Participatory local governance and cultural practices in Thailand

Abstract How do cultural practices influence the process of participatory governance within local administrative structures? We address this question by reflecting cultural dimensions, such as collectivism, uncertainty avoidance, and power distance within a contextual sensitive “cultural sensemaking” model. The modelled context refers to community development meetings held in Thai communities. Here people gather to discuss development plans, which are later finalized at the Tambon Administrative Organization (TAO). We hypothesize that cultural practices influence action by giving the people the rules for action. The analysis of culture in the context of these meetings is an interpretive one and involves qualitative observation of and structured interviews. Our results show that the prevailing cultural practices do not promote truly transparent and open discussions as ascribed to participatory governance tools. Consequently, local leaders employ a subtle paternalistic leadership style. Yet, cultural practices change towards a more open and participatory rural society.


ABOUT THE AUTHORS
The research group authoring this paper has a long standing working tradition with regard to social science issues relating to socio-economic development in rural regions of developing and transition countries. Other cutting-edge topics are rural financial market development and social capital as substitute for classical productive factors, such as human and financial capital. The research reported in this paper relates to the project "Institutions in transition-Challenges for new modes of rural governance in Thailand", which was financially supported by the German Science Foundation (DFG).

PUBLIC INTEREST STATEMENT
The majority of poor people worldwide, around 70%, live in rural regions and are employed in small-scale farming. Many scholars and policy makers assume that the poor know best, which public policy measures improve their livelihoods. This calls for giving them a voice when it comes to selecting public policy measures at the local level; this is known as participatory governance. We assume, however, that cultural practices influence the interaction between the people and the public leaders. To find out more about this interaction, we analyze community development meetings in rural Thailand. Here ordinary villagers and public leaders gather to discuss development plans. We find that the prevailing cultural practices still hinder transparent and open public decision making. Because ordinary participants appreciate a conflict-free and harmonious communication practice, local leaders still display a paternalistic leadership style, although in a subtle way.

Introduction
Institutions are considered to be the means by which a society overcomes collective action problems. While endogenous cultural institutions 1 affect governance and economic development at the macro level (de Jong, 2011, p. 525), they are also crucial for forming the process of civic participation and are therefore highly relevant for processes of micro level governance (Edelenbos, 2005, p. 115). Endogenous institutions are sanctioned by society, usually without state assistance. Exogenous institutions comprise the formal rules and regulations sanctioned by the state. Governance structures are organizational solutions for effectively implementing exogenous institutions and accommodating endogenous institutions (e.g. Ostrom, 1990). By doing so, governance structures ought to comprise various mechanisms, e.g. for conflict resolution, information sharing, monitoring, sanctioning, and participation. Therefore, the analysis of participatory governance has to take account of the exogenous institutions in which people are embedded, but also need to pay attention to cultural practices (Mercer, 2002, p. 103) which, as Theesfeld, Schleyer, and Aznar (2010, p. 382) point out, may countervail exogenous rules or policies (designed at other administrative levels). With regard to culture, we follow Geertz (1973, p. 44) who emphasizes that cultural practices are best seen "[…] not as complexes of concrete behaviour patterns […], but as a set of control mechanisms […] for the governing of behaviour". Following this line of thinking, we are interested in the meaning that cultural practices give to local governance structures (Geertz, 1973, p. 311). Subsequently, culture influences action by shaping a repertoire of practices from which people construct "strategies of action" or "practices" in interaction with discourse (Swidler, 1986(Swidler, , p. 273, 2005. Cultural practices thus provide the ends towards which actors (individual and collective) direct their action (Swidler, 1995, p. 83). The analysis of culture in the context of discourse and practice is an interpretive one in search of meaning.
Thailand offers a particularly good case for studying the interaction of cultural practices and the seemingly participatory administrative rural development planning procedures, which were established as part of Thailand's decentralization policy. There exists a rich body of literature on cultural practices that covers areas from business administration (e.g. Begley & Tan, 2001), through psychology (e.g. Komin, 1990a) to gender dynamics (e.g. Bowie, 2008a) to name just a few. A great deal of research has also been done around Thai community culture. For a summary on this issue we refer to Johnson (2001, pp. 115-119). However, within the broad area of participatory research in Thailand (or elsewhere), cultural practices are acknowledged but have rarely been investigated in detail (a recent exception is Nuttavuthisit, Jindahra, and Prasarnphanich (2014)). This encourages our effort to shed more light on how certain cultural practices in interaction with discourse may influence the process of participatory governance within local administrative structures. 2 Focusing the attention on trans-personal practices in interaction with discourse gives the study of culture an empirical object. Our empirical research focuses on the Thai administrative unit called Tambon, consisting of 10-15 villages. Within the Tambon, Tambon Administration Organizations (TAOs) were created during the decentralization process in the 1990s. TAOs are the administrative level where local participation can take place. Each village in a Tambon elects two TAO representatives. The president of the TAO is directly voted in by all residents of the Tambon. Apart from elections, the major entry point for villager participation is the creation of the community development plan as the major outcome of the community development meeting (Chaowarat, 2010, p. 106). 3 Therefore, we tied our empirical work to this presumably participatory planning process to study the influence of cultural practices on deliberative governance mechanisms. 4 We concentrate our attention on the creation of the community development plan and analyze the backward and forward linkages of Thai cultural practices and participatory governance tools (i.e. public forums) that rely on discourse. Mosse (2001, p. 19) suggests that public forums in particular entail the danger of silencing marginalized groups. Moreover, the nature of group dynamics suggests that the power usually lies in the hands of the most articulate, thus normally reinforcing the status of local elites (Hailey, 2001, p. 94). These critical issues are widely acknowledged (in Thailand and elsewhere) (Gaventa & Barrett, 2012, p. 2,405).
Cultural practices are embedded in the civic (national) culture of Thailand. Present day dominant civic culture is based on the perception that Thailand is the land of the central Thais, with a single religion, Theravada Buddhism, under the Chakri royal dynasty. 5 Obviously, the resulting cultural practices in Thailand are complex and, depending on the context (i.e. meaning), lead to differing outcomes. In order to cut down on complexity to some extent, we align our thought experiment about how cultural practices affect discourse, i.e. participatory governance within local administrative structures, to the research framework developed by Hofstede (2001). The framework distinguishes five cultural dimensions: (1) collectivism, (2) uncertainty avoidance, (3) power distance, (4) masculinity/femininity, and (5) long-term/short-term orientation. These cultural dimensions are widely applied in business and academia (Triandis, 2001, p. 907). Nevertheless, they have also attracted criticism; for a summary we refer to Webster (2007, pp. 65-68). We account for the criticism by combining the cultural dimensions with a "cultural sensemaking model" developed by Osland and Bird (2000) to frame our analytical context. By indexing the cultural context, we can more genuinely inquire about the interaction of cultural practices and local governance, which is important for the identification and explanation of behaviour that is seemingly paradoxical. As outlined above, cultural practices influence the participatory agency within local governance structures. However, evidence suggests that participation in governance structures is strongly influenced by a combination of (1) collectivism, (2) uncertainty avoidance, and (3) power distance dimensions (Gudykunst et al., 1996, p. 510;Hofstede, 2001, p. 180;Triandis & Suh, 2002, p. 143;Webster, 2007, p. 76). Therefore, we focus on these three cultural dimensions within the context of our research, the presumably participatory governance within community development meetings.

