Taking sustainable eating practices from niche to mainstream: the perspectives of Swedish food-provisioning actors on barriers and potentials

Abstract The food system is a major driver of anthropogenic environmental impacts and in Sweden a sizeable proportion of the country’s relatively large per capita ecological footprint is attributable to food. In short, sustainable eating practices need to become mainstream. Actors within the food-provisioning system likely have valuable insights into how such a transition could be enabled. This article presents the results of a qualitative study that aimed to examine the perspectives of these individuals on such a transition in Sweden using a social practice framework to identify framings of barriers and potentials for mainstreaming sustainable eating practices. We found that conventional framings and models for explaining change and transitions dominate. These approaches center on providing alternative food products, with some attention devoted to normalizing sustainable eating through product design, communication, and marketing. However, exceptions to these strategies include calls for redefining business profitability in terms of human and planetary health and notions of a decentralized food-provisioning system consisting of small-scale actors and limited by the regional and seasonal supply of food. Our analysis suggests that interventions for mainstreaming sustainable eating practices need to move beyond a constrained recrafting of mainstream eating practices and toward systematic practice substitution that favors considerations regarding how eating practices connect to other practices that constitute people’s everyday lives. We conclude by discussing implications for the food-provisioning system and suggest directions for further research that could lead to the development of strategies for mainstreaming sustainable eating practices in Sweden and elsewhere.


Introduction
Globally, food provisioning is one of the main drivers of anthropogenic environmental impacts, including greenhouse-gas emissions and land and water use (Aleksandrowicz et al. 2016;Foley et al. 2011;Reisch, Eberle, and Lorek 2013). The situation is no different in the case of Sweden, a country that is often hailed as being a frontrunner in terms of sustainability but which has one of the largest per capita ecological footprints in the world (WWF 2016), and where food plays a significant part (Cederberg et al. 2019). To reach sustainability targets, such as the United Nations Sustainable Development Goals and Sweden's environmental quality objectives, what and how Swedes eat clearly needs to change, and research on the environmental impacts of alternative ways of eating provides some insights into the kinds of changes that might be required (Aleksandrowicz et al. 2016). Many potentially sustainable ways of eating are already practiced today, but in nichesthat is, by relatively few people. Consequently, it is essential to understand how such eating practices can be mainstreamed and who should adopt these new routines.
When considering how this "mainstreaming" can be supported, the provisioning of food and the actors involved emerge as a focal point. Previous research relating to sustainable consumption and food provisioning includes, for example, studies on in-store behavior-change interventions (Bernard, Bertrandias, and Elgaaied-Gambier 2015;Hanss and B€ ohm 2013;Lehner 2015) and the potential roles of mass-market retailers for sustainable food consumption (Chkanikova and Mont 2015;Oosterveer and Sonnenfeld 2012). Considering that actors in the processing and retailing sectors are increasingly the people closest to the consumer (Mylan et al. 2015;Spaargaren and van Koppen 2009), it stands to reason that it isimportant to understand their views on how to make sustainable eating practices more common.
A set of conceptual approaches that have gained in popularity in research on sustainable consumption over the last few decades are theories of social practice (Corsini et al. 2019;Jaeger-Erben and Offenberger 2014). With a focus on practices instead of individual attitudes and behavior (Shove 2010), these perspectives depart from behavioristic approaches to transitions to sustainable consumption (McMeekin and Southerton 2012) and might be one way of understanding the complexities of food transitions and how such transitions could be supported (Sahakian and Wilhite 2014). While important contributions to the sustainable consumption literature explore policy interventions in social practices (Cohen and Ilieva 2015;John, Jaeger-Erben, and R€ uckert-John 2016; see also Strengers andMaller 2015 andSoutherton 2019), the influence and views of commercial actors have received less attention, though exceptions include, for instance, Spaargaren and van Koppen (2009) and Seyfang and Gilbert-Squires (2019). Concerning eating practices, specifically, examples of recent intervention studies include a university campussustainability project (Middha 2020), the interventions of civil society organizations (Parekh and Klintman 2021), the efforts of home builders to construct low-carbon housing (Breadsell and Morrison 2020), and the initiatives of community-based groups (Sch€ afer et al. 2018). Still, the perspectives of commercial actors on and potential influence over eating practices remain relatively underexplored in the social practice-based sustainable consumption literature. This article examines from a practice perspective how food provisioning actors who sell products associated with sustainable niche-eating practices view barriers to and potentials for mainstreaming those practices in Sweden. In the next section, we introduce the theoretical perspectives and concepts at the heart of our analysis as well as our conceptual framework. We then present our methods for data collection and analysis, the results from the application of our approach, and further discuss these outcomes. We conclude this article with some remarks about the implications for transitions toward sustainable food consumption in both research and practice.

Theoretical background and analytical framework
Theories of social practice: from food consumption to eating practices This article is primarily premised on social practice theory. However, we will use broader terms like "perspective" or "approach" since "social practice theory" in fact denotes a movement of thought rather than a unified understanding (Røpke 2009). An important characteristic of the practice approach is its mediation between structuralist and behavioristic formulations to understand human action in social theory (Halkier, Katz-Gerro, and Martens 2011). Instead of giving primacy to either social structure or agentic behavior, a practice approach places "practices" at the center of analysis (Shove, Pantzar, and Watson 2012). People, then, are neither rational consumers nor passive users, but humans who engage in some social practices that they share with others (Oosterveer and Sonnenfeld 2012). This means that they are seen as active and creative practitioners, or "carriers," of a practice (Shove and Pantzar 2005).
One commonly used conceptualization of a "practice," not least in the consumption literature (Halkier, Katz-Gerro, and Martens 2011;Warde 2014), is provided by Shove, Pantzar, and Watson (2012) who in turn build upon the work of Giddens (1984) and, more recently, Schatzki (1996) (see also Schatzki, Knorr Cetina, andvon Savigny 2001 andReckwitz 2002, among others). Shove, Pantzar, and Watson (2012) describe practices as "organized sets of doings and sayings" (Schatzki 2017, 129) that comprise three, interdependently linked elements: materials (physical objects, infrastructure, tools, the body), competence (practical skills and know-how), and meaning (symbolic and social reasons for doing something). For example, the practice of cooking a meal links material elements (e.g., ingredients and a stove, which is connected to public utility infrastructure), competence (e.g., cooking skills and knowledge of how to operate the stove and combine the ingredients), and meaning (e.g., cooking for a family dinner). Furthermore, it is useful to make a conceptual distinction between a practice-as-entity and a practice-as-performance (Warde 2005). The former refers to a socially shared understanding of practice (such as the notion of a family dinner) and the latter to separate individual performances of a practice. Each performance draws upon the practice-as-entity but often with slight variations. For example, different families will cook different food at different times and in different ways, with varying notions of social significance attached to the meal occasion.
Our practice-informed view of food consumption places eating practices at the center of analysis, and we regard eating and drinking-henceforth referred to as eating for brevity-as being more than simply the intake of food and beverages (Warde 2005). Instead, we define it broadly as a set of practices that includes grocery shopping; cooking; making judgments about taste and other meanings such as health, ethics, and sustainability; organizing the meal occasions during which the actual intake of food and drink happens and, conceivably, other related practices (cf. Oosterveer and Sonnenfeld 2012;Warde 2013).

