Rethinking Alcohol and Conflict: The Politics of Alcohol in Post-war Sri Lanka and Nepal

ABSTRACT Existing research on the relationship between alcohol and conflict has focused on displaced populations and viewed alcohol largely as a driver of post-war social problems such as trauma and violence. We draw on qualitative research in Sri Lanka and Nepal to build a more complex picture of alcohol’s role in post-war societies that is attuned to its political economy dimensions and its politically symbolic role. This article develops a comparative and multi-disciplinary approach to demonstrate how narratives about alcohol can reinforce existing processes of social marginalisation and how alcohol can play an important symbolic role in post-war politics.


Introduction
What is the relationship between alcohol and the ending of war? Existing research on this question has been dominated by the field of public health and has tended to frame alcohol as a driver or indicator of war-related social problems such as trauma and violence. Public health research has focused on drawing direct causal connections between the ending of war and rising alcohol consumption, concentrating on displaced or conflict-affected populations, who may be traumatised, and where a rise in drinking may be associated a post-war growth in interpersonal violence.
In this article, we argue that alcohol plays a wider and more complex role in post-war societies than is usually acknowledged, and that to grasp this role, a multi-disciplinary approach is required, which is attentive not only to public-health related challenges associated with alcohol use, but which also considers alcohol's social, economic, and political dimensions. Such an approach is important since alcohol's impacts stretch beyond health, and because the stigmatising and politicisation of narratives surrounding alcohol use can have important real-world consequences. We Interviewees were selected based on their expertise in understanding different aspects of alcohol (its economic, social and political dimensions) from the two countries. We also sought to capture diverse perspectives on alcohol-related challenges (from civil society representatives and industry) and selected interviewees with knowledge of alcohol in different regions and communities. In Sri Lanka, we interviewed five representatives from civil society groups who worked on alcohol abuse, public health and violence, and two alcohol industry representatives. We interviewed seven academics with expertise in alcohol and with experience of living and working in all parts of the country (including direct observations from Jaffna and the Eastern Province, which we draw on below). Our findings from these interviewees are also supported by insights from the research team's extensive experience of conducting research on issues relating to poverty, development and conflict over several years across the island, including in the Northern and Eastern provinces.
In Nepal, we adopted a similar approach, conducting interviews with eight local civil society representatives and three UN representatives who were working both on community development or more specifically on the issues of alcohol and violence. We conducted six interviews with industry representatives, four with government officials, and four academics with expertise in alcohol. As noted above, our research in Nepal was complemented by a series of field observations and interviews conducted in and around Birgunj, a key site in the booming alcohol tourism trade. Here, we interviewed six small businessowners and four civil society representatives to gain insights into the political economy of alcohol in this region. As in Sri Lanka, we were able to support interview findings with insights gained through research team's extensive experience of conducting research on issues relating to poverty, development and conflict over several years across Nepal, particularly in the southern Tarai region.
The article proceeds as follows. The next section reviews existing academic literature on alcohol and conflict, noting the dominance of public health perspectives and exploring the benefits of integrating historical, political economy and anthropological perspectives. The next two sections present short case studies of Sri Lanka and Nepal. Each case examines post-war alcohol trends before examining two features that have been neglected in the existing literature on alcohol and conflict -the political economy of alcohol (drawing attention to key features of government and rebel regulation, the alcohol industry, and the politics-industry nexus) and its political instrumental role (examining how alcohol has been used in a variety of political narratives deployed by leading politicians, rebel groups, and civil society organisations). The article then provides some comparative analysis before a concluding section that draws some wider analytical and policy implications.

Understanding Alcohol and Conflict: A Review of Existing Literature
Alcohol misuse is a major global health problem. The WHO estimates that 3 million deaths (5.3 per cent of all deaths globally) were caused by harmful use of alcohol in 2016, a figure that outstrips harms associated with diabetes (2.8 per cent), HIV/AIDS (1.8 per cent) or direct physical violence (0.8 per cent) (WHO 2018, 63) and was roughly in line with the global death toll of the COVID-19 in 2020 (WHO 2021). While the WHO (2018) has found that alcohol increases inequalities both between and within countries, and findings from the general literature on addiction show that alcohol is closely associated with social marginalisation and stigma (Room 2005), there has been limited analysis of how these processes and intersections play out in post-war contexts.
Existing research on alcohol and conflict has primarily, though not exclusively, come from the field of public health and has shown that countries emerging from armed conflict experience higher rates of alcohol abuse (Ezard 2012, Roberts et al. 2012, Weaver & Roberts 2010. These studies stress that in such contexts, alcohol is used as a coping strategy in response to displacement, unemployment, trauma and other associated mental health problems (Ezard et al. 2011). Some public health studies have taken a wider view of the links between conflict and alcohol by examining the role of loose regulation, low government capacity, and efforts by alcohol companies to expand their influence into previously untapped markets (Moskalewicz and Simpura 2000, Wallace and Roberts 2013, Roberts and Ezard 2015. Recent studies in the field of public health have focused on establishing an evidence base for the links between conflict, alcohol (mis-) use and violence (Lo et al. 2017, Mootz et al. 2018, Black et al. 2019, and examining the effectiveness of prevention or treatment programmes Ezard 2015, Greene et al. 2018). In both areas, recent studies emphasise the need for stronger evidence and further research.
