where people had tried to undertake more environmentally friendly behaviours and found that they had adversely affected their family, they were unwilling to repeat the behaviour. This reaction appeared to remove the need to feel guilty about it. Holidays abroad, for example, fell into this category.

21

Stuart: My two kids have been away every year since they have been born. They take it as normal that they are going away abroad, and to be fair to them we swapped our holidays around last year to take a holiday in the UK and we were so bored. Never again will I do that, it will either be we have a week or two weeks away.

5. Discussion
The empirical research has suggested that it may be useful to recognise the difference between trait, state and moral standards guilt when encouraging pro-environmental behaviour. From this conceptualisation it appears likely that those with trait guilt would be the most susceptible to guilt as a primary motivator for behaviour change. We have highlighted existential guilt as a background feeling of over-privilege for some of the sample. However, we acknowledge that this requires further research in order to recognise its impacts on pro-environmental behaviours; indeed, we suggest any attempts to utilise existential guilt could result in negative attitudes to consumer society which leave the individual impotent in the face of the enormity of the task to change the perceived levels of injustice. Moreover, as Cote et al (2005) suggest, existential guilt has already been over-used in advertising such that people are desensitised to these appeals. Nonetheless, we have reintroduced this concept to our work in response to unprompted discussions about this sort of guilt in the interviews. For those who have a moral standard, perhaps around carbon reduction or waste reduction more generally, guilt could be more related to the moral than any particular actions. In this, guilt may be part of the mechanism which encourages and enforces action across a range of behaviours. We have argued in this paper that messaging which hoped to target behaviour change through guilt would need to be specific to the moral standard held. However, we would also suggest that for most people guilt is limited to state guilt related to specific, probably quite small proenvironmental actions for which the individual has developed a personal norm or internalised a social norm. In this, the reach of guilt to promote behaviour change, rather than enforce some limited actions, remains questionable. Even where the individual feels guilty about their environmental impacts, we have suggested some key mechanisms that individuals use to alleviate this guilt. These appear to be primarily comparative in nature, allowing the individual to feel that they have compensated for their impacts if they believe they have done their bit in comparison to others. Moreover, these small actions can allow the individual to feel they have compensated in some part for their impacts and thus can alleviate their guilt. This essentially places the individual within the norms of society by ensuring those smaller or more accessible behaviours are undertaken, but those which are not facilitated by the systems and structures of society are less likely to be motivated by guilt for the majority of people. Indeed, given that guilt is a negative affect and uncomfortable for the individual, it may be as well that it can be alleviated until such time as alternative actions are facilitated and embedded within the norms of society. As suggested by Cote et al (2005), those wishing to promote behaviour change

22

through guilt should ensure that individuals are aware of and able to engage in more pro-environmental behaviours.