Research objective and hypothesis
Our research objective is thus to illustrate (1) the role of cultural practices with regard to the making of the community development plan during community development meetings (i.e. context) and (2) their influence on elite capture in the village. We are particularly interested in the meaning that cultural practices give these local governance structures and in the way in which leaders and ordinary people differ in their cultural resources and the skill with which they deploy these (Bourdieu, 1977, p. 82 ff;Swidler, 1995, p. 31). We would like to recall that the civic culture propagated by the conservative Thai elite emphasizes a certain type of behaviour, which is explored in a variety of activities and spheres such as education, the media, usage of public space, and economic public policy. Nobody will challenge the notion that the spread of civic culture propaganda is uneven in the sense that it dominates in the greater Bangkok region. Nevertheless, its tentacles reach-to some degree-out to the whole nation; for instance, via the education system where teaching of civic and moral values plays an important role (Pitiyanuwat & Sujiva, 2000, p. 92). "Moral values" in this context refers to the values of the elite such as respect for authority, seniority, and filial piety (Mulder, 1996, p. 77, 122). The "moral values" taught promote submissiveness and deference to elite rule. Jongudomkarn and Camfield (2006, p. 500) also point out that obedience to parents is still a key practice (even of young people). Elites have an interest and the ability to uphold such cultural practices, thereby perpetuating their dominant position (Jakimow, 2013, p. 499).
Consequently, it is plausible to assume that cultural practices are as important for participation as the explicit design of decision-making rules for a public policy tool. This leads us to the central hypothesis of this article: besides the explicit design of exogenous governance structures, popular participation in public governance arenas (such as TAOs) is guided by cultural practices, specifically three cultural dimensions: (1) collectivism, which subordinates individual participation to conflictavoiding majority choices; (2) uncertainty avoidance, which shifts decision-making away from ordinary to prominent individuals; (3) power distance, which inhibits individual participation in the presence of individuals of superior status (for a deeper discussion see Section 2).

Research region and survey design
Our research area lies in north-eastern Thailand (a region called Isan). The Isan consists of nineteen provinces and hosts approximately one-third of the Thai population. The empirical work took place in the central province, Khon Kaen, in four Tambons in 2011/12. All our interview partners belonged to a population group known as Thai-Lao. The survey design followed a mixed method qualitative approach, based on observations, semi-structured interviews, and standardized interviews, resulting in an interpretive analysis.
As mentioned earlier, we focused on TAOs, the most widespread decentralized administrative bodies, and thus the local governance arenas, in which local deliberation and participation ought to take place. In 2009, there were 5,767 TAOs in Thailand. The Khon Kaen province hosted 158 TAOs, on average eight to ten TAOs per district. We worked in three districts: the Nom Pong, the Sri Chompoo and the Manchachiri districts. We purposely selected four TAOs based on their different administrative performances (which was important for our research project but not particularly relevant for the topic at hand). Basic characteristics of the four TAOs are shown in Table 1. The livelihood conditions in all four Tambons are very similar and mainly based on smallholder agriculture with a focus on paddy rice and sugarcane production. Moreover, many households receive remittances from relatives in Bangkok or abroad.
The field work was based on repeated visits over a period of two years. We worked with Thai research assistants from the Thai-Lao Ethnicity to build up confidence (most of the time interviews were conducted in the local Thai-Lao language). First, we conducted about 60 semi-structured interviews and group discussions with farmers and key persons, i.e. village leaders and TAO executives, plus numerous subsequent telephone interviews for cross checking and validating data. Based on these results, we designed and conducted two structured surveys. (1) The first survey comprised 50 village leaders, which corresponds to one leader per village, except for three villages which we could not reach. (2) The second survey covered 104 ordinary villagers, who were randomly selected in 26 out of the 53 villages belonging to the respective TAOs. The survey contained questions regarding the mode of participation and deliberation within community development meetings. We were interested in their subjective information relating to their own experiences. Overall, the questions were formulated in such a way that respondents could give subjective information about their own experiences. Box 1 displays a number of relevant questions in this regard.

Cultural practices and participatory governance in Thailand
A better understanding of cultural practices is of the utmost importance for improving the prognostics of institutional change; for our case study, this change relates to the level of participatory involvement in the local development planning process. More specifically, cultural practices are multidimensional and context sensitive. "Context" in the first instance means the face-to-face situation (e.g. a community development meeting), where issues (e.g. development plans) are debated and decided. In such a setting, the dynamics of the meeting itself can give cultural practices a coherent, systematic influence, even when participants are ambivalent. Context can also mean a more general situation, e.g. of crisis or politics (see Swidler (1995, p. 35)). Therefore, we start out with the heuristic device of sophisticated cultural labeling ((1) collectivity, (2) uncertainty avoidance, and (3) power distance) before evaluating the behavior in the context of the community development meetings of Thai TAOs.