A small number of national-level and cross-country studies in the public health literature have highlighted the risks associated with rising alcohol consumption in post-war contexts, especially amongst youth and displaced populations living in camps (Al-Hemiary et al. 2015, Ertl et al. 2016, exploring how armed conflict may intersect with rising rates of alcohol use and different types of interpersonal violence (Okello and Hovil 2007, Sriskandarajah et al. 2015, Mootz et al. 2018. While existing work in this area has improved understanding of the links between war, trauma, displacement, and alcohol misuse amongst communities directly affected by war, it has neglected how changing patterns of alcohol consumption and related health impacts in post-war societies may be shaped by complex intersections with wider, social, political and developmental processes. Bøås et al.'s (2008) study of Sierra Leone is one of the few detailed studies of alcohol use in a post-war setting. This study challenges the dominant narrative that a growth in drug and alcohol use in post-war settings is a direct result of increased trauma or lack of economic opportunity. Instead Bøås et al. (2008) find that changing patterns of alcohol use are generally shaped by a complex combination of factors including (1) growth in warassociated migration or increased mobility, (2) new income opportunities associated with the post-war period driving growing consumption and growing interest from industry, and (3) extreme social stress. Our findings confirm the relevance of these three factors, but point to three additional factors not considered in their study: (4) shifting social norms related to the end of war, (5) a regulatory regime supportive to growth in the formal alcohol sector (a point also highlighted by Wallace and Roberts (2013)), and (6) the political instrumentalisation of alcohol in post-war settings. 2 Our study draws on the wider anthropological, historical, and political economy literatures on alcohol (Singer 1986, Douglas 1987, Akyeampong 1995, Dietler 2006 to develop a more multi-dimensional view of alcohol's role in conflict settings. Anthropological research shows that alcohol can serve as a key mechanism through which social norms and identities are conveyed, performed and policed (Dietler 2006). Another key finding from the anthropological literature is that alcohol has 'long been a prime political tool' (Dietler 2006, 237) and has often been a central feature of colonial strategies of rule (Jankowiak and Bradburd 2003), whilst also serving as a focal point for nationalist forms of political mobilisation, as Akyeampong (1995) shows in his study of Ghana and as we highlight below in the case of Sri Lanka.
While existing research focuses mainly on the needs, vulnerabilities or 'demands' for alcohol generated by marginalised communities, we argue that this perspective needs to be combined with an analysis of the 'supply' side: the drivers of alcohol use and especially the role of global and local industry players (both formal and informal) in creating demand. 3 Rather than focusing on the causal relationship between war or the conditions of postwar and alcohol (ab)use, our approach is more contextually rooted and examines these relationships alongside an understanding of the emerging social and political meanings of alcohol in post-war transition. This approach is inspired by Gamburd's (2008) anthropological study of illicit alcohol in Sri Lanka. Like Gamburd (2008), we argue that to understand alcohol's complex role, it is important to explore a variety of perspectives including its social and cultural meanings, the role of alcohol in livelihoods, the alcohol industry's role in politics, and how social norms around alcohol use vary according to caste, ethnicity, and gender. We also build on insights from other classic studies of power and commodities, such as Sidney Mintz's (1986) Sweetness and Power, which examines the growing consumption of sugar in Europe through a wide-ranging investigation of the institution of slavery, the political economy of production, the political struggles that underpinned rising consumption, and the new social meanings that became attached to sugar during this period. Mintz's (1986) work can be situated within a wider tradition of work on the anthropology of violence, which is 'committed to ethnographically embedding evidence within the historically given social and economic structures that shape life so dramatically on the edge of life and death' and focuses on tracing the transnational and historical roots of contemporary structural violence (Farmer 2004, 312;Mintz 1997). Our perspective on alcohol also chimes with similar insights from the field of critical medical anthropology, which seeks to 'understand health issues within the context of encompassing political and economic forces that pattern human relationships' (Singer et al. 1992(Singer et al. , 78, 1992. We also draw inspiration from the historical literature on alcohol such as McGirr's (2015) study of prohibition in the USA, which illustrates how strict regulation of alcohol reinforced existing efforts to marginalise and control minority groups, especially black communities in the South.
Our approach recognises the importance of narratives linking alcohol with social problems, whose importance has been acknowledged in alcohol policy research (National Research Council 1981). By examining such narratives in post-war settings, we explore how alcohol can intersect with other vulnerabilities and reinforce processes of marginalisation (Room 2005). While we stress the importance of examining the material interests that shape alcohol production, regulation and consumption, we argue that analysis of these factors is usefully complemented by a study of the shifting social and political meaning of alcohol or narratives about alcohol (mis)use and how these can be used in the mobilisation of political agendas.
This article is concerned with post-war contexts. The literature on post-war transitions has challenged the widespread assumption that the ending of war marks the end of deadly violence and shown how violence more commonly mutates rather than disappears after war (Cramer and Goodhand 2002, Wood 2008, Suhrke 2012. While organised violence may diminish with the ending of war, other forms of interpersonal, domestic or structural violence (Galtung 1969) are likely to persist or increase. The ending of war often coincides with an acceleration in economic development, which drives the rapid formalisation and monetisation of parts of the economy that had remained informal during wartime, often exposing marginalised groups (especially women) to greater violence and vulnerability in the process (Gunasekera and Nagaraj 2019).
There has been widespread critique of the teleological thrust behind the concepts of the war-to-peace or 'post-war' transition, which tend to imply a temporal break and a linear progression towards a condition of peace (Cramer andGoodhand 2002, Walton 2016). In this article, we use the term 'post-war' not to 'signal a definitive after', but rather to indicate 'the process and ambition to "move beyond"' war (Klem 2018, 244). We adopt a political economy approach which is alert to the legacies of war, without viewing these only in direct causal terms. In doing so, we demonstrate how, in addition to its own substantive impacts, alcohol serves as a useful lens through which to understand the constraints and agency, continuities and changes associated with post-war periods.