Sustainable eating practices from niche to mainstream
In this article, potentials and barriers for mainstreaming potentially sustainable niche practices of eating are in focus. Mainstreaming implies a transition in which a practice goes from being niche to becoming a widespread routine; in other words, it is spread or upscaled to become the common way of doing things. There are many and sometimes diverging perspectives and interpretations of what mainstreaming, transitioning, or upscaling could mean (Augenstein et al. 2020). In terms of the multi-level perspective (MLP), mainstreaming is described as innovations thriving in protective niches, that can be nurtured to develop and become part of the dominant, and perhaps modified, regime (e.g., Geels and Schot 2007). Since our focus is on social practices rather than technological innovations, we use the term "niche practices," and define them as practices that are currently carried out by a small number of individuals or actors, be it citizens or organizations. These practices are also generally carried out in relatively benign conditions, for example among individuals with high income. By the mainstreaming of niche practices, we refer to a process of transition in which these practices become the common way of doing things. An example would be that eating vegan or vegetarian food emerges as the new normal in the Swedish population. This kind of shift would imply complexity and require the making and breaking of interconnected elements such as infrastructure and symbolic meaning (Pantzar and Shove 2010;Shove, Pantzar, and Watson 2012).
Structural factors have important effects on the changes or evolution of practices (Sonnberger and Gross 2018). Indeed, Spaargaren (2003) illuminates how systems of food provision shape the practices and possibilities made available to human agents within different domains of consumption such as food. Warde (2014) and others have argued that practice-based consumption research needs to consider these structural factors. In this article, we have therefore sought to understand the view of practitioners within the Swedish system of food provision on what could hinder or enable the mainstreaming of sustainable niche practices. Such impeding factors we term barriers and the means for enabling we term potentials.
Understanding the roles of these food-provisioning actors, by which we mean producers, processors, retailers, distributors, and restaurants, is central to the mainstreaming of sustainable eating practices (Fine 1994). These actors can also be seen through the lens of the three main elements of social practices, as they can affect or "intervene in" (Spurling et al. 2013) certain practices by, for example, providing the material infrastructure for buying certain food and influencing people's practical knowledge or competence, and perhaps also the symbolic or cultural meanings of eating. A focus on practices, rather than on individuals' behavior and choices, can emphasize the complexity and interactions that have an effect on both consumers and the systems of provision, as well as the context in which they are embedded (Spaargaren 2011).
The niche practices in question in this article are potentially sustainable eating practices that we have identified within the research program "Mistra Sustainable Consumption-From Niche to Mainstream." We identified and clustered 300 eating practices that Swedish actors perceived as sustainable into six clusters (Kamb et al. 2019) (Figure 1). This study considered sustainable eating practices as practices that can both improve health and decrease environmental impact. Furthermore, the identified practices concerned both eating and drinking. For example, a suggestion within "eating and buying less food" was to decrease alcohol consumption to lower environmental impact and improve public health.