References
Ajzen, I. & Fishbein, M. (1980) Understanding attitudes and predicting behaviour Open University Press: Milton Keynes Ajzen, I. (2002). Perceived behavioural control, self-efficacy, locus of control, and the theory of planned behaviour. Journal of Applied Social Psychology, 32, 4, 665-683. Bagozzi, R. and Moore, D (1994) Public service advertisements: emotions and empathy guide pro-social behaviour Journal of Marketing 58, 1, 56-70 Bandura, A. (1989). Human agency in Social Cognitive Theory. American Psychologist, 44, 1175-1184. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215. Bedford, T. (1999) Ethical consumerism: everyday negotiations in the construction of an ethical self Unpublished PhD: UCL Berndsen, M. & Manstead, A (2007) On the relationship between responsibility and guilt: antecedent appraisal or elaborated appraisal? European Journal of Social Psychology 37, 774-792 Bicchieri, C. (2006) Covenants without swords: group identity, norms and communication in social dilemmas Rationality and Society 14, 192-228 Biel, A, & Thorgesen, J, (2006) Activation of social norms in social dilemmas: a review of the evidence and reflections on the implications for environmental behaviour Journal of Economic Psychology 28, 93-112 Breakwell, G. M. (1986). Coping with Threatened Identities. London: Metheun. Burke, P. J., & Reitzes, D. C. (1991). An identity approach to commitment. Social Psychology Quarterly, 54, 239-251. Burnett, M & Lunsford, D. (1994) Conceptualising guilt in the consumer decision-making process Journal of Consumer Marketing, 11, 3, 33-43 Christensen, N., Rothberger, H., Wood, W., & Matz, D. (2004) Social norms and identity relevance: a motivational approach to normative behaviour Personality and Social Psychology Bulletin, 30, 10, 1295-1309 Cialdini, R., Reno, R. & Kallgren, C. (1991) A focus theory of normative conduct: a theoretical refinement and evaluation Advances in Experimental Social Psychology 24, 201-234 Cialdini, R. & Goldstein, N. (2004) Social influence: compliance and conformity Annual Review of Psychology 55, 591-621 Cote, S., Coulter, R., & Moore, M. (2005) Enhancing or disrupting guilt: the role of ad credibility and perceived manipulative content Journal of Business Research 361-368 Deci, E & Ryan, R (2000) The what and why of goal pursuits: human needs and the selfdetermination of behaviour Psychological Enquiry 11, 4, 227-268 Jackson, T. (2005a). Motivating Sustainable Behaviour, A review of evidence on consumer behaviour and behaviour change, SDRN/Policy Studies Institute, London. Jackson, T. (2006) Consuming Paradise? In Jackson, T. (ed) Sustainable Consumption Earthscan Reader: London Kaiser, F. & Schimoda, T. (1999) Responsibility as a predictor of ecological behaviour Journal of Environmental Psychology 19, 243-253 Kugler, K. & Jones, W. (1992). On conceptualising guilt. Journal of Personality and Social Psychology, 62, 2, 318-327

23

Lindenberg, S & Steg, L. (2007) Normative, gain and hedonic goal frames guiding environmental behaviour. Journal of Social Issues 63, 1, 117-137 Schmitt, M., Behner, R., Montada, L., Muller, L. & Muller- Fohrbrodt, G. (2000). Gender, ethnicity and education as privileges: exploring the generalizability of the existential guilt reaction. Social Justice Research 13, 4, 313-337 Scheff, T (2000) Shame and the social bond: a sociological theory Sociological Theory 18, 1, 84-99 Schnader, T. & Lickel, B (2006) The approach and avoidance function of guilt and shame emotions: comparing reactions to self-caused and other-caused wrong doing Motivation and Emotion 30, 1, 43-58 Schultz, W. (1999) Changing behaviour with normative feedback interventions: a field experiment of curbside recycling Basic and Applied Social Psychology 21, 25-36 Schultz, W., Nolan, J., Cialdini, R., Goldstein, N & Griskevicius, V,. (2007) The constructive, deconstructive and reconstructive power of social norms Psychological Science 18, 5, 429-434 Schwartz, S. (1977) Normative influences on altruism in Berkowitz, L (ed) Advances in Experimental Social Psychology Academic Press: New York Schmitt, M., Behner, R., Montada, L., Muller, L. & Muller- Fohrbrodt, G. (2000). Gender, ethnicity and education as privileges: exploring the generalizability of the existential guilt reaction. Social Justice Research 13, 4, 313-337 Stern, P. (2000) Towards a coherent theory of environmentally significant behaviour Journal of Social Issues 56, 3, 407-424 Soper, K. (2007). Re-thinking the Good Life: the citizenship dimension of consumer disaffection with consumerism. Journal of Consumer Culture 7, 2 205-229 Thogerson, J. (2004). A cognitive dissonance interpretation of consistencies and inconsistencies in environmentally responsible behaviour. Journal of Environmental Psychology, 24, 93-103. Thogersen, J (2006) Norms for environmentally responsible behaviour: an extended taxonomy Journal of Environmental Psychology, 26, 4, 247-261 Thogersen, J (2007) The motivational roots of norms for environmentally responsible behaviour Proceedings of the Nordic Consumer Policy Reseach Conference 1-16 Tory Higgins, E. (1987). Self-discrepency: A theory relating self and affect. Psychological Review, 94, 319-340.

24