We argue that understanding alcohol's role may be improved by broadening our conception of post-war and the timeframes through which war's effects are felt, which in turn may modify our understanding of the relations between alcohol, conflict and violence. We advocate a more open understanding of the concept of 'post-war', moving away from simply seeing this as a temporal break, and instead an understanding of post-war transition as a process via which identities, practices and norms are re-articulated, providing scope for new meanings and relationships, but where this process of change is also constrained by what has come before (Klem 2018). Our analysis below illustrates how alcohol can come to symbolise new threats and anxieties associated with the ending of war and often serves as a key mechanism for the remaking of social identities or reinforcing processes of marginalisation. Most existing research on alcohol, conflict and violence has focused on direct causal links between alcohol and physical violence. Our findings highlight a more indirect and complex set of relationships and resonate more with work on structural or slow violence, which explores how harms stemming from social inequalities may be largely invisible, taking place over years and decades (Galtung 1969, Nixon 2013. Harms associated with alcohol in postwar settings may be better captured by concepts such as 'slow violence' which 'insist we take seriously forms of violence that have, over time, become unmoored from their original causes' (Davies 2019, 2).

Background to Nepal and Sri Lanka
This article examines alcohol in post-war settings through case studies of Sri Lanka and Nepal -two countries that share some common features. Both countries concluded long-running civil wars at around the same time: the Tamil separatist conflict in Sri Lanka ending in 2009, and the Maoist conflict in Nepal ending in 2006. Both conflicts were driven by grievances from marginalised communities based in peripheral regions, which have persisted in the post-war period (Goodhand et al. 2016(Goodhand et al. , 2021. In both countries, rebel groups -the Liberation Tigers of Tamil Eelam (LTTE) in Sri Lanka and the Maoists in Nepal -controlled territory during the war and prohibited or severely restricted the sale and consumption of alcohol. Sri Lanka and Nepal have roughly similar levels of alcohol consumption and both are experiencing sustained periods of rising incomes, leading to a growth in the formal alcohol sector. In both countries, these trends are linked to growing labour migration or the return of members of the diaspora, and the opening of conflict zones to the alcohol industry. South Asia is one of the few world regions where alcohol use is rising and predicted to grow over the next five years (WHO 2018). The region has the highest rate of informal alcohol consumption in the world (45.4 per cent of the total) (WHO 2018, 43). 4 The two countries also exhibit three important differences, which provide scope to explore variation in post-war trajectories and trends. First, war ended with a negotiated settlement in Nepal and with a military victory for the government in Sri Lanka. These different war endings have shaped post-war politics -leading in Nepal to a period of sustained 'political unsettlement' (Goodhand et al. 2021) characterised by frequent changes of government, and in Sri Lanka to increasingly centralised and authoritarian governance.
Second, the two countries vary considerably in their social demographics. Nepal has a much higher degree of ethnic diversity, with over 125 ethnic groups listed in the 2011 census. Sri Lanka, by contrast, has a large majority community -the Sinhalese -who make up 75 per cent of the population and have dominated the political system at the expense of the larger minority groups: Sri Lankan Tamils (11 per cent) and Muslims (10 per cent).
Third, there are important differences in patterns of alcohol consumption and regulation. Overall levels of consumption are higher in Sri Lanka (growing from 4.0 to 4.3 litres between 2010 and 2016) than in Nepal (declining slightly from 2.1 to 2.0 litres over the same period), and the formal sector plays a bigger role in Sri Lanka (63 per cent) than in Nepal (30 per cent) (WHO 2018). Regulation of alcohol is more rigorously enforced in Sri Lanka. Excise duties from alcohol constitute an important source of revenue in both countries, though are of greater importance in Sri Lanka where there is a declining tax base, and where excise constitutes a higher proportion of total revenues (28 per cent) compared with Nepal (14 per cent). As will be explored in greater detail below, alcohol is a more politically salient issue in Sri Lanka than in Nepal and is more frequently used by politicians to signal commitment to traditional values.
There are some key differences in the non-formal sector. In Nepal, most traditional forms of alcohol are produced and consumed in the home (or informal bhattis) and are socially acceptable for some ethnic groups. 5 In Sri Lanka, by contrast, most non-formal or illicit production takes place on a larger scale and is typically organised by criminal gangs, who often maintain close ties to local-level politicians (Gamburd 2008, chapter 1).
There are sharp variations in drinking patterns within both countriesbetween social groups and across sub-national regions -with consumption rates in Nepal much higher in parts of the hill and mountain regions, and in Sri Lanka in the hill country regions and the war-affected North. In Sri Lanka, drinking rates are generally much higher amongst men and heavy drinking is most common among men engaged in manual work such as farmers, fishermen, and labourers (Gamburd 2008, Hettige andParanagama 2005). Existing work has highlighted that drinking plays a key role in the construction of masculinity in Sri Lanka and can help to enhance social bonds amongst men (Gamburd 2008, Widger 2015. Narratives about alcohol are also strongly influenced by dominant Sinhala Buddhist principles which prohibit the consumption of alcohol, and by gendered and class-based social norms which view drinking amongst women and lower classes as less acceptable. As noted above, Nepal has unusually high levels of ethnic diversity and while alcohol consumption is seen as socially unacceptable for some higher caste groups, for other groups including the Newars, alcohol or raksi has important symbolic and ritual value. 6 Homemade production of alcohol is more widespread in Nepal than in other South Asian countries where informal alcohol tends to be mass produced. For many women from so-called matwali (lit: 'alcohol drinkers') caste groups, brewing homemade alcohol is an important source of income and economic independence (Manchanda 2004). Dhital et al.'s (2001) survey of 2,400 households from across Nepal, shows that a third of sampled households were producing alcohol for both consumption and sale and that alcohol constituted a major source of income in poorer households. Amongst some janajati groups, drinking alcohol is considered socially acceptable for men, women and (in some cases) children (Dhital et al. 2001). 7 Although there is some evidence that social norms around abstinence are breaking down amongst higher caste groups, a study by Adhikari et al. (2019) found that disadvantaged groups (including Dalits and janajatis) were still six times more likely to consume alcohol over the life course compared to upper-caste groups. 8 In the following case studies, we start each case by analysing trends in post-war alcohol consumption, production and regulation. The analysis highlights several important differences in consumption trends and the social meaning of alcohol between and within these two contexts, with stronger evidence for rising consumption in the conflict-affected regions of Sri Lanka, but with evidence of a national-level growth in formal consumption in both countries. This section also examines various competing accounts for these trends that emerged from our interviews and existing studies.