Collectivism
The philosophical concept of methodological individualism is important for the better understanding of why individuals presumably make irrational decisions-also within collectives. The accepted view is that social phenomena can only be understood by examining how they result from the motivations and actions of individuals. A collective, e.g. the "society", the "state", or the "people", is not the same as an "individual actor" and thus does not act independently of the individuals constituting the collective. Consequently, political economics perceives "collective actions" as being determined by the choices of individuals in the collective. Osland and Bird (2000, p. 69) add to this and explain that the dimension of individualism-collectivism has to be supplemented by the personality dimension labeled idiocentrism versus allocentrism. Idiocentric people, those who derive more utility from paying attention to their own needs than the needs of others, can also be found in presumably collective cultures (and similarly for the line of arguing regarding allocentric persons and individualistic cultures). The choice of action of these individuals depends on the utility they generate from individual versus collective welfare generation. Collectivism is then the degree to which people prefer to subordinate their individual choice to the choice of the majority (Hofstede, 2001, p. 225;Triandis, 1995, p. 2).

Box 1: Participation and deliberation within community development meetings
Questions to ordinary villagers, participating in the community development meetings comprised among others: • Who can join the community development plan meeting?
• Do you usually attend the community plan meeting, and if not, why not?
• Are there any persons in the village who never take part in the community plan meeting, and if yes, what kind of persons and why do they not take part?
• Who is proposing projects at the meeting?
• Have you ever proposed a project?
• Are TAO officials participating in the meetings?
• What do villager leaders or TAO officials say or do if they do not agree with a proposed project?
• What do village leaders or TAO officials say or do if villagers still want to vote for that project?
• Are there people in the village who feel shy or do not speak at village meetings or in front of government officials?
• Are you or other persons satisfied with these procedures?
Many but not all aspects of Thai society reflect collectivist culture. 6 Change is mediated through the eyes of the group or one's peers (Vallance, 1999, p. 92) and people seem to be more willing to uphold community norms, draw signals on appropriate behaviour from authorities, and seek to blend into the assigned collective (Begley & Tan, 2001, p. 549). This may be the reason why Intachakra (2012, p. 622) speculates that many Thai people opt for an "anticipatory" discourse style, trying to anticipate what others think and how they may react, as opposed to a "participatory" style, where importance is given to the individual's contribution on a more or less equal footing with others. 7 Along the same lines, people may emphasize social harmony within the group and try to avert conflict (Hallinger & Kantamara, 2001, p. 394). Consequently, there may be a preference for interpersonal interactions that are smooth and conflict-free. A "surface harmony" must be maintained (Komin, 1990a, p. 691, 692;Mulder, 1996, p. 30, 64, 179), especially when a relatively powerless group is involved which cannot afford open defiance (Scott, 1989, p. 5). This can lead people to tend to one side of a bipolar behavioural cultural dimension, e.g. obedience and listening. It is plausible to assume that the poor are especially in need of maintaining harmony as they are the ones most likely to depend on help from others. However once the relationship of power changes, the relative value of a cultural practice may shift too, and covert discontent (hidden transcripts of anger) may turn quickly into overt discontent (Scott, 1989, p. 22, 27, 28, 30).
In this context, saving face is said to be an essential characteristic of conflict settlement in many close-knit village societies (Abraham & Platteau, 2004, p. 214). Whether Thai villages still host closeknit societies is debatable, but face-saving is still a prominent feature. In this context, the cultural norm of "Krengjai" in regulating communication and interpersonal interactions is not yet extinct in Thai society. Krengjai allows for a harmonious resolution of differing opinions. Involved parties soften opinions, restrain emotions, and refrain from strong direct criticism (Panpothong & Phakdeephasook, 2014, p. 103;Roongrengsuke & Chansuthus, 1998, p. 185).

Uncertainty avoidance
In economics, context-specific rational behaviour implies that individuals aim to maximize utility but are bound by constraints. Income is the best known constraint. However, lack of information about the future is probably the next important constraint because it causes strategic uncertainty.
Uncertainty avoidance reflects the extent to which a society establishes formal and informal rules and procedures. It includes the belief that experts, as representatives of the elite, may be better equipped to make collectively beneficial decisions through their knowledge. In societies with high uncertainty avoidance, such as parts of Thailand (Hallinger & Kantamara, 2001, p. 397;Hofstede, 2001, p. 148, 171), people may be more inclined to follow the dominant views, which are most likely the views of the local elite (or experts), assuming they know best. This type of agency is also promoted by the dominant Thai elite in its definition of a good Thai; namely one who respects the three pillars of the country: monarchy, religion and nation (Connors, 2007, p. 80). As pointed out by Roncoli, Orlove, Kabugo, and Waiswa (2011, p. 135), in a situational context where community affiliation and consensual decision-making are highly valued, people often feel the pressure to rally round the preferences of prominent individuals. This may even culminate in patron-client relationships where loyalty, obedience, and gratitude from the client to his patron are perceived as essential (Vichit-Vadakan, 2011, p. 85). Subsequently, only a few people at the top are supposed to make decisions and take risks (Holmes & Tangtongtavy, 1997, p. 84;Thanasankit, 2002, p. 132). Swidler (1995, p. 31) states that cultural practices and power are fundamentally linked, i.e. people differ in their cultural resources and the skill with which they deploy these. Social context plays an important role here. Furthermore, cultural practices shape individuals' knowledge of how others, e.g. elite groups, will interpret their actions (Swidler, 1995, p. 39).

Power distance
The cultural practices embedded in the civic culture propagated by Thailand's elite seem to promote submissiveness and deference to elite rule (von Feigenblatt, 2009, p. 592). They lay the foundation for power distance. Although to varying degrees and dependent on the context in which the cultural practice is considered, this observation supports the statement of Mulder (1996, p. 106) and Bechstedt (2002, p. 241) that Thai society is made up of positions which are hierarchically connected and people are supposed to adapt their behaviour accordingly, due to implicit social pressure. 8 Power distance refers to the extent to which the less powerful members of hierarchical structures expect and accept that power is distributed unequally and that the presumed elite has more privileges and power (de Jong, 2011, p. 526;Hofstede, 2001, p. 98;Triandis, 1999, p. 130). It is evident that Thailand's traditional elite is aware of the fact that culture has an important influence over the control of resources, i.e. human capital and political influence. This is why the elite's view of the dominant civic culture is propagated in the way it is. Furthermore, power distance describes the degree of freedom in decision-making that a superior gives to his/her subordinates (van Oudenhoven, 2001, p. 91). Not surprisingly, Thailand is often perceived as a high power distance culture (Hallinger & Kantamara, 2001, p. 391). As pointed out above, children are made aware of their position within society and are usually discouraged from raising contradicting views or challenging those of superior status. In this context, superiority can be attributed to socio-economic status, position, title, or age (Mulder, 1996, p. 108;Roongrengsuke & Chansuthus, 1998, pp. 172-173). In the traditional view, superiors ought to be obeyed simply because of their status. Since Thailand's civic culture encourages a paternalistic style of leadership which is also quite visible in the moral values taught in school which promote deference to the elite (von Feigenblatt, 2009, p. 592), many people are still reluctant to openly question authority (Hallinger & Kantamara, 2001, p. 391). Moreover, Thai people often feel inhibited and restricted in their self-expression in the presence of people with superior status (Mulder, 1996, p. 107). Finally, in high power distance cultures, information normally follows hierarchical channels (Hofstede, 2001, p. 108), resulting in asymmetric information. Thus, any kind of information can be regarded as power, or as comparative advantage (Jirachiefpattana, 1997, on-line).

