Theoretical suggestions for future research on constructive resistance: strategies of representation of the Japanese civil society

ABSTRACT Currently, studies on collectively organized and confrontational resistance tend to dominate the field of resistance studies. In a different vein, this paper argues that the most powerful practice of dissent works as a form of constructive resistance that produce a slow transformation of values. The paper discusses strategies used by civil society organizations in Japan in order to establish alternative truths and thereby mobilize against poverty and pesticides. In the paper it is revealed how knowledge, which previously has remained invisible to mainstream perspectives, is more easily diffused if understood to represent ‘the real’ and, by this, provoke emotions.


Introduction
Studies of resistance have gone through different stages; an early focus on the more obvious and dramatic forms of resistance, have later been supplemented by a recognition of more subtle and diffused articulations. In spite of this development, studies on public, collectively organized and confrontational (sometimes violent) forms of resistance tend to dominate the field of resistance studies . As a response to this, this paper argues that the most powerful practices of dissent '(. . .) work in discursive ways, that is, by engendering a slow transformation of values' (Bleiker 2000, p. 276).
Challenging, negotiating or replacing dominant discourses could be embraced as a form of constructive resistance, a concept that provide a unique vantage point, considering that: '[d]iscourses are powerful forms of domination. They frame the parameters of thinking processes. They shape political and social interactions' (Bleiker 2000: 277; for a discussion on constructive resistance see Vinthagen 2005, Lilja and Stellan 2007, Lilja and Vinthagen 2014, Sørensen 2016, Koefoed 2017. Probing discursive change, the study of resistance moves into new territories; often mundane and unrecognized domains (14, Bleiker 2000, cf Foucault 1990. In this paper, this means scrutinising how some Japanese civil society organisations use specific strategies of narration and apply specific representations in order to construct truths that make their members act upon poverty and pesticides. The strategies of democracy advocates) in Tokyo, Japan in 2013 and 2014. 1 Among the civil society actors who were interviewed, this paper particularly draws upon interviews with APLA (Alternative People's Linkage in Asia), which has its headquartered in Tokyo. 2 APLA aims to encourage self-reliant local communities that are based on agriculture and fishing (APLA 2014). Among other things, APLA assists farmer co-operatives in their attempts to lessen the impact and power of bigger companies that tend to pay low salaries and use a lot of pesticides. Their efforts to support sugar cane plantation workers, micro-farmers, and the co-operation for regional independence movement, can be seen as a political struggle against the exploitation of local farmers (APLA 2014).
The respondents addressed different issues during the interviews, such as how they can mobilise Japanese citizens and members of their organisations to become politically active or contribute financially to the organisations. In regard to the latter, one APLA respondent discussed and analysed, in great detail, Yoshiyuki Tsurumi's (1982) widely sold, short study, Banana to Nihonjin [Bananas and the Japanese], which provides a very intimate reflection of Japan's relationship with its region, through an exploration of the rise of mass consumption in Japan and how it is linked to the massive expansion of plantation agriculture on the island of Mindanao in the Philippines. Tsurumi's imaginative and innovative writing approach, which he theorised as 'thinking while walking' (arukinagara kangaeru), implies: 'that research is a physical activity carried out by the body as well as the mind: it engages all the senses of sight, hearing, touch, taste, and smell, as well as faculties of reasoning and reflection' (Morris-Suzuki 2011), p. 136).
In the sections below, the respondents' reading of Tsurumi's book, as well as the impact it has had on the Japanese civil society, is discussed and analyzed. Actual meetings with precarious farmers, movie clips and the everyday resistance of social media posts is also dealt with in the analytical sections. These and other discussions illustrate the different strategies of representation that are used by some Japanese civil society organisations to produce knowledge about poverty in order to mobilise the Japanese population to act upon material and social inequalities. But before moving on to addressing different strategies of representation, I will provide an overview of some key insights on the interlinkages between knowledge and resistance suggested in the power-resistance literature.

