Introduction

1According to a recent report (IPCC, 2018), published by the Intergovernmental Panel on Climate Change (IPCC), the energy transition required to limit climate change demands policy-driven pathways that encompass accelerated change away from fossil fuels, large-scale deployment of low-carbon energy supplies and improved energy efficiency, but also sustainable consumption lifestyles (Rogelj et al., 2018: 149). Previous studies have reported that although people often accept responsibility for their carbon footprint, their voluntary efforts are not sufficient to reach the 1.5 degrees Celsius goal (e.g. Dubois et al., 2019). This gap between attitudes and behaviour (Kollmuss & Agyeman, 2002) indicates a need to develop an understanding of how the population justify their (un)willingness to change their lifestyle in a more sustainable direction and in particular how this justification is expressed through language. With the onset of climate change, analysing the role of language is increasingly important to understand the plethora of discourses on the topic. Furthermore, given how language influences not only beliefs but also actions, such analyses may provide the tools to engage lay people to limit climate change and its dangerous consequences.

2This article seeks to address the following questions: How do people reason about individual climate action, and what kinds of linguistic devices do they use? What kind of reasoning is common among those who agree with individual climate action and those who disagree with it? How do the different kinds of reasoning relate dialogically to the ongoing political and public debate? The study is based on an open-ended survey question from the Norwegian Citizen Panel, where respondents are asked to state the reason why they agree or disagree that individuals should change their lifestyle in order to limit climate change. The analysis of respondents’ reasoning will be undertaken through a mixed methods approach. Themes among the survey answers are identified through a qualitative content analysis (Schreier, 2012). This analysis is further enriched with a dialogical perspective on the way in which the answers relate to and reply to climate change discourses circulating in society (cf. Bres, 2007; Bres et al., 2019). This explorative study provides new and important knowledge about how the public argues concerning climate action on the individual level, and how the arguments relate to various discourses that are currently developing in the context of the climate change debate.

3The structure of the paper will be as follows: We first discuss how much can be gained from lifestyle changes, to what extent people are willing to change, and how people have previously reasoned about their climate-related opinions. We will then go on to the theoretical-methodological approach of this study. The third section brings forth the results. The remaining part of the paper discusses these results and their implications. Finally, the conclusion gives a summary, and some possible paths for further research are identified.

1. Climate change and lifestyle changes as discourse topics

4Previous studies have investigated different factors related to lifestyle changes, corresponding to different discourse topics. According to Hertwich & Peters (2009), households are responsible for 72 percent of greenhouse gas emissions on the global level, while 10 percent are related to government consumption and 18 percent to investments (Hertwich & Peters, 2009). The carbon footprint of households in France, Germany, Norway and Sweden are dominated by car and plane mobility, meat and dairy consumption, and heating (Dubois et al., 2019: 147). Previous research has reported that lifestyle changes can reduce CO2 emissions substantially compared to baseline emissions (Cafaro, 2011; van Sluisveld et al., 2016). For example, reducing the intake of meat and other animal products can make a considerable contribution to climate change mitigation (Berners-Lee et al., 2012; Creutzig et al., 2016; Joyce et al., 2014; Raphaely & Marinova, 2014; Scarborough et al., 2014). Among scientists, it is widely agreed that demand-side actions are part of the climate solution (Alfredsson et al., 2018; Creutzig et al., 2016: 157; Rogelj et al., 2018; von Stechow et al., 2016). However, demand-side solutions alone cannot mitigate climate change (Creutzig et al., 2016; van Sluisveld et al., 2016), and although lifestyle changes have direct environmental consequences, they only have significant impact when a high number of people decide to implement such changes (Stern, 2000: 410).