For each case, we then move to examine the political economy dimensions of alcohol, examining the status and composition of the alcohol industry and the politics of regulation. In Sri Lanka, higher taxes on beer were introduced in 2016, while Nepal introduced a wave of strict new regulations surrounding the sale of alcohol in 2017 and 2020. In both countries, the alcohol industry is dominated by a small number of players, tighter regulation has been closely linked to demand for tax revenues, and our research uncovered evidence of collusion between the alcohol industry and politicians.
Finally in each case study, we draw attention to what we have termed the 'political instrumental' role of alcohol in post-war societies. Alcohol is often used as a 'political object' by government, political parties, rebel organisations and civil society groups to assert a variety of moral and political agendas. As we will explore below, alcohol often functions as an important 'political object' that can be strategically deployed by politicians, rebel groups or civil society organisations to police, perform or reinforce values and norms. Alcohol also frequently serves as a 'cultural flashpoint' that crystalises conflicting worldviews and assumptions (Conway 2013).

Post-war Trends
Sri Lanka saw a nationwide increase in alcohol consumption and production after the end of the war in 2009. 9 Growth in formal production preceded the ending of the war but the rate increased thereafter, particularly in the period up to 2013, growing more gradually from that point (Dayaratne 2013). A study by Nugawela et al. (2017, 550) found a 'marked increase' in consumption in non-conflict-affected areas, a trend they suggest may be linked to a variety of factors including 'rapid socio-economic development, alcohol industry penetration and lack of alcohol control strategies during the postconflict period'.
Government revenues from alcohol between 2009 and 2013 increased 28 per cent, while consumption of formal alcohol increased 16 per cent; trends that have continued since and which Dayaratne (2013, 25) attributes this to 'the renewed liquor sales in war-affected areas as a result of the opening of the alcohol market . . . to meet the suppressed and denied demand for alcohol' (WHO 2018). 10 Illicit consumption has become more concentrated amongst older drinkers, and those living in more isolated rural areas (including conflict-affected regions such as Mannar) (WHO 2018). 11 Our analysis of district variations in alcohol consumption per capita found that the top three districts were in the conflict-affected north: Killinochchi, Jaffna, and Mannar. 12 This supports Nugawela et al.'s (2017) finding that there had been a 'notable' post-war increase in alcohol consumption in conflictaffected areas.
This rise in conflict-affected regions was explained by our interviewees with reference to both demand and supply-side factors. On the supply side, several interviewees provided anecdotal evidence about a sharp post-war rise in the availability of formal alcohol in the North. 13 Alcohol companies were amongst the first businesses to arrive after the end of the war in Jaffna (in the far North). 14 Following the war, new licences had been granted in Jaffna, with new hotels opening up and people shifting away from toddy towards beer consumption. 15 In the immediate post-war period, the percentage growth in beer sales in the Northern province reached double digits and Kadirgamar (2017, 263) has argued that in Jaffna this expansion was 'drastically reducing the demand for local toddy and arrack'. 16 This rapid expansion of the formal sector in the North can be seen as a response to pent-up demand and lack of competition resulting from the tight restrictions placed on alcohol production and consumption by the LTTE, which controlled large parts of the North and East during the war and imposed a strict moral code (Sarvanathan 2007, Terpstra andFrerks 2017).
On the demand side, Herath's (2018) study of community cohesion in the North and East, observed a sharp rise in post-war drinking driven by joblessness and trauma, alongside growing availability of alcohol. Those living in the Northern Province during the final stages of war in Sri Lanka were subjected to repeated shelling into civilian areas, with an estimated 40,000 civilians killed in this region during the spring of 2009 (United Nations 2011). Sriskandarajah et al. (2015) explain how this experience, compounded by the Asian tsunami of 2004, created widespread exposure to mass trauma. One of our informants suggested that a rise in post-war drinking in Jaffna had been partly driven by the return of members of the diaspora with more permissive attitudes to drinking, and the fact that people had more disposable income following the end of the war. These trends have meant that practices such as drinking in the family home or in public have become more widespread. 17 The shifting caste composition in Jaffna may have contributed to these changes, with the lower castes now outnumbering the traditionally more conservative vellalar community due to migration or displacement. These shifts did not go uncontested: in Jaffna there had been a sizeable youth mobilisation against alcohol: around 300 youth were mobilised with the support of a local NGO. 18 Interviewees cited a number of additional factors leading to increased consumption in the north following the end of the war including: 'a sense of hopelessness', a lack of work, and a sense of freedom prompted by the ending of tight regulations on alcohol enforced by the LTTE. As one interviewee put it, there was a general attitude that could be summed up as: 'when we are free, we think it is time to use alcohol'. 19 Others, however, argued that narratives around rapidly increased alcohol consumption following war's end should be treated with caution without stronger pre-war or wartime data. As one respondent argued, drinking 'is always couched in moral terms', and therefore reports of increased alcohol use can be used to stigmatise marginalised groups (including 'the poor', lower caste groups or those from other ethnic or religious groups). 20 Some studies on Sri Lanka have presented evidence of a link between alcohol consumption and domestic violence, which may be particularly strong in regions directly affected by conflict. Jayasuriya and Gibson (2013) conducted a study in four conflict-affected districts in the North and East to assess perceived civilian safety during the post-war period. They found that many Tamils felt less safe in the post-war period and that a growth in alcoholrelated domestic and sexual violence is a key factor driving this growing sense of insecurity linked to the growing availability of alcohol. These findings are supported by other studies (Sriskandarajah et al. 2015) and by our own analysis of the 2016 Demographic and Health Survey, which shows that rates of domestic violence generally appear to be higher in the conflict-affected regions of the North particularly in Killinochchi and Jaffna. Similar connections have also been identified in other contexts, for example in a study by Stites and Howe (2019), who describe a shift in patterns of insecurity and violence in post-war Uganda ('from the border to the bedroom'), a trend they link to the growing availability of alcohol (Stites & Howe 2019).