Knowledge as power and knowledge as resistance
In The History of Sexuality (Foucault 1990), Michel Foucault emphasises how resistance appears as discursive, creative and small-scaled occurrences when power and knowledge are joined together in discourse, yet it has the ability to create social change. Political struggles and subversive acts, then, occur as micro-complexities. It is scattered signs that, when re-appearing, can be thought of as accumulated resistance. A single act of resistance, which to a great extent is interwoven with power discourses, might be hidden and negligible, but when amassingfor example, when resistance inspires other acts of resistanceit might lead to social modifications and transformations (Foucault 1990, pp. 96-101, Baaz, Lilja andVinthagen 2017).
Resistance could then be played out through discursive strategies that negotiate or build alternative discourses. It could be about repeating things differently or talking from new venues (Lilja and Vinthagen 2018). Or, as Foucault discusses in Power/knowledge (Foucault 1980), discursive resistance could revolve around ongoing battles between competing discourses. Foucault also discusses how disqualified knowledge could challenge more dominant narratives (Foucault 1980).
Overall, this paper is inspired by Foucault in the sense that knowledge-making, in different forms, is embraced as a form of (power and) resistance. This resistance is to be seen as a constructive form of resistance, which establishes new knowledge.
The concept of 'constructive resistance' denotes both individual and collective forms of resistance, which propose new truths or subject positions. Constructive resistance can be grand, but it can also be a matter of producing ongoing small-scale differences that might look trivial but sometimes have major impacts. It is resistance that targets disciplinary and biopolitical strategies or institutions, which produces and structures subjectivities, ways of life, desires and bodies. This kind of resistance destabilises, displaces or replaces established truths or and various claims to 'the real' by suggesting other ways of life, and constructing discourses, subjectivities and institutions. In this, it often promotes pluralistic views of truth and makes constructive use of the friction that the diversity of perspectives enables (cf. Vinthagen 2005, Lilja and Vinthagen 2007, Lilja and Vinthagen 2014, Sørensen 2016, Koefoed 2017, Wiksell 2020. In addition, as pinpointed by Minoo Koefoed, a more constructive form of resistance distinguishes itself from other forms of resistance by 'its particular temporality of change, where change is approached, not in linear terms as something opted for in a near or distant future, but as something which is implemented directly here and now through the resistance act itself' (Koefoed 2017).

Constructive resistance and strategies of representation
In the forthcoming sections, I further elaborate on constructive forms of resistance, which produce new discourses; in this case, around the poverty of farmers who live precarious lives. It is my intention to highlight multiple strategies of representation in order to display the diversity of forms of constructive resistance.
To uncover how truths are made, I draw on, among other things, Roland Barthes' theoretical elaboration on meaning-making. Barthes discusses how we are experiencing various representations differently through the example of photos, which is a special kind of representation that prevail in the cross-roads between the present and past. In the moment a photo is taken, the moment that is captured is simultaneously immortalised and gone forever (Barthes 1977).
Barthes addresses photos as 'a message without a code': 'a specific photograph, is never distinguished from its referent (from what it represents)' (Barthes 2000, p. 5). Let us paus to elaborate on Barthe's thinking. Language consists of a system of signs that represent objects. For example, the word 'dog' represents the notion of dogs. However, in a photograph we have a more direct relationship with the object that it represents (although the dog is only to reach through the photographic mediation of the actual dog). Barthes means that we can see/reach the object through the image. Reality, then, in this case does not use a code that represents it. Barthes argues that it is a type of representation that reaches a reality behind the form. This idea has been challenged by the breakthrough of digital photography. In addition, scholars of social semiotics, such as Kress and van Leeuwen (1996), states that the photograph itself could be comprehended as a series of codes. Still, photos have a unique status. Or as put by Sontag: What is written about a person or an event is frankly an interpretation, as are handmade visual statements, like paintings and drawings. Photographed images do not seem to be statements about the world so much as pieces of it, miniatures of reality that anyone can make or acquire. (Sontag 1977, p. 5).
Photographs are sometimes experienced as proof of material facts: 'a testament to the existence of a specific thing in a specific place at a specific time' (Rice 2016). However, as Susan Sontag (1979) has rightly argued, that we can reach the real behind/through the image does not mean that photographs are not ideology-free proof of material facts.
