Implications of the Sociology of Emotions for the Restoration of Social Order

Basic Preconditions of Social Order

Following Berger and Luckmann (1966, p. 52), we conceive social order as a “human product,” or an “ongoing human production,” which is carried out through social interactions in everyday life. Predictability and cooperation are essential to social order. By cooperation we assume any voluntary act of giving up something that some other person will gain. Such acts may be reciprocated, but we do not require them to be so in order to count as instances of cooperation.

Predictability of social interactions is not sufficient for social order: The war of all against all in a Hobbesian “state of nature” is quite predictable, but not cooperative, and therefore not ordered. Only when individuals, at least to a degree, comply with social norms and laws that permit and promote cooperation, can we speak about the emergence of order (Hechter & Kabiri, 2008).

Such a notion of social order need not exclude competition or conflict. Market competition is an example. As long as market participants roughly know what they can expect from each other (“condition of predictability”), that is, as long as there is compliance with norms of market behavior, property rights, monopoly laws, etcetera, we can say that the market is ordered—there is a degree of cooperation in market interactions and they are predictable.

In modern societies order is almost unthinkable without a state. Occasionally, however, in states such as Somalia, or Afghanistan, for example, we are witnessing collapses of the most basic social order, resulting in the absence of a minimum of predictability and cooperation needed for continuation of social life. For example, social conditions in contemporary Afghan society have been described as “normlessness,” that is

a state of deranged order in which one no longer knows what is feasible and what is not; what appears appropriate and what does not; nor which claims are adequate and which ones are not. Individual aspirations become boundless, the result of which is also an increasing appetite for personal satisfaction and gain. (Geller, 2010, p. 58)

We refer to the states lacking basic social order as “failing states.” The bulk of this article deals with the problem of how social order in failing states collapses and how it can be restored.

Social Order as the Endorsement of a Psychological Social Contract

While the notion of social order refers to the macro societal level, the concept of a social contract suggests that the emergence of order at the macro level has its roots in a mutual agreement between individuals, and that internalizing such an agreement on the part of individuals contributes to the maintenance of order at the societal level (Moghaddam, 2008).

Whereas classical conceptualizations of the social contract (Hobbes, 1651/1991; Locke, 1689/1986; Rousseau, 1762/1997) have been founded upon the image of human individuals as rational and self-interested beings, the concept of a “psychological social contract” (Moghaddam, 2008) focuses on the socialization of “psychological citizens,” who may be self-serving, but need not be rational. A psychological social contract assumes an individual’s cognitive and affective acceptance of his or her place within a wider order. This acceptance need not be explicit or conscious and it is developed in the course of an individual’s socialization. A strong point of the psychological social contract approach is that one may, by referring to its psychological dimension, explain compliance to social norms and practices such as wearing a veil or genital mutilation, which are generally detrimental to rational self-interest and therefore cannot be explained within the classical social contract approach (Moghaddam, 2008).

If the social contract is not a matter of rational calculation of self-interested individuals, as Moghaddam (2008) convincingly argues, then how exactly do individuals come to approve of it? We contend that the psychological social contract, which is primarily an abstract, theoretical construct, takes on a more concrete form as a collectively held set of emotional beliefs about the place of individual members of a society within a wider social order. The emotional beliefs underlying the psychological social contract are socially constructed through interpersonal interactions in everyday life.

Mercer (2010, p. 2; following Frijda, Manstead, & Bem, 2000; Frijda & Mesquita, 2000) defines emotional belief as “one where emotion constitutes and strengthens a belief and which makes possible a generalization about an actor that involves certainty beyond evidence.” In order to make the notion more general, we should add that generalizations are possible not only “about an actor,” but also “about an object, or an event,” in the aforementioned definition.

We regard emotional beliefs as a sort of “cognitive shortcut” which spares human beings from laborious, time-consuming searches through all possible courses of action, again and again, each time they encounter a particular social situation. Instead, we assume that humans keep in their memory a set of emotional beliefs, which are formed, and updated as well, in the course of everyday interpersonal encounters. Whenever a subject encounters a particular social situation, instead of considering all possible courses of action, the “record” of the most similar previously encountered social situation is retrieved and the corresponding emotional response is elicited. If the situation is found to connote a positive emotional response, then what we call “cooperativeness” of the subject will generally increase. If, on the other hand, the situation is found to connote a negative emotional response, the subject’s “cooperativeness” will tend to decrease.