Analytical framework
To understand the perspectives of food-provisioning actors, it is helpful to interpret their statements through the lens of a practice perspective. Spurling et al. (2013) suggest that the way sustainability problems are framed-in our case, the problem of mainstreaming sustainable eating practices-results in its own logic, which in turn dictates the interventions that are considered viable. Indeed, Schatzki (2017, 134), when discussing language from a practice perspective, posits that sayings, texts, and discourses can constitute normative uses of language in that "sayings and texts contribute to what makes sense to people to do and the intelligibility of things in the world." In our analysis of the statements of food-provisioning actors, we utilize Spurling et al.'s (2013) framework which identifies three common approaches for conceptualizing sustainability problems in contemporary public policy. It then juxtaposes them with three potential ways that sustainability problems could be framed from a practice perspective, highlighting the types of interventions that these latter framings generate (see Table 1). Spurling et al.'s (2013) framework was originally developed to analyze how sustainability problems are understood in public policy interventions, but we argue that it applies to other actors' statements or formulations about how to achieve social change as well (cf. Sch€ afer et al. 2018). Practices are formed and evolve in relation to infrastructures and institutions in a cultural and historical context (Shove 2010). Accordingly, when viewing "interventions" for transitions to sustainable consumption as initiating some form of social change, it becomes clear that policy makers are not the only societal actors who actively intervene in people's everyday lives. Business actors are another prominent example (Vihalemm, Keller, and Kiisel 2015). Interpreting their understandings through a practice perspective on social change can teach us about how effective the measures conceivable to food-provisioning actors are, as well as discuss in what ways they could be re-framed to achieve better results.
The three common framings tend to formulate interventions that focus on technical measures to reduce the resource intensity of existing consumption patterns (Framing 1), consumers' choices between more or less sustainable options (Framing 2), or more generally sustainable behaviors (Framing 3).
The first framing encapsulates a logic where the objective is to increase resource efficiency while maintaining current consumption patterns. In the case of food, this could be likened to efforts by retailers to introduce more energy-efficient refrigerators. This logic, however, is fundamentally at odds with the practice approach which holds that technology and everyday practices are interdependent and evolve in conjunction with each other (McMeekin and Southerton 2012;Spurling et al. 2013). Furthermore, practices are non-static entities that sometimes change due to new technologies, sometimes due to other factors (such as novel social norms and tastes), and other times are unresponsive to such factors (so the effects on consumption patterns from the introduction of new technology cannot be reliably predicted) (Spurling et al. 2013). In either case, the potential efficiency gains are unlikely to be sufficient to achieve the drastic reductions in resource use required for consumption consistent with climate targets (Alfredsson et al. 2018).
The second and third framings are associated with a behavioral approach to social change where the underlying logic is based on a set of overlapping notions about human action: the notions of the rational consumer who makes decisions based on prices and information, the idea of the individual consumer who makes choices based on attitudes and values, and the view that people's consumption patterns arise from habits that are largely unconscious and responsive to interventions that gently "nudge" them to choose more sustainable products or services (Spurling et al. 2013). Nudging, here, refers to changing the environments in which decisions take place in a way "that alters people's behavior in a predictable way without forbidding any options or significantly changing their economic incentives" (Thaler and Sunstein 2009).
The practice theory-informed framings, by contrast, can result in interventions focused on reducing the resource intensity of existing practices by changing their constituent elements (Framing 4), substituting less sustainable practices with equivalent but more sustainable ones (Framing 5), or taking advantage of practice theory's insights into how practices interlock and interact, temporally and spatially, to cause changes in practice to spread through interconnected practices (Framing 6).
These framings go beyond individuals' behaviors and attitudes or technological fixes and address the disparate elements of practices. Practice-inspired interventions can target practices as the social phenomena they are, going beyond individual attitudes, behaviors, and single observable practices-as-performances (Spurling et al. 2013). Instead, such interventions target practices-as-entities (Spurling et al. 2013), taking into consideration their recursive relationships with institutions and infrastructures.
Framing 4 is most similar to more conventional approaches to behavior change, both in the level of change and the comprehensiveness of the resulting interventions, as this framing alludes to changing the elements of existing practices (Spurling et al. 2013). However, it also requires systematic analysis and intervening in the constituent elements (Spurling et al. 2013). Framing 5 is based on two fundamental practice-informed insights. First, since practices require resources, space, and time to be performed, engaging in some practices entails foregoing engagement in other practices, such as how taking baths has been replaced by taking showers. Second, practices often have different "variants"for example, a plant-based meal can be cooked or bought ready-made-and these variants can have trends with trajectories that can be exploited. A notable instance is how ready-made meals have increasingly come to be perceived as nutritionally adequate (Spurling et al. 2013). Regarding Framing 6, scholars, such as Shove, Pantzar, and Watson (2012) and Southerton (2013) highlight the temporal aspects of coordinating the interconnecting practices of daily life, where the latter argues that understanding the different temporalities of consumption and how its associated practices interlock is crucial to successfully implement interventions for sustainable consumption. Spurling et al. (2013) suggest that interventions for sustainable consumption can harness two different ways in which practices interlock in people's daily lives: synchronization (many people performing the same practice at the same time of day, such as school or workplace lunches) and sequences (the order in which practices, such as grocery shopping, cooking, eating, and disposing are performed).
We use Spurling et al.'s (2013) framework as a practice lens through which we analyze the statements of food-provisioning actors about how they view barriers and potentials for the mainstreaming of sustainable eating practices. This analytical approach allows us to elevate what they say-and do not say-to a theoretical level in practice terms, to understand and discuss the potential efficacy of the types of interventions to which their perspectives would give rise. In the following section, we present the interviewed actors as well as how we carried out our analysis of the empirical material.

Methods
Our overall approach was qualitative, including semi-structured interviews, that were analyzed in relation to the analytical framework. We opted to use this approach to attain descriptions of the respondents' perspectives and reasoning, in their own words and based on their lifeworld. A series of interviews, rather than, say, focus groups, was chosen to get a diversity of opinions and to allow the participants to speak freely.

Interviews
To obtain a rich picture of potentials for mainstreaming of the eating practices in question ( Figure  1), we strived for a diversity of respondents regarding the representation of the six clusters (Table 2). Accordingly, the food-provisioning actors in our study enabled or promoted one of the six sustainable eating practices in a Swedish context. The specific cluster that each actor "represents" is indicated in Table 2. Interviewees were then sought via research-program partners and snowball sampling. All the respondents are people with responsibility for, or with intimate knowledge of, their organization's sustainability work. The participants varied in size and type of organization, much depending on the characteristics of the cluster in which the actor worked.
We carried out the eleven semi-structured interviews during the first half of 2019, before the onset of the COVID-19 pandemic. An interview guide was used, an abbreviated version of which was given to the respondents beforehand. The thematic focus of its questions was on barriers and potentials for the actors and in society at large. Examples of the interview guide's questions and themes were (1) perspectives on expansion and organizational growth; (2) which actors are considered to be barriers for enabling the practice they represent; (3) desired changes required for enabling the practice; (4) and with whom the actors place the responsibility for achieving these changes. For most of the interviews, both authors of this article participated, one asking questions and the other taking notes and asking clarifying questions. When possible, the interviews were conducted face-to-face, and when not, videoconferencing software was used. Nine interviews were recorded after permission was given by the interviewees and then transcribed. For two interviews only notes were taken since one was originally part of an unrecorded and previously conducted pilot study, and the other upon the request of the respondent.

Analytical method
The analysis took the form of what Kvale (2007, 117-118) describes as a "theoretical reading" of the transcripts and notes. This was done through repeated readings and using computer software for qualitative data analysis. We coded the interviewees' statements into themes created with the aim of the study in mind, such as "barriers" and "potentials" to mainstream sustainable eating practices. In addition, we used themes that encapsulated fundamental practice theory concepts, such as the practice elements "material," "competence," and "meaning." These thematically coded batches of text were then compared to the framings in Spurling et al.'s framework, i.e., common framings, re-crafting, substitution, and practice interlocking (Table 1)-and grouped according to which of Spurling et al.'s framings we interpreted them to most closely mirror. Finally, the resulting structure of the data provided an overview from which we could collate the main types of barriers and potentials and develop a running text to facilitate better comprehension of the findings from the interview material as a whole. The results of this process and the different framings are further analyzed and discussed in the sections below.
Following Spurling and McMeekin (2014), who analyze the common framings together with the recrafting framing, we did not distinguish between statements corresponding to different types of common framings in the analysis. Unlike these authors, however, we noted the common framings under a separate heading to maintain a clear distinction between the common framing and the practice-style framings and to place a greater emphasis on the latter. That being said, these constructed categories are by no means mutually exclusive but sometimes overlap and depend on how we interpreted the framings.