5In previous studies which took linguistic-discursive perspectives and used open-ended survey questions about what should be done when it comes to climate change, many respondents mentioned individual action, although there was an overall prevalence of suggestions for government action over individual action (Fløttum, 2017; Tvinnereim et al., 2017). However, when it comes to actual behaviour, already undertaken research shows that there are many barriers for pro-environmental behaviour (Kollmuss & Agyeman, 2002; Lorenzoni et al., 2007). According to Ockwell et al. (2009: 310), only a minority of the public in the United Kingdom take measures to reduce their energy consumption. Numbers from Norway show similar tendencies. Most people (67 percent) have the intention of reducing their own climate emissions. However, when asked about concrete actions, they are mostly willing to make efforts that do not reduce their own comfort, such as recycling (TNS Gallup, 2015). Sköld et al. (2018) also find that the greater the CO2 reduction potential of mitigation actions, the less willingness to implement them, because the actions with greater mitigation potential reflect greater lifestyle changes. When asked which climate actions they were willing to take, households only reached a 25 percent footprint reduction by 2030 (Sköld et al., 2018). According to IPCC (2018), we need to reduce our emissions and lifestyle changes can provide a considerable contribution. Through discourse studies of people’s opinions, we observe that they also claim they want to reduce their carbon footprint (see Fløttum et al. in this volume). Still, the reductions they make are not sufficient to reach our global climate targets (Dubois et al., 2019). Disagreement concerning climate change (Hulme, 2009) raises questions about what kinds of framings individuals provide in order to reach different conclusions. This study offers some insights into how ordinary citizens reason about individual climate action (see also Atanasova, this volume).

6Previous studies using qualitative interviews to investigate how people reason when it comes to climate-related lifestyle choices have identified several themes. Becker & Sparks’ (2018) in-depth interviews with twenty people in the UK revealed three main themes in the responses: “representations of climate change: uncertainty about its reality and severity”, “responsibility for action”, and “opposing environmental and economic interests”. Uren et al. (2018) discovered the two themes “do my bit” and “seeking a higher purpose”. When Dubois (2015) asked households in France about the gains from a low-carbon lifestyle, they emphasized the improved quality of life (Dubois, 2015: 17). Shwom et al. (2010) who used open-ended survey questions to examine how respondents justify why they are supporting or opposing various environmental policies in two US states, find four broad categories of answers: economic, moral, political and technology. Furthermore, they examined the correlation between different justifications and support for environmental policies. The theme of reasoning did correlate with an individual’s level of support for environmental policy, although this was generally a weaker predictor of policy support than values, beliefs and certain socio-demographic variables, such as political orientation (Shwom et al., 2010: 479).

2. Theoretical-methodological approach

7This paper employs two different but complementing theoretical-methodological approaches. First, qualitative content analysis is used to evidence themes in the respondents’ answers to the open-ended survey questions. A discursively oriented analysis is also used to examine how survey participants are interacting dialogically with different voices in the climate change debate.

2.1. Qualitative content analysis and survey data

8Qualitative content analysis offers an effective way to find out what a certain group of people say about a given topic (Schreier, 2012: 42), and allows a focus on latent meaning that is not immediately obvious (Schreier, 2012: 15). The aim is to gain insights into how justifications of individual climate action sub-divide into a range of themes and therefore determine the relative prominence of each of these justifications in the open-ended survey answers (Hansen & Machin, 2013: 103). The analysis is based on an open-ended survey question with 2046 participants in wave 11 of the Norwegian Citizen Panel. The survey was fielded between March 6 and April 3, 2018. The Norwegian Citizen Panel is part of the Digital Social Science Core Facility (DIGSSCORE, https://www.uib.no/​en/​citizen), based at the University of Bergen, and is a unique Norwegian online panel where participants are randomly recruited from the Norwegian population register of people above the age of 18.

9The respondents were first asked the following question: “To prevent or limit harmful climate change, it is argued from time to time that we as individuals must contribute by changing our way of life. Do you agree or disagree with this?” This question had two alternatives, “agree” and “disagree”, in addition to the option of not answering. An overwhelming majority, 89 percent, answered that they agreed with the statement. Subsequently the respondents were asked the following question: “We would like to ask you to state the reason for the answer you just gave. All replies would be welcome, preferably a couple of sentences, or just a few words if you would prefer.” 2025 respondents replied to the agree/disagree question (21 did not answer). 1446 respondents gave a justification for their reply. 10 of those who gave a justification had not answered the agree/disagree question. The average response consisted of 20 words. The text material was de-identified by removing potentially identifying words, such as workplace or home city (see Fløttum et al. in this volume for a parallel question from the same survey).

10The answers were read thoroughly while taking notes to gain familiarity with the content. The objective was to discover frequent justifications respondents gave for their stated answer, which could work as appropriate themes for the analysis. Based on this inductive, data-driven approach, a coding frame of nine frequent themes among the justifications was developed. Table 1 provides an overview of the themes, which will be presented in detail in the next section. The two most recurrent themes are “listing measures” and “miscellaneous”. The answers in “listing measures” did not provide an argument for why they agreed or disagreed with individual climate action, but just listed possible climate actions. The “miscellaneous” theme was used for various answers that did not fit into any of the other themes. These two themes are therefore excluded from the further analysis.

