Risk Sensitivity and the Sikh Uprising in the Punjab

Risk Sensitivity

Risk sensitivity is a measure of an individual’s attraction to or aversion to risk. Risk is defined probabilistically when an individual has some sense of the probability distribution of outcomes that will confer satisfaction or utility on the individual (Tversky & Fox, 2000, p. 96). Risk can be assessed for any form of gains or losses, both economic and non-economic (social status, prestige, social identity, religious aspiration), although material wealth is often a key variable. Risk sensitivity can vary from risk adverse to neutral to risk prone. A risk neutral decision maker would be indifferent between a definite outcome and a probabilistic lottery with an expected utility (utility of outcome = probability of outcome) equal to the definite outcome. Thus, the utility curve defining a risk neutral preference is linear. Individuals frequently do not behave in a risk neutral manner; instead, they often are risk averse and prefer a certain outcome rather than a lottery outcome. On the other hand, individuals who prefer the lottery to the certain amount of equal expected value are risk prone. Figure 1 contains an example of a distribution of wealth and status among groups that are likely to be risk prone and those that would be more likely to be risk averse.

OPEN IN VIEWER

Figure 1.

Source: Adapted from Kuznar and Lutz (2007).

An individual’s preferences towards uncertainty can vary throughout a range of wealth or other utility conferring goods. Friedman and Savage (1948) proposed that class differences would create S-shaped (sigmoid) utility curves that alternated concave downward and concave upward. The flattened areas of sigmoid utility curves represent social classes and the steep boundaries between these flattened areas represent class boundaries as can be seen in Figure 1. Individuals in the concave sections of these curves should be risk averse while individuals in the convex sections should be risk prone or attracted to risky prospects. The implication of the sigmoid utility model is that members of groups or classes close to ascending a boundary should be willing to take chances to rise up to the next social class. Sigmoid utility theory also suggests that when qualitative jumps in social status are possible, taking risks outweighs the potential loss in status that could occur (Kuznar, 2007, p. 321). Further, it is also possible that individuals in danger of suffering status loss could become risk prone.

People often judge their well-being and their status by comparison with the material well-being of others. The material indicators of relative social status may be particularly important for analysing political behaviour. Security is another vitally important political good, but it is often difficult to obtain without first acquiring material support. Perhaps most important, the same factors that negatively affect wealth can also negatively affect status. Status considerations are linked to wealth when individuals have frustrated career expectations, which can lead to violent behaviour of groups and their members (Kuznar, 2007; Kuznar & Lutz, 2007; Waldmann, 2005, p. 157). Status, of course, is not strictly linked to wealth. Caste systems are a prime example of a status system that may not be directly related to wealth.

The extension of individual goals as well as frustrations to groups, and the reciprocal influence of the group on individuals, is another important dimension that is linked to differences in wealth. In cases such as these, a threat to the group or its material well-being logically becomes a threat to the individual group members and can lead individuals to political violence (Cooney & Begman, 2015, pp. 28, 33–34). The fact that material threats to groups create problems for the member’s individual goal is obvious in the rhetoric of dissident groups. Whether they are Marxist, Islamic, white supremacist or ethnic separatist, advocates often have invoked material inequality as a barometer of social and moral inequity. What is striking is that often the stated objectives of these groups are antimaterialistic, but at the same time resentment over inequality or perceived inequality is evident in their rhetoric (Bueno de Mesquita, 2005; Kuznar, 2007).

Loss or deprivation alone, however, does not explain outbreaks of terrorism. It would appear logical to assume that persons in poverty would have a lot to gain by joining groups willing to engage in terrorism. Poverty and terrorism have been linked in some analyses (Collier, 2000; Gurr & Moore, 1997; Khan & Azam, 2008). Mousseau (2011) found indications that it was specifically urban poverty that was linked to terrorism, but the improvement in the explanation of variance with this distinction was relatively small overall. Many other, and more, analyses, however, have found that poverty and terrorism have not been linked in any identifiable or consistent way (Coggins, 2015; Dalacoura, 2011, p. 63; Horgan, 2003; Krueger & Maleckova, 2003; Piazza, 2011; Riaz, 2016). Poverty is very likely to have contributed to some outbreaks of terrorism in specific circumstances, but the overall conclusion is that any connections between poverty and terrorism are indirect and complex if they exist (Lutz & Lutz, 2011, p. 17). Social anomie would be another obvious factor that could explain the link with those in urban settings. Such anomie would be similar to that which attracted recruits to fascism and communism after World War I (Mousseau, 2011, p. 42).