Common framings
We present here the interviewees' statements about barriers and potentials for mainstreaming sustainable eating that we interpret in accordance with Spurling et al.'s (2013) three common framingsinnovating technology for sustainability, shifting consumer choices, and changing behavior. Several respondents (Retailer Limiting Consumption, Organic Co-Op, Organic Wholesaler, Plant-Based Producer, and Fast-Food Restaurant Chain) emphasized the importance of shifting consumer choices and formulated their own business's role as an enabler of better choices or a provider of better products. This could be, for instance, by driving the production of more sustainable alternatives to existing products such as organic (Retailer Limiting Consumption and Organic Wholesaler) and plant-based meat alternatives (Fast-Food Restaurant Chain).
Several respondents also constructed a barrier for mainstreaming sustainable eating as the struggle to reach beyond a relatively small group of consumers focused on consumption of supposedly healthy and socially responsible products and services, identified by some as LOHAS ("Lifestyle of Health and Sustainability;" see Cohen 2007) and toward mainstream consumers. By contrast, the Plant-Based Producer explicitly identified LOHAS as the firm's intended target group and considered it to be an expanding part of the population while the other interviewees had more general discussions about attracting other people, including non-LOHAS consumers.
[E]arly adopters etc., those who are the first and quickest to buy our type of products, they're vegans. News spreads and the word gets around, 'Some new product! I've got to try it.' Then there's a second wave and a critical mass starts to form; 'Yeah, I think I'll start to cut down on meat, too.' That's when the masses get onboard. But not the tail; twenty percent of the population is uninterested in food (Plant-Based Producer).
Also speaking in terms of consumer demand and choice, Mainstream Wholesaler and Organic Wholesaler said that the sales of organic foods have recently leveled out, formulating the issue as people prioritizing regional and local food over organic.
It is trumping organic right now. And I think, we think, that the drought last summer [in 2018] meant that our customers and consumers generally began to understand that it is important that we have food production in Sweden, and that they simply want to support Swedish production. So, we see that quite clearly in our figures (Mainstream Wholesaler).
Another perspective raised was that a company expanding into global markets can be a positive driver for change in consumption patterns (Plant-Based Producer and Department-Store Chain). In so doing, there is potential to promote sustainable eating in other countries and thereby get more people to adopt a specific practice. Furthermore, the interviewee from the fast-food restaurant chain argued that growth is not a problem with regard to sustainable eating, in contrast with other types of consumption.
I think that if we were selling furniture or clothes or something like that, we could say -Yes, but does this mean that people buy more clothes than they need and do they then throw away more clothes than they need? But we … you need to eat, otherwise you die. So, we don't think that if people come to us to eat lunch with us, they then go on to a competitor and eat a second lunch because our lunch was cheap. So, the kind of consumerism that is discussed today is not really relevant for us (Fast-Food Restaurant Chain).
Others (Organic Co-Op and Local Producer) described their main customers as being "more aware" and financially well-off than the mainstream, attributing it to the frequently higher prices of their products. Accordingly, convincing people to pay more for certain products was mentioned (Surplus Restaurant, Organic Wholesaler, and Local Retailer) as a significant challenge, and formulated as an issue of people's lack of understanding of why, for instance, organic or locally produced products are more expensive. Local Retailer and Local Producer claimed that eating organic or locally produced food is, in fact, affordable to most people if they adjust their consumption patterns.
In that regard, I think almost everybody can afford it, but rather, one has to prioritize it over something else. So, it's a mental thing that we have to work on over the long term, I think (Local Producer).
Some respondents raised the importance of nudging customers, for instance by placing plant-based items before animal-based ones in the buffet line, to encourage them to fill up their plates with plantbased items (Department-Store Chain and Surplus Restaurant). It was added (Department-Store Chain) that it is important to inform customers of their choice after it has been made, with an emphasis on the food's appetizing qualities and environmental and health benefits. The surplus restaurant also nudges guests to consume less, as the price of the lunch is determined by the weight of the food they serve themselves from the buffet. This reportedly makes the guest more careful and creates discussion about how much and what they eat. Organic Co-Op claimed that their volume discounts, that do not include meat products can discourage meat consumption, and selling the food in bulk can reduce the use of packaging. Fast-Food Restaurant Chain discussed the dilemma of how much control to exercise over consumers in light of this great potential for consumers to instigate change in eating practices. Retailer Limiting Consumption discussed the issue of whether to steer customers more directly or instead facilitate sustainable choices but thereby risk overwhelming them with options-for instance through a multitude of different eco-labels. While several interviewees (Organic Wholesaler, Fast-Food Restaurant Chain, and Organic Co-Op) pointed out people's confusion about which products and what types of eating are sustainable, Retailer Limiting Consumption claimed that most of their customers are not sufficiently interested to seek out such options and criticized placing the responsibility on consumers to make the right choices. Organic Co-Op expressed similar criticism.
[S]ometimes I thinkand this is my personal opinionthat there's a lot of talk about customers having to choose and empowered consumers and all that. Fundamentally, I think companies have a responsibility to procure good stuff. Then you can get the especially interested customers to buy even better [things], but it can't just be that 'We have everything, all mixed, and it's the customer who chooses.' But sure, like everybody else, we are dependent on the customers' satisfaction (Retailer Limiting Consumption).