While we often read the photographic image as a captured reality, when taking a picture, a reduction or change 'in proportion, perspective and colour' is, according to Barthes, taking place. This means that the photographic image is not real, although it is still experienced as analogous to reality (Rice 2016).
Even though this paper focuses on the impact and strategies of descriptions, Facebook posts, movie clips and factual meetings rather than photographs Barthes' theories are still relevant analysing the different truths that are generated by civil society actors in Japan. Barthes, among others, elaborates on representations and our experiences of comprehending 'the real'. Here it is important to pinpoint that we never fully experience the real. First of all, there is always a gap between the description of an object and the object (Edkins 1999, p. 99). Moreover, there are non-symbolised remains of the real that exist outside what we experience as the real (Žižek 1989(Žižek , 2000(Žižek , p. 132, Edkins 1999. As elaborated by Sum and Jessop (2013), complexity is reduced in sense-making and meaning-making and: 'sense and meaningmaking not only reduce complexity for actors (and observers) but also give meaning to the world' (Sum and Jessop 2013, p. 3). Thus, even when we think that we experience the reality, there is a gap between our interpretations of the real and the real.
At the same time, somewhat paradoxically, it is not possible to separate between the representation and the real. There is no clear border between socially produced understandings of, for example, a tree and the matter of the tree in the moment we experience and read it. In addition, we must acknowledge the processual character of material 'things'. This means that the real that Barthes addresses is not the real in the sense that it is objective facts corresponding to an absolute real, but rather what is experienced as the real. As we will see below, when members of the Japanese civil society are visiting villages in the Philippines and Cambodia, the visitors experience that they understand the reality of these people, while overlooking the fact that their reading of their meeting has been coloured by their own interpretative categories and frameworks.
Moreover, the knowledge-making of the civil society around farmers in Philippines and Cambodia, is not only dependent upon images and factual meetings with these farmers, but also includes descriptions of these lives. As I argue below, descriptions could be understood as a form of constructive resistance. According to, Marcus et al. (2016), academics may not always have a clear view of what a 'description' is or what it does, but they know that they do not like it. Descriptions are considered to be: 'Boring and static, rote rather than creative, reconstructive rather than constructive ' (Marcus, Love and Best 2016). Still, when producing counter-narratives, descriptions can, as I will argue, be important components that mobilise emotions and questions, and rupture silences in regard to, for example, poverty.
The forthcoming sections elaborate on the example of Japanese civil society organisations and how their way of representing poverty seems to invoke dynamic, affective, and emotional processes that are evoked through a 'reality effect'. Emotions will be addressed as an individual experiences that are embedded in, and emerging from, social processes. Emotions play an important role when mobilising people for political reasons.
In the text below, I address different ways in which strategies of representation can be played out as constructive resistance. This resistance is performed on behalf of and/or in solidarity with subjects in precarious positions and is known as proxy resistance. Proxy resistance does not always challenge power; but sometimes provokes and thereby simultaneously strengthens the power that is being challenged (Foucault 1990, p. 96;Baaz, Lilja andVinthagen 2017, Baaz, Lilja, Schulz andVinthagen 2017). Images of poverty easily (re)construct stereotypical comprehensions of the 'other'.