Prospect theory adds another element to risk sensitivity. The elements of prospect theory most relevant to the current approach are the concepts of loss aversion, framing effects and status quo bias. Loss aversion is the recognition that losses are regarded differently than gains, fundamentally challenging the notion of expected utility values. The basic finding from loss aversion is that there is more displeasure at a loss than pleasure from a gain (Jervis, 1992, 1994; Kahneman & Tversky, 1979; Kahneman, Knetsch, & Thaler, 2000; Kuhberger, 2002; Levy, 1994; Tversky & Kahneman, 1992). Loss aversion increases risk sensitivity and suggests that individuals will be more likely to take action to avoid loss than to achieve gains. Framing effects refer to the fact that people are generally risk prone when considering losses—they will take a chance to avoid a loss, and they are risk averse when considering gains. Frames over losses are termed negative and frames over gains are termed positive. Status quo bias refers to people having sticky reference points and refusing to give them up when it would be in their interest to do so (Kahneman et al., 2000; Tversky & Kahneman, 1981). This situation suggests that a frame, once established, may be particularly persistent. Such entrenched attitudes would help to explain why terrorist organisations will often continue to survive and even attract new recruits even when the security forces have been successful in dealing with the group or when government programmes have ameliorated the conditions that contributed to the violence.

Connecting the framing effects to the risk sensitivity approach outlined above suggests that the sigmoid utility will have a predictive value when individuals are expecting gains and when people find themselves in a situation in which losses are possible. The interactions between framing effects and loss aversion indicate that decisions made in a loss frame are not only likely to be risk prone, but they are much more risk prone than what would be predicted in a gain frame. In other words, a conservative, risk-averse decision maker should become a radicalised, risk-prone decision maker in a situation where previous economic gains or social privilege or status are being lost. This effect has been considered as a factor in the fomenting of revolutions, terrorism and political activism (Kuznar, 2007; Kuznar & Frederick, 2003; Kuznar & Lutz, 2007).

While some individuals or groups might become involved in terrorist actions in order to achieve major gains, other individuals and groups will become more risk prone if their current status is threatened and they face major losses. Protestants in Northern Ireland when faced with efforts by Catholics to achieve greater equality used violence to protect their position in Northern Ireland, which helped to fuel violence by Catholics (Kuznar & Lutz, 2007). Migrants that have moved to a new country have often suffered from a loss of status in their new circumstances compared to their position in the past which can facilitate their radicalisation and movement towards terrorism (Cooney & Bigman, 2015, p. 33). Poverty, according to earlier research on rebellions, has not been a consistent explanation for terrorism. The poor were too focused on survival to join violent groups. When economic conditions became so bad that their survival was threatened, however, the poor did become politically active and join in rebellions. When this ‘subsistence ethic’ was violated, violence resulted (Welch, 1980, esp. pp. 27–31). Obviously, when groups fell below a survival level, they would become more risk prone and would attempt to return to a higher level of existence, even one that was less than ideal. The threat or reality of loss generated risky behaviour, not the anticipation of gain. Loss of status would explain the violence involving the Tamil minority in Sri Lanka. For a variety of reasons, the Tamils were relatively privileged compared to the majority Sinhalese. One of the causes for the start of the quarter century of violence was the efforts of the majority to reduce that position of privilege and to empower the majority. The Tamils resorted to terrorism and civil war to protect their position that was threatened with loss; they were not fighting to enhance their position in the Sri Lankan society and economy (Kuznar & Lutz, 2007).

The Conflict in the Punjab

While there are many variables that contribute to outbreaks of terrorism and political violence, the idea of risk sensitivity and the associated concepts of framing and prospect theory help to provide insight into why both better-off groups and less well-off groups would resort to violent tactics. The violence in the Punjab, like most major outbreaks of terrorism or insurgency, has had multiple causes. One factor that was important in this case was the desire of many Sikhs to protect their political and economic positions in the state and country. A number of other factors came into play, including the distribution of powers between the national government and individual states, disagreements over the allocation of water for irrigation, religious tensions between Sikhs and Hindus, external support for the Sikh dissidents (principally from Pakistan) and overall economic conditions of groups in the state. Some of these considerations were linked to status concerns that were present for many Sikhs, while others were largely independent of these concerns or only had indirect connections.