Political Economy Dimensions
The National Alcohol and Tobacco Act was enacted in 2006, banning any form of advertising, sponsorship or promotion of alcohol or tobacco products. The Act also established the National Authority on Tobacco and Alcohol (NATA) to advise the Government on policies aimed at eliminating tobacco and alcohol related harm. In 2016, Sri Lanka introduced the National Policy on Alcohol Control, but NATA representatives acknowledge that work to enforce alcohol-related measures have lagged far behind efforts to control tobacco (Wijewardena 2014). 21 Liquor licencing has remained tightly controlled by politicians. Most licences today are owned by leading politicians or their proxies (Dayaratne 2013). 22 New governments typically issue new licences to generate patronage. Dayaratne (2013, 43) notes, for example, that while the Rajapaksa government adopted a hostile rhetorical position on alcohol (as part of the "Mathata Thitha' ('End to Alcoholism') campaign), 377 new licences were issued between 2005 and 2011. Several interviewees provided anecdotal accounts of the connections between politicians and the alcohol industry. These links are widely known and deemed important as shown for example in a speech by the former President Sirisena, who said that commenting on these connections 'might affect the stability of his government' (Asian Tribune 2016).
Some existing studies have suggested that there is local-level political involvement in the production of illicit alcohol (Dayaratne 2013). While local and regional politicians tend to have an interest in perpetuating illicit production, national-level politicians benefit from and therefore seek to promote the expansion of the large-scale formal sector. 23 A post-war decline in illicit production can be partly attributed to improved prosecution, a claim borne out by official data. 24 One respondent claimed that alcohol companies had provided funds to the police to incentivise raids on illicit alcohol. 25 The alcohol industry has generally found it harder to gain a foothold in the north and especially Jaffna because of the strength of the toddy industry there and some respondents argued that alcohol companies tried to undermine toddy production by supporting a more complex licencing regime for toddy and gradually eroding the knowledge base around palmyrah toddy production. 26 Kadirgamar (2017) found that a decline in toddy tapper cooperatives in Jaffna was being driven by generational change and a reluctance amongst younger men to participate in toddy tapping due to social stigma. He also argued that growing post-war connectivity was undermining the market for toddy, by making beer and arrack from the South more freely available (Kadirgamar 2017).

Political Instrumental Role
Successive presidents (Chandrika Kumaratunga, Mahinda Rajapaksa, Maithripala Sirisena, and Gotabaya Rajapaksa) have introduced anti-alcohol campaigns or legislation early on in their presidencies. The National Authority on Tobacco and Alcohol (NATA) was initially established under President Kumaratunga but eventually implemented by Mahinda Rajapaksa as part of his Mathatha Thitha ('End to Alcoholism') campaign. Maithripala Sirisena made tackling alcohol and drug abuse a prominent policy theme throughout his troubled presidency. Gotabaya Rajapaksa became president in November 2019 shortly before the COVID pandemic hit and in March 2020, he announced a lockdown which included a blanket ban on the production and sale of liquor.
Some respondents argued that anti-alcohol campaigns were designed to appeal to poorer female voters who are generally supportive of efforts to restrict sale and consumption of alcohol. One informant argued that presidents often introduce these bold policies earlier in their period of office but over time become constrained by the influence of the alcohol lobby (Dayaratne 2013). 27 The topic of alcohol has long served as a tool with which Sri Lankan politicians can signal commitment to the defence of Sinhala nationalist positions and to Buddhist values in the face of corrupting western influences. The Temperance Movement was closely tied to the independence movement in Sri Lanka, with Anagarika Dharmapala launching an antialcohol campaign against British rulers in 1895. The campaign sought simultaneously to reassert Buddhist values and challenge the British colonial regime by undermining a key source of revenue (de Silva 2001). This connection between alcohol and nationalist rhetoric has remained influential as Sinhala nationalism has come to dominate the post-war political landscape (Venugopal 2018). Contemporary nationalist social movements such as Nil Balakaya (a youth organisation set up by Mahinda Rajapaksa's son Namal) have continued to mobilise around the issue of alcohol.
In 2015, a new coalition government defeated Mahinda Rajapaksa in presidential and parliamentary elections. The coalition involved a pragmatic union between the United National Party (UNP), which adopted a more western-oriented outlook, and the more nationalistic President Maithripala Sirisena, a former ally of Rajapaksa who led a breakaway faction of Rajapaksa's Sri Lanka Freedom Party (SLFP). In 2018, alcohol served as a focal point for tensions within the coalition when the UNP-led government removed a 38-year ban on selling alcohol to women and employing women in places where alcohol is sold or produced. This move proved controversial and was quickly reversed by the President Maithripala Sirisena, who reasserted his commitment to 'building a cultured society with values such as freedom, morality and democracy'. Another decision to extend opening hours of liquor shops to 10pm was also revoked by Sirisena.
Contemporary anti-alcohol campaigns both from politicians and civil society groups have adopted a socially conservative orientation, tied to the preservation of Buddhist values and stressing the corrupting influence of western practices. 28 The activation of nationalist politics can be heightened in transitional post-war moments or during periods of economic liberalisation and may reflect wider anxieties about moral decay associated with the opening of society. As Klem (2014) has shown in his study of postwar Trincomalee (a conflict-affected zone in the Eastern Province), the ending of war led to growing connectivity, which generated growing anxiety about emerging practices including drug-taking, prostitution, and drinking.