Thai society and conflict
Thailand is described as one of the most politically active and socioeconomic vibrant countries in Asia and there is a well-documented history of peasant struggle and civil disobedience. See, for instance, Haberkorn (2011) on the Farmers' Federation of Thailand or Missingham (2003) on the Assembly of the Poor, or the long list of peasant resistance against big infrastructure projects. These political conflicts and the culture of not engaging in public debate and open conflict can create what Osland and Bird (2000) term a cultural paradox. Swidler (1995, p. 35) explains that certain contexts, particularly those that are important in social movements, give culture a coherent organization that it normally lacks in the minds of most actors. She continues by saying that "this accounts for some of the difficulty in trying to pin down just where and why culture makes a difference in social action To deal with this issue, we restrict our analysis to the contextual behaviour in community development meetings and place it in a "cultural sensemaking model". The proposed cultural dimensions serve as heuristic, but flexible, structures that guide agency, but not necessarily in a cause-reaction type of chain, but rather dependent on context. Subsequently, the same cultural dimension, e.g. collectivism, which is working towards group harmony (see our discussion of collectivism above) also regulates relationships with outsiders. When people of collectivist societies interact with out-group members, the former are usually indifferent and if the two groups have incompatible goals, they can become hostile. Moreover, once the in-group takes action against an out-group called for by ingroup authorities, people are likely to become aggressive (Triandis, 2000, p. 150). Thus, the cultural paradox is somewhat attenuated. Therefore, we would like to point out that in the context of community development meetings, people acknowledge the existence of conflict and also its importance in negotiating different interests. Still, the actual task is often delegated to local leaders, thus the more powerless groups such as ordinary farmers avoid that a possible conflict is detected (Scott, 1989, p. 34). For instance, discussions with farmers revealed that they value leaders who are capable of addressing conflicts. But the ordinary community member may also take action. As pointed out by Bechstedt (2002, p. 249, 250), deference to people higher up in the social hierarchy often ends when farmers' interests and ethical standards are affected beyond a certain extent; a cultural trait known as value trumping (Osland & Bird, 2000, p. 69). In addition, a superior is expected to meet his/her obligations towards his/her inferiors. Thus, a broad variety of sentiments ranging from fear, high respect and pride, to ignorance, disrespect and even disparagement can be observed simultaneously.

Cultural practices under change
Institutions are well-established, stable sets of rules backed by sanctions. Thus, they create obdurate structures that are both constraints and opportunities for individuals. Cultural practices are created around rules, i.e. institutions. Individuals can then come to act in culturally uniform ways because they develop common scripts in response to the institutions they confront (Swidler, 1995, p. 36, 38). Yet, the cultural practices of people from different social strata (and/or in different contexts) result in a different organization of their overall patterns of behaviour (Swidler, 1986, p. 275). With reference to the New Institutionalism in sociology, this could be interpreted as context-bound rational behaviour, presuming purposive action, albeit under conditions of incomplete information and costly transactions (Nee, 1998, p. 1ff, 4). 9 Transaction costs here pertain to the principal-agent relationship, and the asymmetric information issue together with the problem of trust, which is also present in hierarchical structures with pronounced power distance (such as TAOs and the community development meetings). Furthermore, the introduction of TAOs with their executive committees can be considered to be an exogenous institutional innovation, sanctioned by the Thai government, thereby supporting more appropriate local public investments through participatory deliberation of the stakeholders and therefore attaining more efficient local economic development. This change in formal institutional structure takes place in light of the existing cultural practices in the Tambons.
Although the theory of institutions suggests that cultural practices (if understood as endogenous institutions) are persistent over a longer term, change is present too, for instance due to societal adaptation to socio-economic development (e.g. the introduction of TAOs) or contact between cultures (Schwartz, 2006, p. 139). 10 Murphy, Mujtaba, Manyak, Sungkhawan, and Greenwood (2010, p. 558, 562) speculate that, although Thailand can still be seen as a primarily collectivistic culture, gradual change is taking place towards accepting cultural practices that are defined as both individualistic and collectivistic. 11 This may be particularly true for the younger generation. For instance, for them, the cultural practice of obedience seems to play a lesser role than for the older generations. McCann, Honeycutt, and Keaton (2010, p. 169) show that young college students from Bangkok favour an equalized form of collectivism (for instance, as in an Israeli Kibbutz) over a strictly hierarchical one. In general, the people of contemporary Thailand are becoming more educated and societal values are undergoing shifts, which may contribute to changes in cultural practices (Sriussadaporn-Charoenngam & Jablin, 1999, p. 412;Yukongdi, 2010, p. 178).
In the past, the North-Easterners for instance, have often been described as apolitical, passively fatalistic, conformist and exhibiting a high tolerance for capricious superiors. However, many aspects of this portrayal have recently come under fire (Streckfuss, 2012, p. 313, 324). Young working migrants from the north-east-in contrast to their parents-are no longer very familiar with farming (Rigg, Salamanca, & Parnwell, 2012, p. 1,475). Furthermore, Walker (2012, p. 18, 84) claims that the economic diversification of farmers has reconfigured old patron-client ties. Moreover, rural-urban social interactions and the spread of information through modern communication technologies challenge existing cultural practices among the rural population (Keyes, 1992, p. 177). In the long run, this can further accelerate cultural change in rural areas (Hirsch, 2009, p. 128). This goes handin-hand with the poor and rural population becoming more aware of being a relatively decisive part of the electorate. Even the cultural hierarchies in north-east Thailand are no longer as strict as before. Phatharathananunth (2016, p. 514) gives the example that, in the past, parents accepted chastisement imposed on their children by teachers, believing that teachers were superiors. Nowadays, if teachers' actions cause injuries to pupils, many parents may even consider suing them. This change also seems to affect local government officials. For instance, starting in the early 1990s, bureaucratic practices and discourses brought about status and power to officials and devalued local culture and participation (Missingham, 1997, p. 161). However, the research of Sinclair et al. (2013, p. 62) on participatory irrigation management shows that attitudes and the behaviour of many government officials towards farmers seem to have changed. This is also supported by ethnographic research by Sopranzetti (2012, p. 367) whose interview respondents mentioned that their relationships with government bureaucrats has changed from that of a solicitor to that of a client where bureaucrats are supposed to refer to citizens with respect and deference.