Strategies of representation as resistance: the case of Japan
Of the organisations that were interviewed in Tokyo in 2013 1nd 2014, several of them discussed the difficulties that they experienced when seeking to mobilise people into political action. People were suggested to be 'short sighted' and live in the 'here-andnow', without any considerations for the future. This section will discuss how these organisations use different strategies of representation, as a form of 'doing' politics, to mobilize the population to take action in regard to, amongs others, poverty. 3

Emotional descriptions as constructive resistance
The respondents who were interviewed pinpointed that rich descriptions are particularly effective for producing knowledge in regard to precarity. This is interesting as descriptions are not usually assigned status but, in line with Amartya Sen, many scholars feel that: 'It is fair to say that description as an intellectual activity is typically not regarded as very challenging. To characterise a work in the social sciences as "purely descriptive" would not normally be regarded as high praise.' (Sen 1980). Still, one of the respondents, when drawing on Tsurumi's 'banana book', illustrated how descriptions have the ability to generate emotions and engender more radical attempts to 'do something'. She said: (. . .) in the 80s here in Japan, a book about bananas and the Japanese people was published. The book described how the banana is a huge problem in the Philippines and in the big plantations. I don't know how many copies of the book they sold, a lot of people come to, when they cooked, feel worried about the bananas. But at the same time there are no organic bananas in the supermarkets in Japan.(. . .) In the book, there were so many stories in detail and until that time we know nothing about the bananas. People don't know about the agriculture business, which will get big money from that. (Interview with APLA, Tokyo 2013) The above quotation displays the lack of knowledge around the cultivation of bananas in Japan before the book by Tsurumi was published in 1982. The respondent simply states that 'until that time we knew nothing about the bananas.' No particular mechanisms of suppressing knowledge are revealed in the quotation; rather, there is simply no information to access about the cultivation of bananas, the farming of bananas, the farmers, poverty and the pesticides. In the context of this, the respondent presents the details of the stories of Tsurumi's book as an eye-opener, or a kind of discursive rupture.
In Tsurumi's book, descriptions prevail as a strategy of constructing new knowledge around the precarious lives that are lived by banana farmers. The descriptions in the book evoked interest and made the readers attentive to the new information. According to Marcus et al. (2016), describing something as a meaning-generating activity makes things intelligible and turns new phenomena or practices into objects of investigation and knowledge. Thus, descriptions 'connects us to othersto those described, to the makers of what we describe, to other describers' (Marcus et al. 2016). How descriptions give an insight into the emotions of others is revealed in the interview with the representative of APLA: I think sometimes, you know often very difficult words are used to explain the situation about world and situation about the economy. But he [Tsurumi] with very easy word described how he himself walked in the heat. And he described his meeting with the people. So maybe it's the very, very first time we really knew . . . So I think there are many people who are moved by the kind of that research. (Interview with APLA, 2013 in Tokyo).
Barthes argues in his essay, 'The Reality Effect ' (1989) that 'non-functional' descriptions and insignificant details, which do not add to the plot or tighten the structure of the text, are still important as they make a story feel real and create a 'reality effect'. The detailed texts of Tsurumi, in line with this, seem to have evoked feelings of 'having been there' for his readership. The descriptions within Banana to Nihonjin touched people who were able to emotionally imagine what it is like to cultivate bananas. The detailed descriptions gave people insights into the emotions of others and provide them with the possibility to temporarily switch subject positions (Marcus et al. 2016). The descriptions in the bookdifferent words piled upon each other to form sentencesrepresent different objects and practices such as starvation, systematic poisoning and hard work. However, as we interpret these objects and practices, they in turn represent, mediate, or are entangled in different emotions such as fear and grief, and give the reader the possibility to understand, read and even experience these emotions. Emotions and interpretations appear as inseparable. As expressed by the respondent: 'So maybe it's very, very first time we really knew . . . So I think there are many people who are moved by the kind of that research'. The emotional 'surplus' of the book about bananas seems crucial for people to feel motivated to act politically, and to 'do' politics through different organisations.
The emotional impact of the simple descriptions of walking 'in the heat' might also be due to the descriptions' capacity to evoke images in the minds of the readers. Concrete representations, such as 'heat' and 'walk', create mental pictures more easily than more abstract nouns, such as justice (Tornborg 2014). Mental images presumably strengthen the impact of speech-acts on the part of the reader, which also evoke emotions, raise public attention, establish counter-narratives and shape political subjectivities .
Resistance is a heterogeneous phenomenon, which sometimes involves establishing knowledge in the context of power. The representative practices used by Tsurumi are to be seen as a particular kind of producing knowledge. Tsurumi's descriptions offer a specific way of displaying precarious experiences or what Foucault calls subjugated knowledge. To write the book about bananas and thereby reveal the precariaty of the banana farmers can be seen as a form of proxy resistance given that it is resistance carried out in solidarity with others. It is also a form of constructive resistance, as it aims to establish meaning around the practices of farming and bananas, thereby mobilising actions around these practices. It is not primarily resistance against something, but rather it builds new discourses around banana cultivations.