Re-crafting practices
We present in this section the respondents' statements about barriers and potentials for mainstreaming sustainable eating which we interpret in accordance with Spurling et al.'s (2013) first practice theory framing-re-crafting practices. When using this as an interpretive lens, most of the interviewees' statements on barriers and potentials for mainstreaming sustainable eating concerned the material elements of eating practices.
One of the key factors discussed, as understood from a practice theory perspective, was the availability of different products in retail stores and restaurants. For example, the participants mentioned that changes in public policy and regulations affect the availability or supply of unsustainable products. Such measures mainly address the material elements of different eating practices by replacing or changing the foodstuffs or products themselves. Examples include calls for the removal of state subsidies that benefit products such as dairy milk in public schools as a result of European Union policy and white sugar (Local Retailer). The respondent from the fast-food restaurant chain suggested financial incentives including a universal "climate tax" on businesses, which, it is claimed, would economically favor food providers that are responsible for smaller volumes of greenhouse-gas emissions. Others asserted that various policies prevent more ambitious sustainability criteria on alcoholic beverages (Retailer Limiting Consumption) and that agricultural policies in general disfavor small-scale actors and the production of sustainable products in favor of large-scale and international actors (Organic Co-Op and Local Producer). Furthermore, several interviewees (Local Retailer, Organic Co-Op, Mainstream Wholesaler, and Local Producer) said that mainstream actors stifle increased sales of sustainable products. Mainstream retailers, especially, are singled out in this critique. For example, Local Producer criticized the low visibility and promotion of small-scale producers' products in retail stores.
Some respondents also discussed their own strategies for targeting the availability and sales of more or less sustainable products. For example, Mainstream Wholesaler worked with a "raise the bottom" model-the worst products from a sustainability perspective should be reduced. Examples given were those where antibiotics and toxic substances are used in production.
Focusing on the meal occasions that occur within their stores, the respondent from the departmentstore chain formulated several potentials for mainstream sustainable eating as a matter of changing the material element of eating practices. The interviewee argued that their firm has a large customer base, and thus could influence many people, for example, through meat-free meals in all store restaurants. The respondent also said that an important potential for changing eating practices lies in the fact that their company had a range of different "business channels"-such as restaurants, caf es, and a retail section-where they can simultaneously replace less sustainable foods with more sustainable ones. The interviewee pointed out that replacing animal-based foods with plant-based foods is necessary to influence the mainstream consumers who eat meat. Similarly, Plant-Based Producer stressed their firm's "dual strategy" of selling products through both retailers and food services-including in restaurants in the form of meals as opposed to products-to an almost equal extent. An important part of this strategy is to sell products under the company's own brand as well as other brands and to produce products through joint ventures.
[Plant-Based Producer's signature product] wouldn't have become such a success if we hadn't also managed to get it into [a major sports bar franchise] and into [Fast-Food Restaurant Chain], 'cause that was a way to get it straight into the mouths of the consumers (Plant-Based Producer).
Considering the material infrastructure that the food-provisioning system provides for eating practices, current arrangements are criticized by two respondents for being operated based on a "just-intime" principle. Accordingly, this strategy gives rise to food waste, since retailers and restaurants are dependent on frequent, and minutely planned deliveries of fresh products, instead of keeping on-site inventories (Mainstream Wholesaler and Local Producer). In contrast, the business model of the surplus restaurant in our sample builds on selling oversupplied food, which means that the firm buys foodstuffs that are at risk of being thrown away by wholesalers. Therefore, overall sales of food can decrease and less has to be produced, indirectly reducing the environmental impacts associated with eating practices. The respondent from the surplus restaurant claimed that for more actors to adopt the same business model, the food sector must change and become "supply-driven," rather than demanddriven, where the local supply of primary products-regarding, for instance, seasonalitydetermines what and how much can be provided for sale to customers. Similarly, Local Retailer observed that sustainable food production requires redefining "profitability" as being about sustainability, in terms of human and planetary well-being.
Most respondents (Department-Store Chain, Mainstream Producer, Surplus Restaurant, Mainstream Wholesaler, Plant-Based Producer, Fast-Food Restaurant Chain, and Local Retailer) identified increased collaboration within the foodprovisioning system as necessary for overcoming several challenges such as raising domestic production volumes of sustainable products and systemic issues including the just-in-time principle. A campaign by the fast-food restaurant chain in our sample that was directed at their competitors exemplified one kind of collaborative initiative that could be used to increase the material availability of more sustainable products.
Check this out, competitors! These are the recipes for those of our burgers that sell really well. Use these recipes, make your own burgers out of them, but rename them (Fast-Food Restaurant Chain). Some interviewees also touched upon the meanings of engaging in sustainable eating practices. Local Retailer formulated the general approach of their firm to promoting sustainable eating in part as a countermovement against a perceived "anonymization" of food and people in retail. In addition to cooking classes held in the Local Retailer's store, where customers were able to engage in various cooking practices using the onhand ingredients, the respondent talked about giving in-store cooking advice and providing "shortcuts" to sustainable eating. For example, during our interview, a previously used concept was mentioned involving the arrangement of items in the store not by product category but into meals. This merchandizing strategy entailed, for instance, placing gravy next to meats, providing ready-made side dishes, and reducing the variety of items to simplify customers' decision-making.
[T]he stores were so sterile and there was no laidout culinary idea … that's why we came up with the idea of building our own sort of "food Meccano." You could put together your own dish and stuff (Local Retailer).
The respondent from the organic wholesaler in our sample talked about the importance of alternative and more resilient crop varieties in the face of climate change, but also connected these material aspects to meanings of history and tradition associated with eating food made from such crops. Other participants talked about their business's role as an educator (Organic Wholesaler, Surplus Restaurant, Plant-Based Producer, Fast-Food Restaurant Chain, and Local Retailer). Two interviewees, in particular, expressed ambitions to achieve fundamental shifts in the mindsets of their customers. The respondent from the plant-based producer, for example, talked about the importance of cultivating "fertile soil"-a popular movement for plant-based food whether based on health, environmental, or other motivations. Speaking in terms of consumer demand, Fast-Food Restaurant Chain raised advertising as a key factor for mainstreaming sustainable eating practices. A barrier was exemplified by current restrictions on the claims that a company could legally make in promotional announcements about its sustainability work. This reaction was likely a reference to an investigation by the Swedish Advertising Ombursman (2019) who found that one of the company's advertising campaigns (highlighting a reportedly "climate positive" menu) contained a misleading environmental claim. The interviewee also talked about "sensitizing stakeholders." It simply means that one mentally prepares stakeholders for the next step within sustainability. So, in a way we can say … that we have helped teach Swedes that our core product, beef, is a climate baddie. That means that we might have helped create the market we see before us now, when we sell more green burgers, and that takes time … [W]e had to have a high enough degree of credibility in the matter, and have the perseverance, and had to help educating (Fast-Food Restaurant Chain).
Some respondents (Organic Co-Op, Local Retailer, Organic Wholesaler, and Local Producer) reported that people perceive their business as genuine and providing a personal experience and connection to their customers. Examples included the socially gratifying nature of purchasing food through their businesses (Local Producer and Local Retailer), where the latter added that this is perhaps the main advantage that smaller actors can have over large, mainstream ones. Another interviewee (Retailer Limiting Consumption) also claimed that their business is perceived as authentic, but attributed this feature to its service-focused and non-profit-oriented business model. Plant-Based Producer claimed that LOHAS consumers are skeptical of big corporations and that this was why many large firms create brands that seem alternative and small-scale, for example, by avoiding the prominent display of their name on the packaging.
Demonstrating a focus on meanings, such as cultural conventions and norms associated with animal-based eating, the representative of the surplus restaurant talked about showing people that plantbased eating can be "classy." Respondents from the fast-food restaurant chain and department-store chain discussed communication strategies that present plant-based eating as an inclusive practice for everyone, where the menu items look familiar and have names that emphasize flavors recognizable to those unaccustomed to plant-based food rather than the items' plant or animal origin. Our interviewee from the plant-based producer also stressed the importance of appealing to everyone and avoiding polarization between, for example, vegans and meat-eaters.
Similarly, in accordance with the company's government commission, the retailer limiting consumption does not promote products or in other ways encourage alcohol consumption. The respondent instead exemplified a similar approach that focused not on directly dissuading people from alcohol consumption but on how to be a good dinner host who accommodates non-drinkers. Local Retailer said that achieving sustainable food consumption requires the reintroduction of the idea of the meal as a social occasion and a means for social interaction. Similarly, it was claimed that awareness of the food's origins is important to gain respect for and a sense of "connection" with the food. The respondent from the surplus restaurant noted that it is necessary to change customers' habits, not just their knowledge.