A subset of these emotional beliefs concerns the individual’s place within a wider social order. Holding such emotional beliefs, knowing that other people hold them as well, and knowing that others’ emotional beliefs are roughly similar to our own is what makes predictability and cooperation—that is, social order—on a large scale possible.

Suppose, for example, I am a woman and I have been told, time and again, that wearing a veil is a good thing for women. If most people around me approve of women wearing a veil and encourage me to do so, I will also come to believe that wearing a veil is a good thing. I will come to encourage others to do so, and I will come to feel good doing so. That wearing a veil is a good thing would for me indeed “involve certainty beyond evidence.”

In the beginning, wearing a veil might well have been an historical accident, but as soon as it became in some way associated with its carrier’s positive feelings, that is, with the emotional belief or “certainty beyond evidence” that it is a proper thing, it could then take root and spread through the population. The ritual theories of emotions, which, as their name suggests, focus on the role of rituals in emotional arousal, can help shed more light on how specific emotional beliefs emerge in the course of human encounters and gatherings, as well as how those beliefs become disseminated throughout a society, forming patterns and contributing to social order.

Interaction Rituals as Sources of Shared Emotional Beliefs

Collins’s (2004) interaction ritual theory, which draws on Durkheim’s (1912/1965) work on elementary forms of religious life, conceptualizes human interactions as chains of rituals exhibiting up-and-down variation in intensity. The theory focuses on how mutual entrainment of emotion and attention during group assemblies produces shared emotional and cognitive experience, group solidarity, heightened emotional energy, symbols of social relationship, and standards of morality (Collins, 2004).

In the view originating from ritual theories, ritual is not only an element, but the essence of the establishment and maintenance of social order. Different cultures only provide different patterns of enactment of the same universal ritual essence.

Viewed through the lenses of ritual theories, the problem of social order becomes the problem of ritual performance effectiveness. The main determinant of the ritual effectiveness is its intensity: The higher the ritual intensity, the more emotional energy is generated, the deeper the individual internalization of the underlying social contract becomes, the longer the ritual’s socially cohesive effects last, and the more cooperative and predictive—that is, “ordered”—social interactions ensue.

Since larger social units are too abstract to be directly associated with strong emotions aroused during rituals, these larger units need to be represented by symbols such as totems or flags, which can be more easily connected with them (Durkheim, 1912/1965, pp. 250–252). Once attached to certain emotional responses, cultural symbols attain the power to reactivate those responses even without the ritual which had been used to establish the connection in the first instance. In other words, marking transient emotions with cultural symbols, by means of a ritual, intensifies and transforms the initiating emotions into a more lasting emotional energy which bonds individuals together (Collins, 2004, pp. 38–40). Going back to our terminology, it can be said that ritual is a source of shared emotional beliefs by means of which a psychological social contract becomes “lodged” within individual minds.

We can now also easily imagine a woman, for whatever reason, or perhaps even by chance, wearing a veil during some particularly successful interaction ritual which established connection between generated emotions and the presence of a veil over the woman’s hair. The emotional belief that wearing a veil is a nice thing was born and it became “valid” not only for the particular woman in question, but also for all female participants. Moreover, in accordance with the ritual theories, as soon as it became connected with the positive feelings generated in the course of the ritual performance, the road was open for the veil over a woman’s hair to become a symbol of belonging to that particular group, as well as a “moral ought” for all the female members.

Just as positive emotions become associated with certain practices or objects, negative emotions begin to arouse when these practices are not followed, or when the consecrated objects are not approached with due respect. This implies that negative emotions are not absent from ordered societies, but they are channeled so that, in effect, they reinforce existing order by signaling to transgressors that they should change their behavior.

State Failure and the Collapse of Order

So, why do states fail, after all? And is there any connection between interaction rituals and state collapse?