Local development planning and thai cultural practices-results
Since their creation in the 1990s, the TAOs are supposed to strengthen the role of rural people in decision-making and policy formulation within local government. To aid this, various administrative and policy tasks have been transferred to the TAOs: (1) local and community planning and development; (2) promotion of local economic development, investment, employment, trade, and tourism; (3) provision of local public services, such as education, primary health care, housing, arts and culture; and (4) promotion of democratic values, civil rights, public participation, law and order, and conflict resolution (Krueathep, 2004, p. 219). As a means of achieving this and as an end in itself, every year a Tambon development plan is set up by the TAO in each Tambon through consultation with farmers in a supposedly participatory way. Although community development plans are now drawn up in all of our four TAOs (see Table 1), they are not mandatory.
The Department of Local Administration, which is part of the Ministry of Interior, has set out formal guidelines for the implementation of Tambon development plans. Procedural guidelines as to how farmers' needs are to be determined are not specified by the regulation (phase one in Figure 1). The community development plan is the main input for the Tambon development plan. The community development plan is outlined either in a single or in two consecutive meetings, whose contents and procedures are almost identical. Formally, farmers should propose projects and then rank them by public vote. If there are two community development meetings scheduled, one meeting is held without TAO officials. The community development plan is a window of opportunity, during which direct consultation between TAO officials and farmers is possible. Nevertheless, the final decision about which projects to implement in the villages is made by TAO officials and certain TAO committees. These decision-making committees consist partly of elected or delegated farmers.
The conceptualization of culture as practices in interaction with discourse facilitates the empirical research with regard to (1) the role of cultural practices during community development meetings (i.e. context) and (2) possible elite capture in these meetings. Therefore, we start our thought experiment following the heuristic of cultural dimensions of Hofstede (2001) seeking to mirror it with local empirical informed context. This is useful in the initial stages of making sense of complex cultural practices. Then we narrow our analysis to the specific context of the community development meetings (see Figure 2). In this context, we first make attributions before describing the cultural script (Osland & Bird, 2000, p. 70ff). This procedure is based on the belief that cultural practices influence the process of governance within local administrative structures.

Collectivism within community development meetings
A common recommendation of participatory deliberation methods is to create homogenous subgroups. This recommendation is based on the assumption that people of the same social status or sex could then discuss more intensively and freely with each other and, thus, a more diverse range of opinions and ideas could be collected (see, e.g. Slocum, 2005, p. 82). However, the majority of farmers (78%) interviewed said that they would prefer a larger general community development meeting to elaborate the development plan instead of several smaller meetings in subgroups.
The reasoning behind this preference is as follows. (1) In a larger meeting, information appears to be spread more evenly. Moreover, in smaller meetings, farmers perceive a risk that arguments could arise among in-between subgroups because of the in-/out-group phenomena described above.
(2) It is important to observe other people during community development meetings to see their reactions (as stated by 83% of the farmers), particularly with regard to one's own comments and choices (see Swidler (1995, p. 39) on the interpretation of cultural practices). Consequently, the majority of farmers (80%) preferred open acclamation by raising their hands rather than a secret vote which would point to individuals having a more allocentric personality (refers to attributions in Figure 2).
The preference for open acclamation comes, however, at the costs of differing opinions, marginalized groups and more timid persons. For instance, we observed a situation where a poor woman was mocked by the village head for her project choice. One could interpret this mocking as a strong sign of disagreement and a way of influencing her. (3) The farmers also pointed out that these community development meetings are not the place to discuss controversial issues. Most farmers emphasized that arguments rarely arise during these meetings (refers to cultural script in Figure 2). If this is the case, then farmers are very irritated. In one village, farmers claimed that they no longer attended community development meetings because the village leaders (such as village head, vice head, village committee members or village elders) argued too much. This can be interpreted as a form of exit. Farmers usually informally complain after the meeting or-if the disagreement grows too big-refuse to take part in village activities and thus escape the confrontation altogether. Besides avoiding disharmony it is important to meld with the group. In this sense one also has to interpret the projects favoured by farmers. Over 90% of the community development projects voted for and implemented referred to public goods addressing the needs of a large part of the community, such as road construction projects. In fact, around 60% of the farmers who participated in the community development meetings stated that they voted for projects which improved the quality of life for the community as a whole and not just a small elite group.