Table 1 : Themes

Theme	Percentage (N)
Every action counts	13% (220)
Morality, responsibility and concerns for the future (morality)	12% (198)
Climate measures should be taken on the societal level (societal level)	7% (114)
Potential influences of lifestyle changes (influence)	5% (79)
Fact doubt	5% (76)
Because climate change is human induced (human-induced)	3% (56)
Too small to make a difference (too small)	3% (53)
Listing measures	33% (545)
Miscellaneous	18% (291)

11Note: 1446 respondents. As one respondent’s answer could be coded as several themes, the total N=1632.

12All the answers were read over again and coded according to the coding frame. Coding involved examining each answer to determine presence or absence of each potential theme. The coding followed the logic of mixed membership, which means that one answer could potentially contain several themes. However, as Table 2 shows, most individuals (88 percent) wrote an answer that was coded into only one theme. Ten percent (145) of the answers were coded by two additional coders to measure inter-coder reliability. The combined kappa value was 0.79. Kappa values for different codes varied from 0.57 (for “miscellaneous”) to 0.94 (for “listing measures”).

Table 2: Themes per answer

Number of themes for one answer	Percentage (N)
1	88,38% (1278)
2	10,24% (148)
3	1,38% (20)

13Note: 1446 respondents. As one respondent’s answer could be coded as several themes, the total N=1632.

2.2. Dialogism

14Since the main issue in the present paper concerns answers given to an open-ended survey question, there is at the outset a dialogue taking place between the survey and the respondent (see Fløttum et al. in this volume). In the present study, the focus will be on the “dialogue”, implicit or explicit, taking place between the respondents and the surrounding climate change debate. Inspired by various approaches to multivoicedness – or polyphony, the analysis undertaken here will mainly adopt the perspective as developed by Gjerstad (2011) in what he calls “discursive polyphony”, integrating the theory of praxematics as developed by Jacques Bres and colleagues (Bres, 2005, 2007; Bres et al., 2019). Following the Russian semiotician Mikhail Bakhtin, we adopt the crucial perspective that discourse is fundamentally dialogical. As expressed by Gjerstad (2011: 5), “[n]o one speaks in complete isolation from what has been said before or without considering how the message will be received. The voice of “the other” is therefore present in any utterance, whether or not this presence is signaled by specific linguistic markers.” A similar perspective is formulated as follows by Bres et al. (2016: 80):

(…) linguistic production is essentially dialogic because it is formed in the process of social interaction; the dynamics of speech rely on the interaction of different values which are expressed through borrowings or echoes from the speech of others.

15Without excluding the form of interaction called “interlocutive dialogism” (when the speaker’s response is oriented towards the anticipated response of the recipient), we hypothesize that in the present study the most important kind of response will be the one that is oriented towards discourse made previously on the same subject, called “interdiscursive dialogism” (see Bres et al., 2016), by which the speaker reproduces or echoes what has already been expressed on a given topic. This analysis perspective requires a constant to and fro comparison between the actual answers given in the survey and the surrounding socio-political context.

3. Findings

16This section provides a brief quantitative overview of what kind of reasoning is frequently used by those who agree and those who disagree with individual climate action, followed by the explanation and linguistic analysis of each theme. The content analysis is complemented by a discursively oriented analysis of how survey participants are interacting dialogically with different voices in the climate change debate.

3.1. Quantitative overview

17Figure 1 shows the percentage of respondents citing each reason. What justifications does the population offer to support or reject the statement that we as individuals must contribute to limiting climate change by changing our way of life?

18Note: 1446 respondents. As one respondent’s answer could be coded as several themes, the total N=1632.

19There are conspicuous differences between those who agree and those who disagree with the statement. Among those who agree, the most frequent themes are by far “every action counts” and “morality”. The themes “societal level” and “influence” both occur in five percent of the justifications provided by those who agree with individual climate action. Interestingly, the theme “societal level” is also the third most frequent theme among those who disagree. The most frequent themes in this group are “too small” and “fact doubt”.