Still, as is displayed in the next section, descriptions are not always enough for Japanese civil society members (or potential members) who want further proof of the pesticides, starving children and poverty. Descriptions are experienced as just descriptions; thus, they are not read as anchored in any real evidence of existence. This is a problem with descriptionthe descriptions cannot be proved. There must be a tremendous amount of trust placed in the describer in order for the descriptions to be relied on, as the description is neither proved nor confirmed (by representations that are read as corresponding with the real). This is contrary to, for example, photographs that are, as stated above, sometimes experienced as proof of material facts (Rice 2016).

The real without a code as an engine for political subjectivities
As indicated above, Tsurumi's book has been highly embraced by the Japanese population (Morris-Suzuki 2011). According to APLA, however, Japanese civil society members are not generally persuaded by descriptions that provide an experience of a language that consists of a system of signs that represent, although emotionally, practices of farming. Instead, civil society activists prefer a direct relationship with the objects that are represented. When reviewing the interviews, I found that several of our respondents worked to facilitate real-life experiences of poverty in order to mobilise support for the organisations' aid interventions. Trips were arranged to different areas, for the civil society members (or potensial members) themselves to see the farmers, who experienced poverty. How can these trips be understood in terms of meaning-making and constructive resistance?
Barthes' theoretical ideas above can shed some light upon civil society activists in Japan, who desire a direct relationship with those who experience poverty. Unsatisfied with descriptions, members of the Japanese civil society want to experience what they consider the reality without a 'code representing it'. They want representations, which reach the real beyond images and descriptions, and that are understood as a proof of material facts. One APLA employee stated in regard to Tsurumi's book: So, those people really want to step forward to directly meet the farmers; not only meeting them in the book. The author of the book visited the Philippines, like Mindanao, and worked in the communities and then made a map of the area. So, I think the readers of the book also, you know that . . . people wanted to meet those people.
APLA has also arranged study trips to Negros Island in the Philippines when farmers at the sugar plantations were facing a starvation: So, they visited there, civil society members or NGO members, and they discovered those kids, who were facing hunger. And they've helped those people, the farmers, and they donated money because they really knew the real situation.
The quotations above displays the link between knowing and acting, in the following way: 'they donated money because they really knew the real situation'. According to the respondent, the feeling of 'really knowing the real situation' occurred after the civil society members had met the farmers themselves; that is, when they experienced that they had encountered the reality 'without a code'.
To get the farmers represented through descriptions does not appear, to the civil society members, as an evidence for the existence of those farmers. According to this logic, the knowledge one acquires through the interaction with the materiality of the farms is different from the one gained through descriptions. The knowledge and entangled emotions that emerge from study trips give rise to political action, and the study trips emerge as a form of constructive resistance.
Interviews carried out with JANIC, confirmed the narrative of APLA. JANIC also experienced difficulties when trying to mobilise people to fight poverty. According to the respondents from JANIC, many of their clientele, or potential clientele, do not trust their representations of the real to be the real, but demand real-life experiences in, for instance, Cambodia in order to embrace notions of poverty and inequality. JANIC explained this scepticism against imposed interpretations as specific to the Japanese people: (. . .) people here, they don't trust the . . . they are interested in what we do but they want to confirm whether what we are reporting to them is real or not by themselves. So they like to take part in the so-called study tours. Tours because they visit to the field and (. . .) Interviewer What is a study tour?