Uncertainty avoidance within community development meetings
High uncertainty avoidance can drive people to follow the dominant views, which are most likely the views of the local elite. Thus, not surprisingly, three-fourths of the farmers stated that, in general, they would rather vote the same way as the majority and a slightly smaller share (64%) the same way as their village leaders. However, people do not always act consistently. As already mentioned above, over 60% of the farmers in community development meetings claimed that they prefer to vote for  projects that potentially benefit all farmers. Only 10% followed the majority vote and 2% said they voted for the ideas of village leaders (differing outcomes with regard to cultural script, see Figure 2).
In about a third of the villages, village leaders made a pre-selection of the projects. Village leaders met before the community development meeting and discussed which projects should be promoted. These projects were then presented to the farmers. Moreover, the formal hierarchy is maintained in the communication procedures and also favoured by farmers. Very few villagers (only around 1%) agreed with the statement that ordinary farmers should spontaneously start talking about possible project ideas in the community development meeting. The most preferred strategy of the villagers is that the leaders or, more specifically, the village head, should start presenting project ideas. But when we asked who should actually propose the project ideas to be included in the plan, about 40% of the farmers declared that this should be done by ordinary farmers. Again, we see this difference between espoused (what people say) vs. manifested cultural practices (what people do) leading to differing outcomes within one cultural script. Also, a large proportion of people (74%) pointed out that they would dare to propose a project themselves and about half of these stated that they had made a substantive comment (in at least one meeting in the past). It may come as a surprise that the majority of farmers (almost 90%) declared that in a community development meeting in the case of an unclear project idea, they would ask the person proposing the project for clarification independent of the respective power distance. Our participatory observations during the meetings, however, seem to support the perception that farmers (in contrast to village elites) hardly make any substantive comments (see attributions in Figure 2). We could interpret this observation such that farmers are aware of their civic rights. On the one hand, they may still find it somewhat intimidating to seize their opportunities pointing to a so-called unresolved cultural issue, in the sense that there are contradictions between the individual liberal and the egalitarian position. On the other hand, this behaviour can be a reaction to a situation in which the farmers do not find it necessary to engage in "value trumping" (the recognition that in specific contexts certain cultural practices take precedence over others) because the projects discussed may be in line with their general preferences. This finding is also in line with the cultural script of making a difference between real vs. espoused values (Osland & Bird, 2000).

Power distance within community development meetings
As pointed out earlier, the power distance between people also guides communicative behaviour in Thailand. However, despite the presumed high power distance between farmers and local government officials, there are two notable observations. First, almost all farmers (over 96%) in our survey Notes: A cultural script is a pattern of social interaction that is characteristic of a particular group. These patterns are accepted and appropriate ways of behaving in certain patterns of interaction. These patterns reduce the uncertainty with regard to the response from others (Osland & Bird, 2000, p. 71). Source: Adapted according to Osland and Bird (2000, pp. 70-72).
stated that during the community development meetings, representatives of government agencies (mostly TAO officials) should be present. The reasons given for this preference usually revolved around receiving information (which was otherwise not available or less reliable) on the implementation process of the village projects, or letting the TAO officials know about local concerns and possible project ideas. This argument could represent an espoused cultural value. Second, a greater number of farmers were expected to join the meeting if government officials were present. Some farmers stated that they felt proud when officials visited their village, and almost all farmers (99%) felt encouraged by the presence of TAO or government officials. Moreover, our semi-structured interviews revealed that TAO officials today are much more approachable than they were about 10-15 years ago. Nowadays, they visit the village regularly (at least once a month). In doing so opportunities to create casual contacts are established. It was mentioned by many farmers that these casual contacts are very important for creating a good relationship between ordinary farmers and TAO officials. But despite the overall positive attitude towards TAO officials, there is a clear preference among the farmers with regard to the discourse with TAO officials. (1) If a TAO official has to be present during a meeting, slightly more than 60% of the farmers would opt for just one meeting. Moreover most farmers preferred having only a few TAO officials present (instead of many). (2) Almost 70% of the farmers preferred to express their opinion to a TAO official in a meeting (see Figure 2 for flux of interaction) rather than on an individual basis.

Cultural practices and village elite capture
As stated earlier, cultural practices are created around rules, thus governing behaviour. As such, they can facilitate elite domination and, in a worst case scenario, can result in village elite capture of implemented projects. Elite capture is a "societal bad", which is to be avoided by participatory governing structures. It therefore deserves specific attention in our analytical context; the committee development meetings.
In our sample, almost 70% of the farmers stated that their leaders strongly influenced how community development projects were ranked which is the primary outcome of the community development meeting (see Figure 1 and cultural script in Figure 2). Furthermore, around the same proportion of the farmers explained that projects were always proposed by the same clique. Although, as we stated above, most farmers do not vote for the projects of village leaders directly, they will end up doing so because the selection and presentation process is heavily influenced by the leaders. Moreover, in most villages, the leaders try to guide ordinary farmers through the planning process, e.g. by way of outspoken recommendations.
We also observed rather bold means to influence the ranking of projects, e.g. village leaders actually changed the list of projects after farmers had decided on them in a public vote. When we compared farmer satisfaction levels across community development meetings with and without village leader interference, we found significant differences. The satisfaction levels of farmers were significantly lower when village leaders strongly interfered in the planning process, such as by deleting projects from the plan without farmers' consent, (9.06 and 7.94, respectively) 12 and in the case of deleting projects from the list, villager participation in community development meetings was reported to have sharply declined. Interestingly, in villages where TAO officials did not participate in community development meetings, village leaders significantly more often pre-selected projects or strongly interfered in the selection process (75% compared to 36%). 13 These observations would imply a danger of village elite capture. Yet our semi-structured interviews provided little evidence that village elites captured benefits from community development projects and only 13% of the farmers had recently observed elite capture. All cultures express a preference for ideal behaviour in the sense of how people should act (Osland & Bird, 2000, p. 70) (see also Figure 2). Nevertheless, in the end, the village elites might simply exploit their potential agency following given cultural practices within the certain political structure, without aiming at personal benefits.

Discussion
In the following, the empirical findings are discussed following the heuristic of cultural dimensions, viewed through a cultural sensemaking lens to account for the specific context, i.e. ordinary farmers and local leaders deliberating development plans in a community development meeting.