3.2. Theme examples combined with linguistic and discursive analysis

20In this section, we will describe the coding scheme and give representative but not exhaustive examples of answers within each theme, accompanied by analyses of the extent to which the answers manifest a kind of interdiscursive dialogism. All examples of text responses are translated. The original Norwegian versions as they appear in the dataset are in the footnotes.

3.2.1. Too small to make a difference

21A unit of coding belongs with this theme if a respondent expresses the opinion that we are too small to make a difference. This could either mean that the effect of a change in an individual’s lifestyle is too small to matter or that Norway as a country is too small. Some respondents claim that the main CO2 emissions are not caused by individuals’ lifestyle but occur elsewhere. This could be natural causes (such as volcanoes), large companies or the government. The answers in this category vary in their arguments, but the reasoning they have in common is that individuals’ actions matter little compared to other factors.

• 1 Original version: «Mengden utslipp jeg som privatperson gjør er så minimal sett i forhold til store (...)

[1] “The amount of emissions I as a private person make is so minimal compared to large companies that I don’t feel that I make a difference”1

• 2 Original version: «vi er for få mennesker i norge til å endre på hele verden»

[2] “we are too few people in norway to change the entire world”2

22Through the expressions “so minimal” and “too few” these examples are clearly oriented towards previous and other voices pointing at the “littleness” of an individual as well as the “littleness” of Norway in the context of reducing consequences of climate change. Through emphasising this situation of being helpless, the respondents are also refusing the quite prominent point of view that Norway is or should be sufficiently equipped and rich to take the lead and go forward as an example in the climate issue (see Fløttum, 2017; Tvinnereim et al., 2017).

3.2.2. Climate measures should be taken on the societal level

23Answers in this theme argue that climate action should be taken on a higher level, either in addition to or instead of the individual level. Many argue that the politicians, the state, or the government should facilitate and incentivize climate-friendly lifestyle choices. Some also emphasize that the politicians should do their share by cutting the state’s emissions. The answers sometimes point out that the actors who really should take action are the large emitters, such as the industry and large companies. It is argued that countries with polluting industries and big emitters, such as China and the US, have to do more. The biggest challenges are related to industry and transport. Others emphasize that all countries have to contribute. Some mention the need for global solutions and international agreements.

• 3 Original version: «Det trengs kollektive, politiske tiltak.»

[3] “Collective, political measures are needed.”3

• 4 Original version: «Begynn med de store kullkraftverkene rundt om i verden, feks. Kina og andre stor (...)

[4] “Start with the big coal power plants around the world, f.ex. China and other big polluters in the east and elsewhere around the world.”4

24These examples are expressing an injunction through the verb form “needed” and the imperative “start with” and can be interpreted as a form of interdiscursive dialogism through an implicit disagreement about the importance of individual action. In this view, individual (non-collective) action is not enough; politicians, authorities, big companies and big countries are called upon to take action. This is a perspective that is becoming increasingly important through the discourse of various environmentalist and activist groups (such as the youth action and Greta Thunberg).

3.2.3. Fact doubt

25Answers in this theme question what we know about climate change, such as whether climate change is human-induced or whether the climate is changing at all. Some argue that scientists do not agree, and that the climate issue is created by politicians. Many point out that the climate has always been changing, also before the industrial revolution. Some say we lack sufficient knowledge about the causes and consequences of climate change.

• 5 Original version: «Tror klimaendringene er en naturlig syklus.....det er alltid i endring...men vi (...)

[5] “Believe the climate changes are a natural cycle….. it is always changing…but we destroy much of nature with our way of life.”5

• 6 Original version: «Klimakrisen er for det meste IKKE menneskeskapt.»

[6] “The climate crisis is for the most part NOT human-induced.”6

26Example [6], through the polemic negation “not”, is an obvious refusal of the commonly accepted point of view that climate change IS human induced. The first part of the first example makes the same point, but it is rather an implicit refusal of this point of view, emphasising that climate change is caused by a “natural cycle”. However, at the same time, through the contrastive and concessive connective “but”, the respondent emphasises that “we” (i.e. human beings) are doing much damage to nature. These linguistic markers (“not” and “but”) contribute to conceptions related to ambiguity and doubt about climate change and are discursive manifestations of a current deliberation in society (see Fløttum et al., in this volume).

3.2.4. Because climate change is human-induced

27This theme is expressed by respondents who argue that individuals should take action because climate change is caused by humans and our lifestyle. Some respondents state explicitly that the logical consequence of this fact is that we change our way of life. In other answers the logical consequence is more implicit.