Both political and civil society campaigns against drinking have often reinforced a class-based narrative of alcohol use, which focuses on the social ills that arise from drinking amongst lower classes (neglecting problems of middle or upper-class drinking) and which have been widely discussed in studies in other post-war or transitional contexts (see e.g., Lincoln (2016) on Vietnam). One civil society group, for example, argued that alcohol and tobacco use was a primary driver of poverty in rural areas and campaigned actively on this basis. 29 This position was also voiced by President Maithripala Sirisena in a public rally in August 2016, where he said that the consumption of tobacco and alcohol is the main cause for poverty in rural Sri Lanka (News. lk 2016).
To summarise, government efforts to regulate alcohol in Sri Lanka have been consistently undermined by political interests and industry pressure. Despite the failures of regulation, politicians have frequently used alcohol as a tool for political mobilisation or demonstrating commitment to Sinhala Buddhist values. Alcohol in Sri Lanka serves as a potent tool for asserting nationalist visions of a good society, often emerging in the form of 'flashpoints' where social norms are debated in the public arena.

Post-war Trends
While overall rates of consumption are lower in Nepal than in other South Asian countries, rates of 'heavy episodic' drinking are higher than average (WHO 2018). Although there has been a downward trend in overall alcohol consumption (from 2.1 litres of pure alcohol per person in 2010 to 2.0 litres per person in 2016), there was a sharp rise in formal alcohol consumption (0.2 to 0.6 litres) and a decline in informal consumption (1.9 to 1.4 litres) over the same period (WHO 2018). As will be discussed below, while the formal sector has grown during the post-war period, this appears to have been driven primarily by wider economic transformations, changes in regulation, and cross-border factors. 30 Representatives from the alcohol industry confirmed in 2019 that alcohol sales were rising, with particularly strong growth in the Tarai (the southern plains region that borders India), the Kathmandu valley area, and in other larger cities. 31 In the Tarai, the rise is largely attributed to social and economic changes connected to migration and cultural shifts such as the emerging expectation that alcohol should be served at wedding ceremonies. 32 Respondents claimed that more affluent groups in the Tarai were shifting from homemade to formal alcohol as incomes rose. 33 The ban on alcohol consumption and sale in the state of Bihar in India in 2015, which borders parts of the Tarai, has led to an alcohol tourism boom in the major towns and villages that straddle the border, where there has been a noticeable growth in the incidence of small shops selling liquor in villages across. 34 Several key informants in Birgunj (a major town and trade crossing in Parsa district) estimated that consumption had gone up 3 or 4 times since the ban, and there has been a sharp rise in the number of liquor licences granted in the town between 2018 and 2019 (rising from 596 to 728). 35 In Kathmandu, growing consumption is associated with shifting social norms and attitudes and growth in the urban middle class (World Bank 2016). In contrast to Sri Lanka, there is only anecdotal evidence for a rise in consumption in conflictaffected regions.
There is some evidence of alcohol playing a growing role in popular perceptions of crime, violence and insecurity. Survey data from 2008 to 2011 saw a growing proportion of respondents (8 per cent to 32 per cent) identifying alcohol as a main cause of crime and violence (Racovita et al. 2013). A 2019 survey found that 'alcohol abuse' was perceived as one of the top two 'main threats to security' (identified by 21 per cent of respondents nationwide, 32 per cent in mountain regions, 22 per cent in hills and 19 per cent in the Tarai) (Thapa 2019). In Sri Lanka, we observed evidence linking conflict, mass trauma, and alcohol, especially in the northern province which experience repeated shelling during the final stages of the war. In Nepal, no populations were subjected to the kind of aerial bombardment experienced by populations in Sri Lanka's northern province. A lower proportion of the population are estimated to have been killed in the civil war (16,000 from a current population of 28 million in Nepal or 0.06 per cent of the population compared with (80-100,000 people from a current population of 22 million in Sri Lanka or roughly 0.45 per cent of the population) (BBC 2009(BBC , 2021.

Political Economy Dimensions
The government of Nepal passed the strict National Policy on Regulation and Control of Alcohol in 2017 to minimise alcohol-related harms, but implementation has been poor. The policy covered a range of measures including mandatory use of pictorial warnings, an outright ban on liquor advertising, and banning the sale of alcohol to people under 21 (Nepali Times 2018). The reasons for poor implementation include push-back from the alcohol industry, the lack of a global framework agreement for regulating alcohol policy (unlike tobacco), the fact that consumption of alcohol is accepted by some social groups, the importance of homemade production of alcohol for livelihoods, and a lack of manpower in the excise department. 36 Alcohol duties are a growing source of revenue for the government. Total alcohol-related revenues rose 125 per cent between 2014 and 2018. In 2015, the Government introduced a directive intending to safeguard and promote local alcohol industries. The directive made it mandatory for foreign liquor companies involved in production in Nepal to carry no more than six percent of the total annual profit back to their country of origin. 37 Some politicians appear to have an economic interest in the alcohol industry. In the case of one distillery in the Tarai, for example, a politician was reportedly taking a 20 per cent cut of profits. 38 One respondent discussed how efforts to develop tighter alcohol regulations in 2017 and 2018 were undermined by high-level leaders from a multi-national corporation from Singapore came to lobby against the policy. 39

Political Instrumental Role
Since the consumption of alcohol is deeply engrained in the culture of many social groups in Nepal, narratives linking alcohol with western influence are less prevalent than in Sri Lanka. During the People's War (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006), the Maoists, however, used alcohol as a focal point for political mobilisation. The All Nepalese Women's Association (Revolutionary) (ANWA(R)), the party's mass women's organisation, led a nationwide campaign to ban alcohol in the late 1990s. The anti-alcohol campaign (or raksi bandh) was particularly vigorous in the hill areas (Manchanda 2004). As Mahat (2003)has argued, the Maoists fought against all forms of oppression, including subjugation of women and the ban had the explicit goal of 'protecting women from intoxicated men'. Some authors (Mahat 2003, Fujikura 2003 describe how this move was widely appreciated by women but others, including Pettigrew and Schneiderman (2004), have argued that the campaign alienated some women from janajati groups who were economically reliant on homemade alcohol production. Several of our respondents noted that the Maoists' strict line on alcohol did not continue into the post-war period as fear of reprisals diminished and other issues were prioritised. 40 While alcohol has been less widely used as a political object in Nepal, it has occasionally become a focus for political contention. In 2020, the Supreme Court introduced measures to promote the formalisation of alcohol, justified on public health and fiscal grounds. These moves were resisted by political representatives of ethnic groups who have traditionally used alcohol. Some politicians representing these groups -such as Usha Kala Rai, the Minister for Social Development of Province 1 in Eastern Nepal -proposed branding rather than banning locally produced liquor, on the grounds that it would support livelihoods of marginalised groups, while also boosting government revenues and reducing imported liquor (Online Khabar Nepal 2021). In this instance, the politicisation of alcohol was linked to a wider politicisation of ethnicity in Nepal (Adhikari and Gellner 2016).