Substituting practices and changing how practices interlock
From the perspective of our conceptual approach, the interviewees made relatively few statements about barriers and potentials for mainstreaming sustainable eating practices that could be categorized under Spurling et al.'s (2013) two other practice framings, namely substituting practices and changing how practices interlock. Therefore, we collect pertinent statements under a single heading.
As presented in the previous section, most respondents in some way discussed the availability of products associated with different eating practices, for example, through plant-based meat analogs (Fast-Food Restaurant Chain, Plant-Based Producer, and Department-Store Chain) and claimed that public policy favors dairy milk in schools (Local Retailer) and mainstream agricultural products (Organic Co-Op and Local Producer). This response could be interpreted as a call for the removal of one practice to make way for another, albeit without systematically addressing several-or all-practice elements to facilitate a more thoroughgoing substitution of practices.
The statements corresponding to Spurling et al.'s (2013) third framing concern how connections between practices can enable or hinder the mainstreaming of sustainable eating practices. More specifically, when and where practices are performed is greatly affected by various institutions and infrastructures, which can therefore be targeted for intervention. For example, when and where the practice of grocery shopping is carried out in relation to other practices is influenced by institutional factors such as store-opening hours and infrastructural factors that include store proximity and how easily it can be reached with different modes of transportation.
The retailer limiting consumption's commission from the government is to reduce the negative effects of alcohol consumption with no obligation to increase sales or to generate profit. This actor works specifically with measures to reduce overall consumption where some interventions entail sequencing and synchronizing daily practices. The respondent specifically mentioned limiting opening hours, keeping all stores closed on Sundays, and having a limited number of stores.
The respondents from the local producer and the local retailer called for a food-provisioning system that consisted of a diversity of small-scale actors, which has implications for how practices interlock, although this was not explicitly addressed in the interviews. The formerly proposed networks that connect local producers and consumers and the latter suggested networks of small-scale producers and small-scale distributors and retailers, especially outside of major cities, as a way for such actors to compete with larger businesses such as mainstream retailers. The respondent from the local producer argued for the importance of effective tools that can connect producers and consumers, mentioning an emerging digital platform that is used in existing networks or "nodes" to manage connections between the producers and a growing number of consumers. The respondent from the local retailer also indicated that small-scale actors needed to provide a diversity of services that go beyond mainstream-retailer offerings to be able to effectively compete with them. An example was the cooking classes hosted in the store's basement. These different statements about mainstreaming their niche practices imply changes to how practices interlock, for example, through adjustments in food-shopping practices. Spurling et al. (2013) argue that when it comes to policy interventions for sustainability, it is better to focus on social practices rather than on behavior or choice. This is because the dynamics of practices offer a window into transitions toward sustainability. What then can be learned from looking at the statements of provisioning about transitions through a practice lens? In this section, we discuss what the respondents said but also focus on what was not said. We then proceed to describe potential implications in terms of how or if a practice perspective could complement currently common framings as well as the further mainstreaming of sustainable eating practices.

Common framings of mainstreaming
Many of the interviewees' formulations of the issue of mainstreaming sustainable food consumption adhere to Spurling et al.'s (2013) common framings and the general approach to social change that these represent. The statements to some extent also mirror more conventional theories on change and growth. For example, the Plant-Based Producer's view on how plant-based eating spreads from the LOHAS group is focused on the acquisition of products and mainstreaming by diffusion. Other examples under the "common framings" heading are related to supply and demand and the prices of products, and wider dissemination by replacing currently purchased food products with more sustainable ones. In a way, such framings are focusing on the observable behavior of individuals, or practiceas-performance according to Spurling et al. (2013), rather than taking in, say, socially shared meanings, tastes, and skills in practices-as-entities.
Though our respondents voiced different views, the correspondence of many of their statements to the common framings was demonstrated by their focus on and mobilization of the "consumer" construct (Evans, Welch, and Swaffield 2017). For instance, they discussed their businesses as enablers of better choices and creators of demand, principally through the introduction of products to a market and through various promotional efforts. Although some of them acknowledged a need to adopt other roles in shifting food consumption-and leveled a critique against mainstream actors such as retailers-the nature of change in food consumption, and thus the ultimate objective, was often formulated in terms of changing the behavior of the consuming subject. This mobilization of the consumer arguably places the main focus, if not the principal responsibility, to successfully change the food system on individual consumers. For example, respondents tended to envisage the difficulties in reaching mainstream consumers-beyond the relatively well-off LOHAS group-as being a major barrier to scaling up sustainable food consumption. Some interviewees assessed efforts to overcome this barrier as a matter of convincing consumers to pay more for sustainable products; for instance, one respondent said, as noted above, that most people can actually afford it and the obstacle is a matter of prioritizing food purchases over something else. This observation points to the great importance the interviewee attributed to consumers' rational choices and illustrates the importance of considering socioeconomic aspects and equity in understanding and enabling the mainstreaming of sustainable food consumption.