• 7 Original version: «Det ER vår levemåte som er hovedgrunn til klimaendringer.»

[7] “It IS our way of life that is the main reason for climate changes.”7

• 8 Original version: «Jeg tror på at klimaendringene er menneskeskapte, og eneste løsning er en innsat (...)

[8] “I believe that all the climate changes are human-induced, and the only solution is an effort from all of us.”8

28These examples are explicit conceptions of the causes of climate change, refused in the examples of the category above (Fact doubt). The formulated points of view are strengthened by using capital letters in “IS” in [7], a form of emphasis that creates an implicit contrast to an opposing point of view, and by using the pronoun “all” in [8] to characterize the scope of the issue. As indicated above these points of view, as examples of interdiscursive dialogism, are here used to justify the necessity of human action.

3.2.5. Every action counts

29An answer belongs to this theme if it argues that many small amounts accumulate to make a large amount. The answers typically consist of the Norwegian proverbs “alle monner drar” and “mange bekker små blir en stor Å”, which could be translated into English as “many pennies make a pound”.

• 9 Original version: «Hvis alle gjør litt blir det totalt mye.»

[9] “If everyone does a little it will be a lot in total.”9

• 10 Original version: «Mange bekker små»

[10] “every action counts”10

30Expressions of degree (“a lot”) and scanning (“everyone”, “every”) are important markers of the respondents’ point of view in these examples. This “every action counts” perspective is in fact a refusal of the points of view expressed under the category “too small”. As the “too small” category just expresses an opinion about the challenge of climate action, the theme “every action counts” indicates more clearly a constructive perspective of the importance of individual action, presupposed in the survey question. Answers in these two themes demonstrate the difficult current debate about climate action and repeat common previous discourses on the same issue. Particularly interesting is the use of proverbs – an important discursive behaviour echoing conceptions that “all” agree about.

3.2.6. Morality, responsibility and concerns for the future

31This theme includes respondents who argue that it is our duty and responsibility to contribute. Answers that emphasize the concern for more vulnerable parts of the planet and future generations are also included in this theme.

• 11 Original version: «Alle må bidra»

[11] “Everyone has to contribute”11

• 12 Original version: «Det er vår plikt overfor kommende generasjoner!»

[12] “It is our duty towards future generations!”12

32The deontic perspective is prominent in these examples, through modal expressions such as “has to” and the noun “duty”. The “Everyone has to contribute” answer [11] refers to an opinion that has almost become a slogan, often repeated in the discourse promoting the importance of individual action. The morality perspective is also related to the responsibility we – human beings – have towards future generations. Example [12] could be interpreted as both interdiscursive dialogism (echoing previous instances of discourse) and interlocutive dialogism (the speaker’s response is oriented towards the anticipated response of the recipient, i.e. the people behind the survey, by seeking to cater to their supposed values).

3.2.7. Potential influences of lifestyle changes

33This theme includes respondents who argue that we should change our way of life in order to influence others. Some assert that by changing the way of life you also improve your own and others’ consciousness about the climate issue. Many point out that individuals should use their consumer power to influence the decisions of producers. A few also point out that climate-friendly lifestyle changes can send a signal to politicians.

• 13 Original version: «En holdningsendring blant forbrukere kan kanskje «presse» produsenter til å tenk (...)

[13] “A change in attitudes among consumers might « induce » producers to think more environmentally friendly”13

• 14 Original version: «Litt håpløst å be folk fly mindre. Heller mot tekno-optimisme. Tenker likevel at (...)

[14] “A little hopeless to ask people to fly less. Lean towards techno optimism. Still believe that small measures in daily life (e.g. recycling) are needed to increase the consciousness that we need a bigger shift.”14

34These two answers include linguistic mechanisms that construct uncertainty (epistemic modality) and ambivalence. By using the modal verb might the first respondent avoids taking responsibility for the content of the utterance, while the second respondent starts a dialogue with an anti-air travel discourse that (s)he characterises as “hopeless”, before pivoting to a contrasting point of view using the adverb still. The use of these linguistic mechanisms reflects the uncertainty surrounding the efficacy of competing measures that circulate in society.