As noted, Sikhs in India were not a group that had been economically disadvantaged or one which faced consistent discrimination from the majority Hindu population. The Sikhs were allied with the Hindus against Muslims in the partition conflicts of 1948, and many Sikhs were forced out of Pakistan in the partition violence. The Punjab, where the Sikh population was concentrated, was one of the wealthier states in India (Svensson & Harding, 2011, pp. 138–139). In fact, the economic situation of Sikhs was impressive, and the Punjab led other states on a variety of economic measures. For the Sikhs there was a clear connection between wealth and status since those that did well were also from groups with higher status (Wallace, 1986, p. 363). Further, Sikhs were prominent and well represented in the national commercial sector. Religious compatriots in the diaspora were important in providing international commercial contacts for their coreligionists in India. Thus, poverty and economic disadvantages did not place the Sikhs in a situation in which they would be driven to violent activities in the hopes of making significant gains in status or wealth, even if there was an apparent opportunity to make a major gain in status, wealth and class standing. They were actually in a relatively privileged position vis-à-vis other linguistic, ethnic or religious segments of multicultural Indian society. As a consequence, while they were not likely to resort to violence in anticipation of making major gains, they held an economic place in Indian society where threats to their position could lead them to become risk prone and to use violence in an effort to prevent losses.

The Punjab was an Indian state that realised great benefits from the Green Revolution, and many Sikhs saw increases in their personal wealth. The Punjab was one of the most productive agricultural areas in India with the highest rural income in the country before the Green Revolution (Wallace, 1986, p. 367), and the new seeds and fertilisers of the Green Revolution resulted in even greater productivity. The advantages of the increased productivity, however, were not spread evenly through the population. The general view is that the Green Revolution did not actually lead to increases in poverty in the rural area of the state, but small farmers and landless peasants did not benefit and there were increased levels of inequality between them and the richer farmers (Bryjak, 1985, p. 30; Paroha, 1993, p. 244). The differential impact of improved agricultural practices and profits created tensions since the poorer members of the agricultural community resented their lack of economic progress (Bryjak, 1985, p. 31). The Green Revolution thus created one group of Sikhs with an improved status that they might want to protect; additionally, there was a second group that may have perceived that the Green Revolution provided an opportunity to improve their status and wealth but those benefits were not realised (Paroha, 1993, p. 244). Marxist analysts have gone further to suggest that the Green Revolution resulted in the creation of a class of capitalist farmers who then exploited agricultural labourers (Singh, 1987, pp. 1275–1277). It has further been suggested that the Green Revolution led to pauperisation in the Punjab, which led to greater potential for ‘individual and collective violence’ (Bryjak, 1985, p. 37). These views overemphasised potential class relationships and ignored the other factors that contributed to the outbreak of violence but they do identify one contributing factor.

Political parties representing the Sikhs had long dominated the state government in the Punjab, but they were facing the increased migration of Hindus into the state seeking to take advantage of opportunities for advancement. The fact that the Punjab had one of the highest average incomes, moreover, made it an attractive destination for Hindus from elsewhere in the country. The Green Revolution actually contributed to the change in population balances since the new agricultural productivity attracted Hindu agricultural workers (Singh, 1987, p. 1272; Wallace, 1986, p. 365). This influx of Hindus created a situation in which there was the potential for increased competition for access to economic opportunities and resources. These resources included control over water for irrigating crops that was essential to the greater yields that came with the Green Revolution. There was even a possibility that the Sikhs could be outnumbered by Hindus in the state. The Sikhs saw themselves becoming a minority in what they considered to be their own state (van Dyke, 2009, p. 994). This fear had first led to Sikh agitation for dividing the initial Punjab state into two parts that gave the Sikhs what appeared to be a secure group majority in the smaller Punjab state that resulted, but even that majority was threatened. In addition, the increased wealth in the farming sector made it easier for Sikhs to migrate to other parts of India and the world (Bryjak, 1985, p. 33; Wallace, 1986, p. 365). This shift raised the danger that parties representing the Sikhs could lose control of the state government and limit access to resources. This changing situation represented a more indirect challenge to the status and position of Sikhs. At the same time, the influx of Hindu farm labourers seeking employment also created the potential for conflicts between poorer Hindu labourers and the wealthier Sikh land owners.