Interviewee: The study tour is the Japanese way of studying about the realities in Cambodia and of the organizations who work in Cambodia. They hold the study tour for a group of 20 students, who are young girls, who are you know people between their 20 s and 30 s. They form a group and they visit. And they speak with you know the people that is the Cambodians and also the people who work there. And also they expose themselves to the reality that those people are living and they sum up . . . of course the degree of reaction is different but they bring back in their own definitions of assisting others and then they may continue to be aware of the issue and then they may become a donor to that specific organization.(. . .) Or, they might be inspired that I want to be part of this action so I might want to change jobs. You know I want to be an NGO staff rather than you know employee at the office. And they might try to study on their own (. . .). Or they think of going to a masters study abroad or at home to gain more specific knowledge. Solve the issues, to be part of some bigger issue. (Interview with JANIC in Tokyo 2013) Overall, the interviews with JICA support the pattern outlined by APLA. Resurging counter-narratives evolve in direct contact with what is experienced as the real. The meeting with the materiality of the farmers creates a 'reality effect'. The strategies of representation that are used to create this effect can be understood as a form of resistance, which draw on what Barthes discusses as 'the endless need to authenticate the "real"' (Barthes 1989). Also, Kress and van Leeuwen 1996, p. 171) argue that one of: The crucial issues in communication is the question of the reliability of messages. Is what we see or hear true, factual, real, or is it a lie, a fiction, something outside reality? (. . .) We routinely attach more credibility to some kinds of messages than to others. The credibility of newspapers, for instance, rests on the 'knowledge' that photographs do not lie and that 'reports' are more reliable than 'stories', though since we wrote the first edition of this book the rise of Photoshop and 'spin' have begun to undermine both these types of knowledge.
Before moving on, it is important to point out that both 'the real' in Barthes terminology (i.e. the villagers or the villages) and texts describing rural areas compose representations. Representations are the building blocks of discourses through which we interpret what we see and hear. They also maintain and create the discourses surrounding the villages of Cambodia and the Philippines. The discourses form how we comprehend reality. Still, my respondents felt that there is a difference between different types of representations. One is perceived as reality 'without a code' while the other (the descriptions) is experienced as representing reality through a code. Hence, in this text, Barthes' theoretisation is used in order to separate between various representations, which are experienced differently.
As indicated above, an increasingly abundant scholarship of multimodal social semiotics has explored the 'reality effect' of specific representations. But, as argued in this paper, not only the credibility of messages are important but also how some messages provoke emotions are central in meaning-making processes. Or as expressed in one interview: When I was watching the TV of course there are many news about the world or conflicts in the world. But I don't know maybe Japanese people think oh it's very far away from here. It is not our business. But maybe when we meet the people. Like I meet a friend from the Palestine or Kosovo there (when I was) in Malmo. And I've heard their story directly. Then I was really moved or shocked and decided to . . . I must get involved and work to resolve those problems. (APLA Tokyo 2013).
The twisted time and space of the movie clips of the evening news seemingly create a distance, and the readers thereby remain emotionally untouched by the messages within them. This creates a distance and a feeling of 'I do not care'.
It is worth adding a critical reflection here; the emotional meetings with the farmers, pictured above, might be strengthened by a fascination of difference. In the recognition of the farmers' experiences, cultural diversity also becomes visible and experienced. The visitors might be struck by what they interpret as exotic differences, which again twist the discourses. When lives are embraced as 'unusual and exciting because of coming from far away', this can be seen as a form of exoticism. 4 According to Kuehn (2014) exoticism is a complex philosophical, historical and representational issue and is concerned with the perception and description of difference. What might be interpreted as 'exotic' representations rarely give the truth or reality about past or far away cultures; rather they are aesthetic understandings that are produced in a specific historical context. Thus, in the meeting between farmers and the civil society actors, resistance could both strengthen and challenge power (in the form of stereotypical comprehensions). It works to produces different twisted discourses and a surplus of meaning around precarious bodies. Thus, the 'reality effect' that is created through the meetings with subjects that are framed as precarious, possibly create a reduced or stereotyped understanding of these lives. There is always a gap between the complex real and our constructions of it. However, as previous research has pinpointed, there are variations in regard to how much we reduce and simplify when producing the real. Exoticism implies an unnecessary reduction of complexity (cf Dyer 1993).

The nodes of resistance: everyday resistance through social media
While Foucault emphasises that power relations form a dense web that permeates social institutions, it is still possible to locate 'local centres' within the production of knowledge. Among other things, Foucault displays how an entire 'watch-crew' of parents, nurses, doctors and educators together compose a 'local centre' of power-knowledge around a child (Foucault 1990).