4. Discussion

35The present study has documented an overwhelming public support for individual climate action. Almost ninety percent of the Norwegian population agree that individuals should change their lifestyle in order to limit climate change. This supports Becker and Sparks’ (2018: 13) finding that the narrative of individual responsibility is widely accepted. Most citizens are in other words willing to act, or at least they say so, but Becker and Sparks (2018: 13) suggest citizens lack the strategies to respond effectively. Furthermore, the current study has documented different kinds of reasoning for agreeing or disagreeing with individual climate action. The four most frequent themes among those who agree with individual climate action were “every action counts” (15 percent), “morality” (13 percent), “societal level” (6 percent) and “influence” (5 percent). Among the few who disagree with individual climate action, the most frequent themes are “fact doubt” (32 percent), “too small” (20 percent), and “societal level” (16 percent).

36Some of these themes have been identified in previous studies on how people think about climate and lifestyle. Shwom et al. (2010) find that moral rationales often lead to environmental policy support. Both themes identified by Uren et al. (2018), “doing my bit” and “seeking a higher purpose”, captured elements of morality. “Doing my bit” focused on the ethics of personal actions, while “seeking a higher purpose” had a moral element of doing the right thing, and also included discussions of future generations (Uren et al. 2018, 10).

37Regarding “societal level”, several studies (Dubois et al., 2019; Tvinnereim et al., 2017) have found that people often accept individual responsibility for their footprint but at the same time call for government action to create consumption changes in areas with large, untapped mitigation potential (Dubois et al., 2019: 152). Becker & Sparks (2018) similarly find that respondents place “responsibility for action” on either the individual level, governmental level, corporate and business level, or a relationship between these levels (Becker & Sparks, 2018).

38Studies of climate lifestyle argue that major policy support is needed for households to change their lifestyle (Dubois et al., 2019: 148-150; Sköld et al., 2018). This is supported by Green & Denniss (2018) who conclude that supply-side policy instruments have numerous advantages over otherwise similar restrictive demand-side instruments. The above discussion supports the notion of responsibility at the societal level. However, those who disagree with individual climate action and justify this with responsibility at the societal level, contradict the IPCC report (2018), which states that increase in resource-intensive consumption is one of the key impediments to achieving the 1.5 degrees Celsius goal, alongside lack of global cooperation and lack of governance of the required energy and land transformation (Rogelj et al., 2018: 95).

39Although only a minor portion of the respondents questioned the causes and consequences of climate change, the presence of such fact doubt demonstrates that uncertainty and scepticism persist after decades of information about climate change. This is not unique to our study. Becker & Sparks’ (2018) theme “representations of climate change” also includes representations of climate change as cyclical and natural (Becker & Sparks, 2018). Tvinnereim et al. (2017) find that those who question mitigation doubt that climate change is a serious, or even real, problem, whatever the cause is, or they think Norway is too small to make a difference globally (Tvinnereim et al., 2017). In order to convince those who belong to the themes “too small” and “societal level” to make climate-friendly lifestyle changes, it could be necessary to demonstrate the extent to which such changes have an effect. In addition, politicians have to take climate action seriously and not just rely on individuals’ lifestyle changes.

Conclusion

40Our findings suggest that, in general, citizens agree that individuals have to change their way of life in order to limit climate change. Furthermore, the analysis expands our knowledge on support for individual climate action by describing the variations in reasons people provide for supporting or opposing such lifestyle changes. Our findings indicate that there are many different viewpoints and no “one size fits all” solution on how to communicate lifestyle changes in a climate perspective. The answers formulated to the open-ended survey question are generally influenced by the ongoing political and public debate through a mainly interdiscursive dialogism. The respondents relate dialogically to previous instances of discourse on the issue of climate change action through various linguistic devices manifesting their different stance: echoing, agreement, disagreement and refusal. Thus, justification of lifestyle choices seems to be dependent on questions related to the causes of climate change, the importance (or not) of the contribution of individual action, and moral preoccupation.

41An arguable limitation of the current investigation is the subjectivity of the selection and definition of themes. Another researcher might have emphasized other aspects of the data material. Notwithstanding this limitation, the study provides insights on the key factors that are in peoples’ minds and need to be addressed if more people are to change their way of life in a more sustainable direction. Further research might explore the link between sociodemographic variables and different kinds of linguistic reasoning to examine who talks about what, and whether people with different values or ideologies use different arguments. Future research could also investigate actual behaviour and examine respondents’ reasoning for why they changed or did not change their lifestyle.