Challenging common framings
Conversely, some respondents challenged, and in some cases dismantled, the common framings of consumer choice and responsibility. The retailer limiting consumption and organic co-op participants in our sample argued that sustainable food consumption is ultimately the responsibility of retailers and policy makers, by ensuring that unsustainable products are not available for purchase. These observations, together with some of the smaller actors' statements, recognized the importance of material infrastructures and the social nature of food consumption. This is better aligned with a practice perspective's more complex understanding of human action.
When interpreting the interviews using the recrafting framing, we found that the interviewees' statements often focused on material elements of eating practices, especially in terms of a supply of products to be made available and attractive to consumers. A possible explanation is that several of the actors represented specific product segments. In our interpretation, this notion of supplying sustainable products is complemented by some respondents' claims that unsustainable types or amounts of food should not be available for sale at all, for instance in terms of "raising the bottom" (Mainstream Wholesaler) or shifting to a supply-driven provisioning system (Surplus Restaurant).
We interpret some of the interviewees' statements that correspond to the re-crafting framing as concerning certain meaning elements of eating practices. This is exemplified by the respondents' statements about how their marketing and other communication strategies could contribute to making sustainable eating practices, such as plant-based eating more widely accepted as normal or acceptable. Additionally, and generally speaking, the interviewees affiliated with smaller actors seemed to reflect on the sociality of food consumption-arguably demonstrating something closer to a practice perspective-and the importance of social relations and socially shared meanings for the engagement in sustainable eating practices (see also, e.g., Seyfang 2006).
When it comes to the respondents representing plant-based eating practices, we interpret their statements to correspond to Spurling et al.'s (2013) recrafting practices framing. This recrafting focused on changing the material-food items themselves, by, for example, promoting alternative products. However, the actors could already be accounting for meanings and competencies as well, despite this ostensible focus on mainly material elements. The actors' plant-based products are not merely nutritional analogs but also aim to be similar to their animal-based counterparts in taste, texture, and appearance. Consequently, the actors already account for competencies in the form of cooking skills, for example, since the meat analogs are prepared the same way as their animal-based counterparts. However, such plant-based eating practices, of course, still require knowledge about the products' existence and that they are prepared in the same way. Furthermore, we consider Spurling et al.'s (2013) first practice framing to best correspond to the interviewees' statements about using non-polarizing language and flavors associated with animal-based eating to attract people to this type of food. Taken together, we can summarize this interpretation as the (re-)crafting and promoting of a variant of plant-based eating that mimics animalbased eating.
We also have made an alternative interpretation. Eating and buying more plant-based food products can be seen as a part of substituting one practice for another, since buying, for example, vegan burgers instead of meat burgers can be part of a change toward flexitarianism, vegetarianism, and veganism. But following this interpretation, food-provisioning actors are not currently and explicitly addressing the meanings and competencies of plant-based eating variants. Such a process of re-framing might result in very different suggestions for changes and interventions. For example, a burger restaurant could with this line of thought market its vegetarian burgers as a way of supporting its customers' transition to eating more plant-based food rather than just substituting one product for a more climatefriendly one.

Changing the systems
Some interviewees discussed the mainstreaming of sustainable eating practices in ways that imply significant changes to the food-provisioning system as a whole. Examples included a greater proportion of, or a complete shift to, small-scale actors and local/ domestic food production and consumption (Organic Co-Op, Local Retailer, and Local Producer), a shift toward a supply-driven system (Surplus Restaurant), and a business logic where "profitability" is redefined in terms of the well-being of people and planet (Local Retailer). Furthermore, some of the actors in our study represented alternatives to well-established retailers (Organic Co-Op and Local Retailer) and producers (Local Producer). Given the aim of this article, it is pertinent to consider which of their perspectives are reconcilable with the current food-provisioning system, and which imply significant changes-or even go beyond and presuppose an entirely different one. A related issue is whether the current system can support different types of solutions to enable sustainable eating implied by contrasting problem framings.
The respondents from the local producer and the local retailer in our sample envisioned a multitude of small-scale retailers and producers, distributed across the country, connecting with local consumers. Such a structure stands in sharp contrast to the globalized food system of today (Oosterveer and Sonnenfeld 2012). Similarly, the mainstreaming of organic eating could, in addition to changes in agricultural methods, have significant consequences for the food system (e.g., Milestad, Kummer, and Hirner 2017;von Oelreich and Milestad 2017). If business models of not-for-profit food retailers, such as those represented by the retailer limiting consumption and the organic co-op, were to grow, this would also be a significant break from corporate business-as-usual. An even greater challenge would be a retail monopoly such as that of the retailer limiting consumption.
Many respondents called for more collaboration and coordination within the food-provisioning system to deal with various problems such as insufficient volumes of sustainable products. Some interviewees also raised competition from larger actors as being a major issue in promoting more sustainable products, where major retailers are said to have a particularly large influence over how people eat. The prospects of such collaborations likely depend on whether the actors and the practices they represent challenge the current system (see also, e.g., von Oelreich and Milestad 2017). For instance, the respondent from the plant-based producer emphasized the benefits of collaborating with mainstream actors. The local producer, however, largely circumvented them, instead connecting directly with consumers, and even claimed that dominant retailers hinder local eating practices, apparently for economic reasons. Several interviewees addressed this competitive relationship between niche and mainstream actors and practices, and outlined ways to overcome it. For example, the local retailer presented social relationships with their customers as the main advantage available to smaller actors over larger ones. However, the respondent from the plant-based producer noted that larger corporations often convey an image of their brands as belonging to smaller companies. A practice perspective shows how such meanings of engaging in local eatingalbeit possibly important-are far from the only factors that determine recruitment to or defection from a practice.
In general, the interviewees representing larger actors placed relatively little explicit emphasis on public policy interventions. Respondents from the smaller actors, however, to some extent criticized, for example, European Union legislation for favoring large-scale and non-domestic actors (Organic Co-Op and Local Producer) and subsidizing products such as animal milk (Plant-Based Producer and Local Retailer). Franz, Schlitz, and Schumacher (2018) and Koretskaya and Feola (2020) argue that present-day food politics is no longer dominated by government-based interventions and that increasing globalization of food provisioning results in a shift where responsibilities need to be taken by other actors as well, not least private entities in the supply chain. However, it seems unlikely that the corporate food-provisioning system would enable the mainstreaming of practices that challenge or require fundamental changes to the system itself-such as local eating practices or practices of consuming lesswithout outside intervention, reasonably imposed by policy makers.