Comparative Analysis
Our comparison of post-war trends in Nepal and Sri Lanka shows that both countries experienced national-level growth in the consumption and production of formal alcohol. In Sri Lanka, consumption grew rapidly in some conflict-affected regions in the north. An absence of data makes it difficult to ascertain a similar effect in Nepal, though it seems likely that this has been less marked in Nepal since the Maoists wielded less powerful administrative control than the LTTE. Key drivers of this rise in Sri Lanka include social stress relating to war, unemployment, growing availability and marketing of alcohol in conflict regions, and the emergence of new social norms linked to migration. The latter factor can also be identified in Nepal, particularly in the Tarai. Comparing the two countries' experiences therefore suggests that while a rise in alcohol consumption in regions directly affected by conflict can be seen in some post-war contexts, this trend is shaped by various factors including the nature of war-time violence, the level of trauma experienced by the population, the extent to which connectivity rises in the post-war period, patterns of migration and diaspora return, and the scope and character of rebel governance, which was unusually extensive in the Sri Lankan case (Wood 2008).
The above analysis highlights three wider contextual factors, which may shape the linkages between post-war transitions and alcohol use and lead to significant sub-national variation. First, varied norms and traditions around drinking between social groups (and associated regional variations), and second, the prevalence and character of homemade or illicit production. A third contextual factor neglected in existing work, and which is not directly related to conflict, concerned the transnational or cross-border dimensions of alcohol use. In Nepal, a sharp rise in alcohol use was particularly marked in towns and villages along the southern border with India due to a ban imposed on alcohol consumption in Bihar imposed in 2015. Sri Lanka lacked these cross-border dimensions, but transnational influences appeared to be enhanced during the post-war period in certain regions due to the growing influence of returning family members from the Tamil diaspora.
Our analysis of the political economy dimensions and political instrumental role of alcohol in Nepal and Sri Lanka has highlighted three key points. First, on the political economy dimensions, the effectiveness of government efforts to regulate alcohol in both countries were driven by both governments' heavy reliance on alcohol-related tax revenues but constrained by a powerful domestic and international alcohol lobby with close ties to the political class. There was evidence of the alcohol industry targeting conflict-affected regions in Sri Lanka but this dynamic was less evident in Nepal.
Second, the political instrumentalization of alcohol by politicians is common to both countries but varies considerably. In Sri Lanka, the political potency of alcohol was rooted in Sinhala nationalist discourse, which emerged out of Sri Lanka's colonial experience, and was particularly amplified in the post-war period. Moral anxieties about the ending of war and growing connectivity also increased the salience of these narratives. In Nepal, by contrast, social norms surrounding alcohol use are more varied and contested and alcohol's role as a political instrument could therefore be less clearly tied to a majoritarian or nationalist agenda. While the state's efforts to reduce homemade production of alcohol were justified on health and fiscal grounds, these measures triggered counter-narratives from janajati groups, which centred on the role of homemade alcohol in traditional culture and livelihoods. The comparison illustrates that the power of alcohol as a political object varies across post-war societies and that political narratives around alcohol are shaped both by the wider cultural climate for drinking (which was more permissive and complex in Nepal), and historically informed discourses about social behaviour. Third, the political instrumentalization of alcohol is not only performed by the state, but also by rebel groups and civil society actors. Both the Maoists and LTTE used regulation of alcohol as a means of social and political control and to demonstrate moral commitment (especially to abstinence as a demonstration of commitment and purity). For the Maoists in particular, regulation of alcohol formed part of a campaign to shore up support from women. In both countries, civil society groups are involved in maintaining or challenging prevalent narratives about alcohol.

Conclusion
In this article we have advocated looking beyond the literature's narrow focus on the associations between conflict, trauma, displacement and alcohol abuse to examine the wider political and political economy dimensions of alcohol in post-war settings. We have shown that it is difficult to generalise about the role of alcohol in post-war societies. While we have argued that the politics of alcohol is often heightened in post-war settings, the potency of social and political narratives around alcohol varies considerably between contexts and is dependent upon history, social demographics, and preexisting social norms around alcohol use.
We have highlighted the important economic and political interests that shape the supply of alcohol in post-war settings and illustrated how the policing of rules and norms around alcohol consumption is frequently used to reinforce stigmatisation of marginalised groups. Furthermore, we have shown that norms around alcohol may shift during post-war transitions, and that alcohol may also become a symbol that encapsulates societal anxieties about growing connectivity, insecurity, and disorder.
Alcohol can serve as a device for accentuating nationalist political agendas, or as a cultural flashpoint for political contestation, as we saw in the brief examples of internal tensions in Sri Lanka's coalition government relating to alcohol policy and women in 2018, or in the tensions that emerged between the central government and representatives of ethnic groups in Eastern Nepal in 2020. These effects are not unique to conflict or post-war settings but are likely to resonate more powerfully in these contexts since conflict-affected countries are often marked by strong divisions between social groups, and since the ending of war often gives rise to anxieties about growing connectivity, consumption and the erosion of traditional norms.