The Tokyo interviews reveal how those who might be assumed to form this 'local centre' of power-knowledge in Japan around povertyfor example aid-workers, 'experts', media and 'governmental staff'are sometimes met with scepticism. Representatives of governing bodies and organisations are often considered to twist the 'truth' from reality, and deliver distorted characterisations that feed off or support dominant ideologies. One respondent explained: (. . .) false information spread during the time of World War II, propaganda that came from the major media and the government. So, people experience that, you know, the government information, even though they constitute the authority, may not necessarily tell the truth. So, you have to be selective. That you have to select on your own (who you trust). When they select, people tend to select the people close to them. (JANIC 2013). According to the interviews, discourses around farmers living precarious lives are most successfully spread by friends and colleagues, often via social media. The 'local centres' in the production of knowledge are composed of 'poverty tourists' and their close ones, who are utilised by the organisations to spread and build discourses about precarious lives: They visited there and after they met the people, kids there, they really determined to, you know, that they involved those movement to support. So that after they came back from the Philippines to their place, they talk, talk, talk with their friends and the movement. And then the movement grow. (APLA 2014).
My JICA respondents similarly stated that their members, after visiting areas of famine, might 'hold reporting sessions with their friends' (. . .): not a formal reporting session, but many of them will write on their Facebook or Twitter. Or maybe write essays to be submitted at their schools or offices. Yeah, so I think people will take action and be the spreading the word of what they have experienced (Interview with JANIC in Tokyo 2013).

Another respondent said:
And the second point I wanted to say is that the Japanese people trust that if . . . trust more that if your friend or your family member or your colleague or the people who are close to you rather than published announcements or reports. So, we ask them to assist us on the study tour or any event that we hold. We ask them to share what you have heard today or experienced today to others starting from people who are close to you (Interview with JANIC in Tokyo 2013).
Jennifer, Robinson (2000) pinpoints friendship as a 'noisy' form of surveillance. She states: 'Power can be understood as a mutual, although rarely equal, relationship (rather than simply a technology) in which active subjects (both "dominated" and "dominator") participate.' (Robinson 2000, p. 68). Civil society organisations in Tokyo, however, embrace relations and practices of friendship, not as power, but to resist the hegemonic knowledge production (or the lack of knowledge) around, among others, bananas and banana cultivation. The very trust in the friendship makes the resistance possible. Friendships become places where political messages such as 'do not eat the wrong bananas' or 'support local farmers not big companies' can be repeated and spread. By this, the members of the Japanese civil society, who communicate via Facebook, can be considered as practicing not only constructive resistance, but also everyday resistance. The mobilising of friends and family members individually through personal channels corresponds to Scott's outline of resistance, where the key characteristic is that it is dispersed and not played out in public (Scott 1990).
The information emanating from friends is, according to JICA, different from other forms of representations. It is information that 'personalises the experience' and it is a specific 'way of communicating'. According to one of my respondents, posters on an information board just provide facts and nothing to reflect on. But if your friends give you information you reflect more and (re)live and (re)experience the information. The trust in friends' knowledge seemingly mobilises the critical agency that is needed for resisting power-knowledge frameworks. Indeed, being able to embrace and (re)live the lives of those who experience precarious situations is a way of emotionally embracing their lives. The critical potential of friendship networks thereby becomes crucial when mobilising the civil society members in order to make them understand, embrace and act upon inequalities, poverty and pesticides. This can be seen as a form of constructive resistance that produces and (re)structures resisting subjectivities and contemporary discourses.

Concluding discussion
This paper has discussed different practices of representing farmers who are (at least during periods of starvation) living precarious lives, and the character and effectiveness of each sign that is used in the knowledge-making process.
The strategies of representation analysed in the paper can be seen as a constructive form of resistance, which aims to establish new narratives, and (re)structure subjectivities and contemporary bodies of knowledge. Overall, the analysis of the interviews with representatives of Japanese social movements and non-governmental organisations (NGOs) displayed how these organisations construct particular discourses around the question of poverty by: (1) drawing on simple and emotional descriptions; (2) representing the reality 'without a code' thereby creating an emotional 'reality effect'; and (3) utilising trusted sources of knowledgein this case friends, colleagues and family.