Collectivism
In the context of community development meetings, the ordinary participants appear to appreciate a conflict-free and harmonious communication practice. This form of communication practice is, on the one hand, propagated by the elite through various channels (see above) and, on the other hand, signals a preference for harmonious meetings of the collective. Interestingly, the participants of the meetings prefer to communicate in the context of larger assemblies, i.e. community development meetings. This is because smaller groups, although recommended in participation deliberation, may create an insider-outsider effect. This can create disharmony and is therefore disregarded by farmers. However, our findings from analyzing community development meetings in four TAOs suggest that farmers favour conflict free meetings and that those meetings are indeed conflict free most of the time. This is in contrast to observations made by Walker (2012, p. 133, 158, 161, 163) in one northern Thai village, despite the fact that his empirical research took place around the same time as ours. He describes lively discussions and public disputes in village meetings. Rather than ignoring this piece of the picture, we would like to interpret it as one of the cultural practices that appear at first sight paradoxical due to the tendency of observers to confuse individual and collective cultural practices by ignoring the personality dimension labelled idiocentrism vs. allocentrism (see above). These two opposing observations may simply reveal the natural diversity within any society. Furthermore, they could also mean that the topics of the village developments projects in our sample were of lesser general interest, that the particular setup of the meeting inhibited controversial discussions, or that there were not many contentious issues to be discussed. However, Walker (2012) also points out that many disputes were generated by the tendency of community development projects to develop into private ventures, which straddle the line between private and communal benefit. In our case almost all selected projects could be attributed as public goods such as infrastructure projects, thus benefiting the collective. Thus, even the choice of projects could be counted as a conflict avoidance approach.
In more collectivist societies, individuals also seek social information about the background or context of others through nonverbal communication (Kapoor, Hughes, Baldwin, & Blue, 2003, p. 685). Thus, the preference of farmers for open acclamation may be linked to social control and to the collectivist character of Thai society. In a secret ballot, farmers would no longer be able to judge what the village leaders or the majority would prefer (this practice would not result in a signalling effect to others). It would make it easier, however, for more timid farmers to choose their true preference without being influenced or intimidated by co-attendants of the meeting.

Uncertainty avoidance
The discrepancy between how farmers "should" vote and how they "actually" vote may indicate that uncertainty avoidance (in this case depicted by "following the ideas of the leaders") seems to be still in place but only influences the agency of the actual vote when one value takes precedence over another. Subsequently the agency of farmers differs from what farmers claim; there is a difference between real vs. espoused cultural practices (Osland & Bird, 2000, p. 69). This presumably cultural paradox of "value trumping" could be interpreted as a sign of an ongoing institutional change. 14 The pre-selection of projects by village leaders can be, on the one hand, understood as a measure to ensure consensus between village elites and farmers because the voting takes place on project options which have already been approved by village elites and thus enhance harmony. On the other hand, the pre-selection of certain projects also reduces the likelihood of other projects being proposed by the farmers and can be labelled as a form of cultural exclusion (see e.g. Peterson, 2011, p. 105).
The few controversial discussions we observed in meetings may not only be a sign of self-censoring but could also be interpreted as an attempt at face saving. In this sense, asking questions is often limited because farmers do not want to display their presumable lack of understanding, since that may cause embarrassment. This further entrenches the role of key persons (experts) since they are not forced by the public to communicate their knowledge in a comprehensible way.

Power distance
TAO officials have been able to lower the power distance to farmers through the establishment of informal contacts during which displays of power or criticism can be interpreted as advice among friends. However, the hierarchical distance is still substantial and reveals the conflict between the desire for an undistorted way of communication and the wish to be in contact with government officials. This dilemma might be resolved by the compromise of having only a few TAO officials present during the community development meetings -which, in fact, is what most farmers wanted. Another way of accommodating power distance between farmers and TAO officials would be to allow for one community development meeting with officials and one without. This procedure is already recommended by the TAO administrations in our sample. Nevertheless, the decision on whether to have one or two meetings is taken by the village head. Most village heads choose to have a single meeting because this is more convenient to them timewise. Only a few village heads decided to have two meetings and a small number of them did so because they had an unresolved issue in the village and they did not want to show their disagreement in front of TAO officials.
Without wanting to generalize, it is plausible to state that power differences still exist in Thai rural contexts. This also influences the initially surprising preference for bigger group meetings instead of several smaller meetings. On the one hand, in smaller meetings participants are less afraid of losing face (Chompunth, 2011, p. 356). On the other hand, a higher number of participants can mitigate the power difference between farmers and officials and, thus, give more room to deliberation (Rubin, 1973, p. 432). Moreover, in bigger meetings the information is shared more equally among all participants.

Village elite capture 15
Regardless of the efforts by TAO officials to facilitate open deliberation and information exchange, these efforts are sometimes impeded by village leaders who act as information gate keepers. As pointed out by Mansuri and Rao (2012, p. 77) it is essential to understand the role of local elites and to differentiate between elite control, which can contribute to effective participation at the local level, clientelism, and capture. The dominance of village leaders in and outside of community development meetings is based on their voiced perception that ordinary farmers alone cannot develop a village. In summary, while guidance seems to be welcomed, a strong paternalistic leadership is more and more frequently rejected by farmers, which may also be interpreted as an indication of gradual changes in the cultural dimension of power distance.
Two institutional designs counteracted the power of village elites during community development meetings. (1) TAO representatives are emerging as a new class of leaders in the village. Thus, political power is spread more widely (Badenoch, 2006, p. 88).
(2) In a more recent reform, TAO officials are now urged by higher administrative bodies (such as the Department of Local Administration) to reach out into the villages, in particular by being present at community development meetings. The presence of a higher authority seems to constrain the village elites in a non-confrontational way.
We mentioned above that holding two community development meetings may positively affect the participation of farmers by paying heed to the power distance to TAO officials. However, internal village hierarchies and power distances may counteract this. Thus, it seems more appropriate to call for only one meeting together with officials. On the one hand, the presence of TAO officials may silence some farmers but, on the other hand, may constrain village elite power. Many village leaders mentioned that one reason why TAO officials nowadays join the community development meetings is to control village elites and to ensure that the projects on the development plans are based on farmers' needs and not on the preferences of village elites. So in the end, decentralization seems to be loosening the power grip of village elites. Yet we conclude that, while meetings were still dominated by village elites, evidence for widespread elite capture could not be found.