There were indications that there were other economic factors that contributed to the tensions in the Punjab and the resulting popular unrest among Sikhs. The general increase in well-being meant that control over local resources for the benefit of the local residents had become even more important (Wallace, 1986, p. 376). A related economic factor was the perception of many Sikhs that the Indian system of capitalism had been structured to favour Hindus to the detriment of the inhabitants of the Punjab (Narang, 1987, p. 544). These perceptions, whether justified or not, would have increased the feeling of some Sikhs that they were facing economic and status losses in a system that was weighted against them; consequently, they were willing to consider violence as a remedy. It is likely that these particular economic considerations were not sufficient to lead to the violent outbreaks. As well as the direct effects that they had on the citizens of the Punjab, they would have to be included among the indirect effects that threatened the position of many Sikhs in the Punjab.

Political decisions in New Delhi reinforced the negative perceptions held by many of the Sikhs. By 1982 and 1983, Prime Minister Indira Gandhi and her Congress Party (I) faced the prospect of an electoral defeat at the national level. One consequence of this possibility was the use of the potential Sikh challenge to Indian unity at the national level to rally support for the Congress Party in elections (Singh, 1987, p. 1273). Further, Gandhi attempted to weaken the Akali Dal, the dominant Sikh party that was loosely allied with a national coalition of parties opposed to her Congress Party (Chowdhury & Krebs, 2009, p. 385). Gandhi was consciously pursuing a policy of divide and conquer in the state by supporting competing Sikh politicians (Fair, 2005, p. 128; Paroha, 1993, p. 241; van Dyke, 2009, p. 979; Yaeger, 1991, p. 224). While Sikhism did not adopt the caste system of Hinduism, social practice did distinguish between higher status and those who worked in lower status occupations (Ram, 2004; Wallace, 1986, p. 366). This social differentiation provided an opening for the Congress to gain voting support among the less well-off Sikhs since the Akali Dal did not have widespread support among the poorer (low-caste) groups (van Dyke, 2009, p. 995). The successful appeals to lower status Sikhs and the strategy of divide and conquer were designed to permit the Congress Party to retain control at the national level by weakening the opposition. In addition, the Congress Party hoped that increasing conflicts between Hindus and Sikhs in the state would prevent any cooperation between the Akali Dal and other Hindu parties strong in urban areas that were also part of the national opposition (Leaf, 1985, p. 484). Ironically, some of this support was even initially provided to Jarnail Singh Bhindranwale, who eventually became the leader of the most threatening violent groups that appeared (Ganguly, 2003, p. 16; van Dyke, 2009, p. 986). The attempts to politically divide the Sikh community combined with the increased immigration of Hindus directly contributed to the fear of many Sikhs that they would be marginalised in their own land (Lutz & Lutz, 2013, p. 86).

The tensions in the Punjab were also exacerbated by national politics in other ways. Differences between the politicians in the Punjab and those in New Delhi were heightened by the inevitable conflicts involved in the division of powers between the national and state governments in India. Federalism always has such tensions. For the Sikhs, given their perceptions of overreach by the central government, victory for Sikh parties at the state level and strong state government would provide an effective combination for the protection of communal interests. Conflicts over the allocation of important water resources were generally resolved at the national level, leading to the suspicion that local Punjabi and Sikh concerns were being compromised to reward others (Singh, 1992, p. 992; Wallace, 1986, pp. 372–373). Such suspicions increased as Gandhi had sought to centralise power at the expense of the states (van Dyke, 2009, p. 979). These conflicts were essentially constitutional in nature reflecting disagreements between the central and state governments. Part of the difficulties involved the fact that central government powers were seen as providing a bias towards industrial and commercial interests in the country at the expense of the agricultural sector (Leaf, 1985, pp. 480, 490). Farmers in the Punjab found themselves subject to national legislation that they perceived to be responsible for the exploitation of their contributions to the nation without corresponding investment for industrial development (van Dyke, 2009, p. 984). This effort was bound to generate concern for a minority population such as the Sikhs. These perceptions would also have increased the willingness of various elements of Sikh society to adopt more risk-prone behaviours to defend or regain an advantageous position. Gandhi for a variety of reasons had rejected suggestions for the regional devolution of powers to local leaders that conceivably might have reduced the increasingly high levels of tension in the state (Svensson & Harding, 2011, p. 139). Some concessions had been made. The boundaries of the Punjab were redrawn so that Sikhs were more likely to politically control the smaller state, even though they remained in competition with allies of the Congress. Ultimately this change was insufficient to prevent the violent outbreak that occurred (Nandi, 1996). In the final analysis, Gandhi eventually chose to present the escalating dispute that occurred in communal terms, which made compromise more difficult (Leaf, 1985, p. 493). She was, in essence, framing the conflict in a way that provided some benefits to her political position, but this approach clearly contributed to the fears of many Sikhs that their positions in the Punjab would continue to decline.