Our analysis generates three wider implications about understanding the relationship between alcohol and conflict.
First, the ending of war does not by itself explain shifts in post-war alcohol consumption. Any explanation for changes in consumption, production or regulation patterns needs to examine how war may interact in complex ways to other processes including migration, wider processes of economic and social change, growing connectivity, the interests and strategies of the alcohol industry, and the political and fiscal interests of the government. As our analysis of Sri Lanka and Nepal's post-war experience has demonstrated, the ending of war often unleashes economic growth, a consumption boom, and the formalisation of the economy; trends that are often associated with growing consumption of formal alcohol. 41 Second, alcohol epitomises many of the contradictions of post-war transitions and as such may be a useful object of study for understanding some of these contradictory and ambiguous elements (see also Ghiabi's (2018) writing on drugs or Mintz (1986) on sugar). One example of this ambiguity can be seen in how growing alcohol consumption may give rise to counterintuitive links between the ending of war and perceptions of (in)security. As we have seen, alcohol is seen by some to embody freedom, can be viewed as a social practice that strengthens group bonds, and can serve as a (pro-poor) livelihood activity. But at the same time, as findings on northern Sri Lanka show most acutely, it is an exclusive activity that may undermine security for more marginalised groups such as women (especially by driving an increase in domestic violence), and alcohol addiction can also be understood as an indicator of psychological distress. Studying the place of alcohol in post-war contexts can therefore challenge teleological thinking about post-war transitions, capturing how the legacies of war reverberate into post-war, and how transitions may involve progress (a greater sense of freedom, rising disposable incomes) while also intensifying ill health, violence, and exclusion for more marginalised groups (Cramer andGoodhand 2002, Suhrke 2012). Such a focus contrasts with much of the mainstream literature on post-war transitions and peacebuilding, which tends to focus either on the positive dimensions (economic reconstruction, reconciliation) or the negative (corruption, emerging forms of violence). Relatively little attention in the literature on post-war transitions is focused on phenomena whose contribution is more ambiguous and which are more tangentially linked to the ending of war, but nevertheless play an important role in shaping post-war transitions and peoples' experience of them.
Third and finally, our analysis has implications for public policy. In both Nepal and Sri Lanka, tighter regulation was approved after the end of war, but implementation was patchy and undermined by industry pressure. If the stated objective of tighter regulation is reduced consumption, then this outcome can be interpreted as a policy failure (Venugopal 2018a). If the goal of these policies, however, is interpreted more broadly and critically, and we acknowledge that alcohol can be used as a tool to mobilise political support by asserting certain moral ideological frameworks, then these policies can be viewed as at least partly successful (Heitmeyer andSimpson 2014, Mosse 2004). Our research indicates that future policy research on alcohol in conflict settings should be alert to the instrumental functions of alcohol when explaining policy 'failure'.

This research was funded by the University of Bath's International Research
Funding Scheme. 2. In this article we distinguish between the formal and informal sectors. 'Formal' alcohol is alcohol that is legally manufactured and taxed. 'Informal' refers to both illicitly mass-produced alcohol (moonshine or hooch) and homemade alcohol which is not always illegal. 3. A notable exception is Saggers and Gray's (1997) study of alcohol in the aboriginal community in Australia, which calls for an analysis of the promotion and supply of alcohol, alongside exploration of demand. 4. These data relate to the WHO's South East Asian region, which includes South Asia. Globally, levels of alcohol consumption are falling, most notably in Europe (down from 12.3 litres in 2005 to 9.8 litres in 2016). In WHO's South-East Asia region consumption has increased over the same period (from 2.1 litres in 2005 to 4.5 litres in 2016) and is projected to rise in the period up to 2025 (WHO 2018). Current drinkers increased their consumption in almost all world regions except Europe (WHO 2018). A country's level of development is positively associated with higher levels of alcohol consumption across the globe (WHO 2018). 5. A bhatti is an informal place where homemade alcohol is served. 6. Raksi is a generic Nepali term for distilled alcoholic drinks but is normally made from fermented millet. 7. Janajatis is a name for indigenous groups, which include more than 60 communities with different languages, cultures and distinct historical homelands. 8. Interview with academic, Kathmandu, 20 May 2019. 9. The beer market in Sri Lanka is dominated by Sri Lankan owned Lion (with more than 85 per cent market share) and Heineken (15 per cent) and Sri Lankan owned Distilleries Company of Sri Lanka (DCSL) (over 65 per cent) has enjoyed a long-standing dominance in the arrack market. 10. In both countries there is evidence of a decline in formal consumption because of the COVID-19 pandemic, which has restricted availability of alcohol and reduced demand from the tourism sector. 11. The topic of illicit alcohol is controversial and highly politicised in Sri Lanka.
Notwithstanding the significant methodological difficulties associated with estimating levels of illicit alcohol consumption, some available studies have been funded by the alcohol industry which has a vested interest in playing up the threat of illicit alcohol to counter proposed rises in excise duties. Demand for illicit alcohol is strongly driven by the lack of affordability of formal alcohol, and difficulties accessing licenced outlets in rural outlets caused by Sri Lanka's relatively strict licencing regime. 12. These figures were generated by comparing sales data from the Dangerous Drug Control Board's Handbook of Drug Abuse Information (2016) with district population data from 2012. These findings are confirmed by post-war annual spot surveys conducted by Alcohol and Drug Information Centre (ADIC), a Sri Lankan NGO. 13. Growth in alcohol sales was slower in the war-affected East because of the large Muslim population in the region. 14. Interview with NGO representative, Colombo, 9 May 2019. Interview with university academics with experience of working in the North, Colombo 10 th May.