Implications of going beyond the consumer and the material
The main part of the barriers and potentials that the interviewees raised demonstrates common framings and re-crafting of material elements. To date, such strategies have not produced the required transitions in food consumption and provision. Embracing a more complex approach could be one way forward. A practice perspective on the respondents' statements about these barriers and potentials suggests that provisioning actors, as well as policy makers and strategic actors in civil society, should shift the focus of their interventions and actions taken to date-from consumers and food products to the practices involved in food consumption. Furthermore, these actors need to address the material elements, meanings, and embodied skills and knowledge of these practices-mainstream practices as well as those that are sustainable, but as yet niche.
One approach could be to utilize Spurling et al.'s (2013) substituting practices framing. For example, the participants in our project from the mainstream wholesaler and the organic wholesaler mentioned the stagnation of sales in organic products in the face of preference for locally sourced products. They discussed this phenomenon as a consequence of shifting consumer preferences, perhaps due to worries about the economic viability of domestic farmers. Although this development, in practice terminology, could be understood as a shift in meanings of responsible or ethical eating, organic food currently only accounts for a fraction of total food sales in Sweden (Ekoweb 2021;SEPA n.d.). That is to say, the practice of organic eating is far from being carried out by a majority of the population (cf. Spurling et al. 2013, 17). Though researchers have observed the phenomenon of consumer preference shifting from organic to locally produced food during the last decade, for instance in the United States and many European countries (Adams and Salois 2010; Winterstein and Habisch 2021), Sweden has not received the same level of attention in the scientific literature (cf. Organic Sweden 2020). Furthermore, this development has not yet been studied through a practice perspective, which would suggest a more comprehensive look at the elements involved. For example, the phenomenon could be framed as a substitution of practices, where one-eating local food-outcompetes another-eating organic food. More specifically, practices compete for time, space, and resources (Spurling et al. 2013). Accordingly, researchers could study these aspects, as well as the respective materials, competencies, and meanings required for each practice, to understand why the two niche practices compete with each other instead of with mainstream practices. Exploring this competitive relationship could provide insights into the potentials for such practices to be made mutually reinforcing (cf. Bj€ orklund et al. 2009) or "collaborative" (Shove, Pantzar, and Watson 2012, 67). Spurling et al. (2013) suggest that another strategy for intervention through practice substitution is to identify and harness existing trends in eating practices (cf. Bergflødt 2012;Naska et al. 2015). Their example, pertinent to several European countries (Healy 2014), including Sweden, is the growing trend of eating out (Gronow and Holm 2019;Statistics Sweden 2016). Food-provisioning actors representing niche practices could harness this shift by focusing their efforts on collaborations with restaurants. Indeed, such a strategy of having readymade plant-based food served in familiar formats in restaurants was advanced by some interviewees (Plant-Based Producer, Fast-Food Restaurant Chain, and Department-Store Chain). But, again, unsustainable mainstream eating practices need to simultaneously be discouraged in some way. Both actors in the food chain and politicians could employ a variety of practice-based strategies such as weakening animal-based eating's symbolic grounding in our culture, and thus its cultural significance, or breaking the connections that grocery shopping at large retailers with limited or no local products has to other practices in most people's everyday lives (cf. Shove, Pantzar, and Watson 2012).
Another approach could be to focus on solutions and measures that have implications for how practices interlock. While Spurling et al. (2013) emphasize that this framing can produce especially powerful interventions, these strategies tend to focus on changing the scale of institutions and infrastructures to affect how practices synchronize and are sequenced. This arguably suggests that such interventions would require more resources and coordination to implement than those implied by the other two practice-based framings. What this could mean for the ability of food-provisioning actors to implement practice-interlocking solutions can be exemplified by the strategies of the retailer limiting consumption to prevent excessive alcohol consumption such as restricted access and opening hours. Albeit not formulated using practice terminology, the company provides extant examples of long-term interventions that harness how practices interlock. The retailer limiting consumption's primary mission to reduce the negative effects of alcohol consumption, the not-for-profit business model, and the centrally coordinated nationwide monopoly on alcohol retailing are likely factors that facilitate the coordination and mobilization of resources to implement these measures on a national scale.

Conclusion
We found that the business actors in this study identify many barriers for enabling sustainable eating in the current food-provisioning system, but they also perceive considerable potential for change. Many of the solutions that they suggested are related to common framings, such as consumer choices and behavior, and fewer are linked to recrafting and substituting practices and reconfiguring how practices currently interlock. An emphasis on conventional changes that are part of a system has not, thus far, been sufficient to enable sustainable eating. We argue that current arrangements need to be questioned to find new ways forward. Such a process can be encouraged through research studies such as this one that scrutinize the prevailing system, but also through reflecting on, and challenging business as usual by other actors, both in the public debate and within food-provisioning organizations.
If the more complex understanding of human action that a practice perspective affords is embraced among food-provisioning actors and relevant governing institutions, more complex interventions and strategies are possible. Our analysis suggests that such changes will need to move beyond a partial re-crafting of mainstream eating practices and toward systematic practice substitution, as well as considerations of how eating practices connect to the other practices that constitute people's everyday lives. The increased embrace of complexity that follows with a practice perspective will likely require the engagement of, as well as the cooperation between, many strategic actors, both within the food-provisioning system as well as in public policy and elsewhere. This would place considerable responsibility for systemic changes beyond the individual level and be in contrast to our interviewees' emphasis on consumer choice.
Additionally, we have not explored issues of justice, equity, and power structures in-depth, but such perspectives need also to be considered in future research and interventions related to mainstreaming sustainable food consumption. The importance of this point is made clear by how a practice perspective highlights the intrinsically sociocultural and material nature of practices of eating. With these considerations in mind, practice-based interventions and strategies could help spread sustainable eating practices beyond a relatively small number of "ecoaware" and financially well-off consumers and begin to move these routines toward the mainstream.