While many of the above discussions of conflicts in the Punjab suggest that religion as such was not an issue in the Punjab, there were definite tensions that were rooted in religious differences. One religious concern for the Sikh community involved the nationalist Hinduvata movement that had appeared in the country and that sought to exclude foreign religious influences. Supporters of the movement in the 1980s were primarily targeting Muslims—and infrequently on occasion, Christians. While Hindus frequently considered Sikhism to fall within the broad bounds of Hinduism which is very inclusive, the Sikhs considered themselves to be a quite separate religion (Hoffman, 1995, p. 279; Nandi, 1996, p. 180). As followers of a religion that see themselves to be part of a different belief and one that has drawn on foreign elements for its theology, the Sikhs could recognise the dangers that Hinduvata might represent. There has been a fear of re-absorption into Hinduism as a consequence (Nandi, 1996, p. 183; Wallace, 1995, p. 361). This fear was increased by the fact that some Hindu religious figures suggested that Hinduism should reabsorb the Sikhs (Narang, 1987, p. 546). The Hinduvata movement thus constituted a possible threat for Sikhs and raised concerns among the adherents (Subramanian, 1999, p. 728; Unnithan, 1995). The fear for Sikhs was exacerbated by the fact that many Sikhs felt the national government had been soft on Hindu fundamentalists and their activities (Paroha, 1993, p. 247).

The Sikh religion came into play in another fashion. Bhindranwale mobilised much of his support among dissidents in religious terms. His arguments were particularly appealing to fundamentalist Sikhs. The fundamentalists among the Sikhs feared increasing secularism that had appeared in India (Narang, 1987, p. 547). In this sense, fundamentalist Sikhs were like fundamentalists of other religions in that the greatest threat to their religious, social and cultural values was secularism (Lutz & Lutz, 2013, p. 80; Pillar, 2001, p. 65). He found the changes in lifestyle that had been occurring among Sikhs to be offensive to his religious beliefs. These lifestyle changes were especially noticeable among the more prosperous members of the community (Bryjak, 1985, p. 31; Wallace, 1986, p. 377). His arguments for traditional values and ability to raise conscious and support led some militants to impose their religious views on all Sikhs in some areas (van Dyke, 2009, p. 990). Bhindranwale’s fundamentalism and his appeal to those who shared his beliefs set him apart from the Akali Dal. He was able to increase the Sikh consciousness of their group identity through his efforts, which also created the potential for increased political power for Sikhs if communal cohesiveness would create a monolithic voting bloc (Singh, 1987, p. 1269). Bhindranwale used other appeals as well; he played on the Sikh fears that they would become a minority in their own lands (Wallace, 1986, p. 363). This interjection of religion into political debate generated benefits for parties and groups that were using the religious theme including some Sikh politicians and Gandhi and her framing of issues in communal terms (Narang, 1987, p. 544). The fundamentalist groups, the Akali Dal and the Congress (I) locked themselves into positions where they were supporting more extreme views as part of their larger electoral strategies (Singh, 1987, p. 1274). The linkage of religious views to political conditions in the Punjab reinforced the status concerns of many Sikhs and directly led to the perception among many of them that the position of the entire group was in need of defence.