Conclusions
In light of Thailand's contested current civic culture, we use the cultural dimensions of Hofstede (2001) as a heuristic analytical device with regard to the effect of cultural practices on the deliberation within local governance structures. Local governance is put into context, applying the cultural sensemaking model of Osland and Bird (2000) to the community development meetings at the TAO level. By doing this, we acknowledge that cultural practices interact with discourse guide behaviour, yet the outcome is context sensitive and cultural practices are just one part of the set of identities available to an individual in determining agency (Geertz, 1973, p. 44;Swidler, 2005, p. 84ff).
The present setup of the community development meetings as part of Thailand's past decentralization efforts appears to be strongly based on idealized concepts of participatory governance. However, prevailing cultural practices make it difficult to create an environment catering for truly transparent and open discussions as ascribed to participatory and deliberation tools. The participatory ideal that open discourse can lead to consensus is unrealistic in a context such as a community development meeting where cultural practices such as harmony and silent conflict trump espoused cultural practices.
Nevertheless, our results point in the direction of endogenous institutional change. Although a paternalistic leadership style is still prevalent in rural environments, it has to be conducted in rather subtle ways and open self-displays of power are disapproved of by the rural collective. Thus, there may be a change in cultural practices towards a more open and participatory rural society. However, most farmers seem to be quite content with their current modest form of participatory involvement in rural governing structures. Therefore, it may be essential to initiate changes in the setup of the meetings to shield marginalized farmers from the influence of others (e.g. introduce secret acclamation during community development meetings). But when those changes are forced upon people from the top, people might simply stop attending the meetings. In line with Scott (1989, p. 5) and Kirsch (2004, p. 55ff), exiting the collective may depict one form of resistance by a relatively powerless group if open vocal deviance is not possible, keeping in mind that any form of resistance (exit or voice) can entail negative consequences. This shows how important it is that governance mechanisms account for the prevailing cultural practices in the respective context. In contrast to the governance bodies of TAOs, which are said to be dominated by rural elites, elite capture of TAO projects at the village level does not seem to be an issue. We found no evidence for widespread elite capture during the community development plan making procedures. In summary, farmers do have possibilities to influence the development of their village through deliberation in local governance structures such as community development meetings-even if they do not have the final say in the decision and even if cultural values may, to some degree, impede participation. Notes 1. Institutions are defined following North (1990, p. 3) as the formal and informal rules that facilitate co-ordination among people by helping them form expectations. They function as constraints that shape human interaction and the enforcement characteristics of these constraints. Swidler (1995, p. 36) points out that individuals create culture around their rules. They then can come to act in culturally uniform ways, i.e. cultural practices, not because individuals share the same cultural ideas or values (see Weber (1946, p. 280) and Parsons (1951, p. 7), respectively), but because they act by the same rules. It is in this sense that the more general term "endogenous cultural institution" is used here. Nevertheless, for the sake of readability, we will use the term cultural practices in the following. 2. While Thailand has a relatively high level of linguistic and religious homogeneity it is not culturally homogenous, not the least due to the many ethnic groups (Streckfuss, 2012). Thus, Thai culture is diverse and has its regional specifics (Jory, 1999, p. 339). 3. Political participation can be exercised, for instance, directly by way of elections, which have been discussed in detail by Callahan (2005) and Bowie (2008b), or indirectly by way of civil society groups or social movements. The latter pathways of participation have been discussed in detail elsewhere, e.g. by Walker (2008, p. 86).

Other forms of institutionalized participation exist in
Thailand, e.g. public hearings on environmentally sensitive projects (Unger & Siroros, 2011, p. 211-212) and participatory irrigation management (Sinclair, Kumnerdpet, & Moyer, 2013, p. 58). 5. The Thai government goes to great efforts to embed this understanding of civic national culture in all religious and ethnic Thai groups. 6. Thailand is also sometimes described as a loosely structured society; first mentioned by Embree (1950). In loose cultures deviation from norms are often tolerated (Triandis & Suh, 2002, p. 139). However, the description of Thailand as a loosely structured society has been controversially debated throughout the decades (see e.g. Potter (1976, p. 1ff, 147ff), Komin (1990b, p. 5, 8, 11, 71, 190), Mulder (1996, p. 61, 76, 178), or Johnson (2001). 7. This is in line with Swidler's (1995, p. 35) statement that cultural practices have a powerful influence because the individual knows beforehand how others will interpret his/her action. 8. The patron-client framework has been frequently used for the analysis of Thai society (Bechstedt, 2002, p. 242). However, the patron-client relationship can be seen as a special form of power distance, thus this dyadic concept has limits when investigating community meetings. Kirsch (2004, p. 55 ff) says that social pressure is present when an individual is forced to act in a way he/she would not normally choose or when the general action is in line with the individual's choice but certain features collide with the individual's preferences. 9. In the individualistic tradition of the New Institutionalism in sociology, individual action following cultural practices is considered rational, but this rationality can take various forms as a function of the context. 10. Swidler (1986Swidler ( , p. 283, 1995) points out that culture influences action differently at some moments than at others, which is again linked to the specific contexts in which culture is brought to bear. This is why it is difficult to try to pin down just where and why culture makes a difference in collective action. 11. The "community culture school" of thought (see e.g. Chatthip Nartsupha, 1991) points out that all change should and must evolve from the village as villages have the local knowledge. In that sense, any intervention, e.g. through regulations that are framing community village meetings, would be detrimental and unsustainable. However, this school of thought assumes that the village is a homogeneous egalitarian entity, thus negating any power differences. This is quite unrealistic. Negating such differences would imply the acceptance of cemented power structures. 12. Two-sample t-test with equal variances significant at the (p) = 0.01 level. 13. χ 2 -test significant at (p) = 0.05. However, the number of village meetings without TAO staff participation was rather small (16% of villages in the sample). 14. In contrast, a survey of the Asia Foundation showed that only about 17% (total Thai sample) and 25% (for the North-East region) of respondents stated that they would follow village leaders' recommendations on whom to elect (Meisburger, 2009, p. 89). Obviously, elections and community development projects are different contexts, which might explain such differing data. 15. TAOs are known for their notorious connections to local construction bigwigs and can be easily corrupted if transparency and participation of the people is lacking (see e.g. McCargo, 2008, p. 83 or Nelson, 2002. Newspaper reports have pointed out that the decentralization of decision-making has become the source of more corruption and abuse of power. The number of corruption-related complaints against local governance organizations is significantly higher than against other public agencies. However, this is not surprising given the sheer number of local administrative bodies. Furthermore, it could also simply mean that corruption higher up in the bureaucracies is better concealed or that local administrations are now more transparent than before, meaning that any wrongdoing would be revealed more easily. Finally, there is no proof that local corruption is more prevalent than at the national level. Hence, despite the dangers of elite capture and the corruption of TAO officials, TAOs seem to have become a better mediator between the central government and villagers than the monopolized bureaucracy of the past (Arghiros, 2001, p. 273, 277).