The respondents in this paper pinpointed that detailed descriptions were eye-openers, or a kind of discursive rupture. Descriptions gave insights into the understandings, practices and emotions of others. Still, the descriptions were experienced as giving no proof of the poverty or pesticides. Therefore, NGOs have posted study groups to different countries in order to establish a direct relationship with the object that was being represented. They want to show (what is comprehended as) the reality 'without a code' representing it. It seems that civil society members in Japan must be convinced that they have experienced the real (or trust others stating this) in order to act upon the real (poverty, starvation, banana cultivation, etc.). Creating a 'reality effect' emerges as a form of constructive resistance.
Establishing a direct experience of the situation of those who live precarious lives, and the meeting and communication with these people, can be seen as a form of proxy resistance that is carried out in solidarity with those living precarious lives. Also, movie clips of the evening news are to be seen as 'proof'; they are verified as being images of the real, but still fail to emotionally touch the audience of the TV-programme. Thereby, they also fail to mobilise people into action. Thus, according to my respondents, the effectiveness of the strategies of representation depends on, among other things, whether or not the messages (about precarious farmers) are creating a reality effect and if they evoke any emotional reactions.
After the study trips, the participants were encouraged to repeat the newly established discourses around the farmers' precarious situation within their social networks. When mobilising some forms of knowledge, the organisations took advantage of everyday channels of communication and relations of trust. In particular, friendships were identified as important platforms in the production and spreading of knowledge around, among other things, banana cultivation. This partaking in the repeating of more emancipatory narratives can be seen as a more individualised and everyday form of resistance. The individuals' experiences of organised and public forms of 'doing politics' (such as the arranged study tours), then inspired civil society members to apply more everyday forms resistance, which displays some linkages between the individuality and collectiveness of resistance. There seems to be an upscaling, rise or acceleration of the resistance where some representations (descriptions) lead to other strategies of representation, such as study trips and (thereafter) Facebook posts. Thus, the organisations provide us with an interesting example of how an escalation of resistance came about.
The above analysis suggest that, when analysing the production of knowledge as a form of constructive resistance, some of the aspects that are listed below could be relevant (cf. Norman Fairclough, Ruth Wodak and Teun van Dijk): The representation's (image, video, descriptions) ability to create a 'reality effect'. Are the representations seen as corresponding to, composing or depicting the real? Is there an illusion that there is no symbolic or material surplus? (the surpluses of matter or meaning, the existence of a code, etc.).
The trust of the readers of the meaning in the speakers. Is there trust in the speaker? What representations are considered to be proven?
The emotional impact of the representations. Are the representations evoking any emotions on the part of the reader? Are the representations forwarding any intensity of anger, pity, et cetera.?
That counter-narratives are repeated over time with new representations (messages on Facebook, in descriptions, study tours, etc.). Is this happening? Is resistance upscaled through different representations?
How individual and collective practices of resistance are feeding each other. Is the resistance upscaled over time, while appearing as both individual and collective practices where one feeds the other?
Are there are any practices/relations of proxy resistance (motivated by solidarity)? This kind of resistance might become very strong, since it creates unexpected alliances across social sections such as race or class (yet it can easily become paternalistic, selfserving, or exploitative, in which non-subalterns utilise subalterns). Together these points should be seen as building blocks for a theory of representation that addresses the production of meaning as a form of constructive resistance. Notes 1. The quotations of the interviews that are presented in this text have been edited for clarity.
This includes removing repetition and correcting grammar. I have also omitted certain fragmented passages that were difficult to comprehend. 2. The interviews were carried out by my research partner Mikael Baaz and me. 3. The analysis rests primarily upon interviews, but other sources of informationincluding scholarly texts, reports, and webpages of the organisations that are in focushave also been considered in order to get a broader picture of the resistance of civil society organisations in Japan. Overall, the approach to this study is qualitative and is designed in an inter-disciplinary way in order to capture a more in-depth understanding of civil society and resistance. 4. https://dictionary.cambridge.org/dictionary/english/exoticism.