Finally, at a very broad level the processes involved with modernisation and globalisation contributed to the tensions in the Punjab. The modernisation of the political system had major effects on the Indian political system, exacerbating the difficulties in the Punjab as it generated disruption in the previously usual processes (Narang, 1987, pp. 542–543; Singh, 1987, p. 1272). Political decisions at the national level to deal with national issues created problems as noted with the perception of unfairness towards the regional and local issues that were important for the Punjab. The Green Revolution clearly involved aspects of modernisation and globalisation that have generated their own set of tensions in many societies and which can undermine social connections that hold a society together (Nieman, 2011, p. 275; Singh, 1987, p. 1271). The trend towards greater secularisation that challenged Sikhism as a religion is another one of the by-products of globalisation and modernisation. Globalisation also facilitated the migration of Sikhs to other parts of the world and the resulting support of these diaspora communities for the struggle of their religious cohorts in India. Ultimately, the effects of globalisation and mobilisation in terms of disrupting societies and economies are inevitable and unavoidable. They can be met with different strategies and policies by governments and local leaders that could ameliorate the problems rather than adding to them. These problems are not particularly unique to India as a society undergoing change even though they provide a broad context for concerns about lost status or hopes for increased status.

The concerns of many of the Sikhs eventually led many of them to engage in attacks against the central government and those in the Punjab that were seen as local collaborators. Bhindranwale in his efforts gave the poor and unemployed youth the prospect of improving their lot (Paroha, 1993, p. 244). The dissident groups, however, attracted support from many economic sectors in the Punjab. Two-thirds of militants were landless labourers or small farmers, 22 per cent were middle class and there were some upper class and wealthy individuals involved (Singh, 2012, p. 147). These percentages, however, are not dissimilar to the general makeup of society in the Punjab, indicating that all elements of Sikh society were involved in the violent confrontation with the central government. Interestingly enough, many lower caste Sikhs tended to continue their support for the Congress Party and not the dissident groups (Wallace, 1995, p. 377). Bhindranwale was the leader of one of the major violent dissident groups, and while he appealed to many segments of the Sikh population, he was especially able to attract young and educated members of the community in addition to the poor and unemployed and those who had not improved their status with the successes of the Green Revolution (Paroha, 1993, p. 244; Yaeger, 1991, p. 227). His recruitment pattern, as well as appeals of a number of different militant groups to smaller farmers, the middle class and wealthy individuals, all indicate that there were many members of the Sikh community who saw the dangers of a reduction in their positions in society and thus became more risk prone.

One final cause has been put forward to explain the violence or which greatly contributed to the violence once it occurred. It has been argued that the violence was fuelled by external actors, principally Pakistan, and that these were supporting the uprising as part of an effort to break up India (Singh, 1987, p. 1270). There is little doubt that there were some Sikh organisations that received support from Pakistan as part of an inexpensive foreign policy method for creating difficulties for an unfriendly and potentially dangerous neighbour (Lutz & Lutz, 2013, p. 88; Ogden, 2013; Paroha, 1993, p. 246). In fact, while Pakistan may have been more than happy to see unrest in the Punjab, external support was not at the core of the violence. Governments often find it useful to displace the blame for domestic unrest onto foreign enemies as opposed to blaming what the domestic government did or did not do (Lutz & Lutz, 2006, p. 92). Such a displacement of blame, of course, can rally domestic support and provide a means of obscuring the failings of the home government. In the case of the Punjab, foreign involvement would appear to have been a minor factor at best that led up to the violence.

The resulting campaign of violence and government responses led to tens of thousands of fatalities. Similar to the objectives of the Tamil Tigers in Sri Lanka, at least some of the Sikh dissidents were attempting to create an autonomous or independent Khalistan (Land of the Pure). The actions that the central government instituted ultimately increased the fears of the Sikh community. From the perspective of the national government the violent groups in the Punjab constituted a major threat not only to government control in the state but as a potentially disastrous example for the overall unity of India. If the Sikhs successfully gained greater autonomy or independence, other ethnic or linguistic groups could follow. Further, the Punjab is the gateway to Jammu and Kashmir, where, of course, other violent dissidents, in this case obviously with the active support of Pakistan, were challenging the government. The violence in the Punjab potentially could have weakened Indian control of this disputed territory. Given the resulting importance that the government placed on dealing with the threat from the Sikh militants, Prime Minister Gandhi declared an emergency and placed the state under presidential rule that superseded the jurisdiction of local elected officials. Presidential rule limited civil liberties in the state and led to the declaration of martial law (Wallace, 1986, p. 375). This situation was another contributory factor to the threatened loss of status for many Sikhs since there was no longer any opportunity for direct political influence through elections and state politics. Gandhi also approved Operation Blue Star in 1984 which was directed against the militant organisation led by Bhindranwale who was using the Golden Temple at Amritsar as his headquarters. The military assault on the holiest site of the Sikh religion killed Bhindranwale and many militants, but it also killed many Sikh pilgrims and damaged the temple and surrounding buildings. While the attack did kill a prominent dissident leader, it also fuelled the perceptions of threat for the Sikh community. The military assault and the large number of deaths confirmed for many Sikhs that Bhindranwale was correct in the arguments that led his followers and other Sikhs to accept that the central government was biased and that they should be suspicious about the goals of the central government (Leaf, 1985, p. 494). Ultimately, the attack on the Golden Temple and the more general indiscriminate use of force by the military and security agencies to deal with dissidents further radicalised the Sikh population (Chowdhury & Krebs, 2009, p. 384; Singh, 1992).

The fears of the Sikhs increased in the aftermath of the assault on the Golden Temple when Sikh bodyguards of Prime Minister Gandhi assassinated her in retaliation for her authorisation of the attack on the Golden Temple and the subsequent casualties and damage. The killing of Gandhi led to riots in New Delhi in which Hindu mobs targeted Sikhs. The Hindus held all Sikhs responsible for Gandhi’s death. The attacks also reflected Hindu resentment of the economic success and position of Sikhs in the capital and the northern parts of the country with Hindi populations. In New Delhi, the attacks often concentrated on well-to-do members of business castes (Bryjak, 1985, p. 32). Thousands were killed in these attacks (Ganguly, 2003, p. 17). This mob violence led to as many deaths as it did because local police remained inactive and because some local politicians even encouraged the rioters (Chowdhury & Krebs, 2009, p. 386; Fair, 2005, p. 136; Nauriya, 1996, p. 301; Pettigrew, 1995, pp. 8–9). The massacres also brought the Sikh diaspora into play resulting in increased support and leading to greater external support for the efforts to create an autonomous or independent state (Fair, 2005, p. 128). The riots also once again heightened the perception of the threat that many Sikhs felt. They not only increased the fear of economic or status loss but now included loss of life as well as many Sikhs became targets. These attacks in the aftermath of the assassination undoubtedly prolonged the anti-government violence by local militants in the Punjab. In addition, the violence was prolonged because the more moderate Sikh political leaders in the Punjab were in jail at the time of the assassination. They had been arrested with the prior imposition of martial law which permitted such detentions. Their imprisonment was counterproductive since it silenced the voices of moderation and left the field open for the more extreme leaders among the dissidents (Leaf, 1985, p. 491).

Conclusion

The violent challenge by the Sikh militants in the Punjab was finally successfully met by government forces, and the state eventually returned to more normal conditions. As the above discussions indicate, there were a large number of factors which were involved in the violence. The circumstances that led to the violence were indeed complex and included religious tensions, the influx of Hindu migrants who threatened the demographic and political position of the Sikhs in the Punjab state, political manoeuvring related to national politics, underlying economic and political issues between the national government and the state, central government actions in response to the initial attacks by dissidents and other factors. What is clear is that many of the recruits to Sikh militant groups had become more risk prone when they faced threats to their economic and social status and their very lives. The Sikhs were not primarily protesting high levels of poverty or any violation of the subsistence ethic; the violence was directed towards preserving the position of the group, especially in the Punjab. The framing of the conflict was one that concentrated on preventing the loss of status for the community. Individuals were willing to use terrorism in defence of the community against this loss. Even though the Sikh uprising provides important support for the idea that groups facing the loss of status, wealth or position are among the ones more likely to resort to violence, there were some indications that other Sikhs participated in the violence in the hopes of improving their position in society. Of course, there is nothing in the nature of societal conflict that would suggest that both those hoping to gain and those hoping to preserve their status could not both be involved in such a complex situation. Clearly, however, a significant number of participants in the struggle were responding to threats to their positions. The situation of the Sikhs provides further evidence as to why there are a number of important cases where poverty and lack of opportunity are not the primary cause for outbreaks of terrorism, and governments and analysts need to be aware of this possibility.