There are many explanations for state failures. According to one prominent explanation (Rotberg, 2003), states fail when they can no longer deliver positive political goods (human security, rule of law, civil and human rights, health care, education, physical infrastructures, and the like) to their citizens. Another explanation (Doornbos, 2002) stresses a number of internal (lack of meaningful linkages between state and society, greed for resources, excessive concentration of power, gross institutional mismanagement) and external (deliberate destabilization from abroad, general economic vulnerability of poor countries) factors, the confluence of which may precipitate state failure. Still further explanations emphasize the role of interclan (Doornbos & Markakis, 1994) or interfactional conflict (Rashid, 2000), external imposition of institutional structures alien to local institutions (Ahmed & Green, 1999), skewed incentives and distorted economic institutions (Acemoglu & Robinson, 2012)–the list could go on and on.

If we look at these explanations more closely, we may find that a common feature of most of them is a claim that, for various political or economic reasons that all such attempts take great pains to specify, something went wrong with the links between the micro and the macro (or, to be more precise, the explanations most often address the links in the other direction, that is between the macro and the micro) societal levels within the failing state under consideration. Rotberg (2003, p. 3), for example, points out that the political goods that collapsing states fail to provide “encompass expectations, conceivably obligations, [that] inform the local political culture, and together give content to the social contract between ruler and ruled that is at the core of regime/government and citizenry interactions.” In other words, a state fails since it loses its capacity to guarantee and enforce the social contract, and it can lose that capacity for various political and/or economic reasons.

The question, however, that is left unanswered in most such accounts is what it really means for a state to lose the capacity for maintaining the social contract. In view of the theory outlined in previous sections of this article, particularly the theorized role of emotional rituals in shaping psychological underpinnings of the social contract, it is not surprising that we suspect that state failure and the concomitant breakdown of the social contract have something to do with the failure of order-generating interaction rituals.

In particular, our claim is that it is not so much that the state is directly involved in interpersonal interaction rituals (although it can be, particularly in totalitarian regimes), but that the state provides the necessary infrastructure for such rituals, and particularly for rituals that stimulate the positive emotional energy of larger groups of people, beyond family and closest neighborhood.

The state itself rarely generates emotionally intensive rituals since the state apparatus seems to be “singularly ill-equipped to mobilize communities” (Turner & Stets, 2005, p. 114). Its authority-based hierarchies simply lack internal capacities for enacting compelling emotionally intensive rituals. Bureaucratic procedures, laws, and regulations are impersonal and austere, while “for codes to be moral [emphasis in original]”—that is, able to induce compliance with social norms and laws—“they must have the capacity to arouse emotions; otherwise, they would be like the contents of a dry instruction manual” (Turner & Stets, 2005, p. 270).

In spite of not being a very effective generator of emotionally arousing rituals, however, the state apparatus can be a very effective facilitator of such rituals. Public schools, hospitals, markets, courtrooms, military barracks, recreational and sporting facilities are all venues of emotionally intensive rituals. Most importantly, these are all places where people from different families, clans, or neighborhoods can meet together and engage in such rituals. And such venues most often crumble with the collapse of the state. In less developed states even religious facilities often depend on state support, which is also diminishing with the collapse of the state.

In addition to providing the necessary infrastructure for interaction rituals, the state can also take a more active part in rituals at the family or neighborhood level in order to incite in the participants positive feelings toward a community that is much larger than the immediate family or neighborhood. Take the example of a state-sponsored war-veteran’s funeral and that powerful moment when the widow receives the state flag that had covered the coffin with the veteran’s body. This ritual act not only transforms the grief of the widow and the closest family into pride, for they have been relatives of such a brave man, but also unites all the participants at the funeral in their feelings of pride and worthiness for they have been citizens of the same country in whose name the deceased man fought on the battlefield. Needless to say, rituals of this kind almost completely vanish with the collapse of the state.

It is therefore not surprising that in the absence of the state’s support for emotionally intensive rituals at the level beyond the immediate family or the neighborhood, the emotional underpinnings of the social contract, together with the social order that rests on them, break down.

One might be tempted to ask: Is it not the other way round—is it not the failure of interaction rituals, in the first place, which leads to the failure of a state? Well, to some extent, this is also true. To the extent that political and economic processes (e.g., interfactional conflict, competition for resources, market system, interpenetration between civil society and political institutions, etc.) can be conceptualized as interaction ritual chains (examples can be found in Collins, 2004), we can also speak of the failure of interaction rituals “causing” state failure. More precisely, if we envision the entire social world as a huge chain (or perhaps a network) of interaction rituals, with emotional energy flowing along the links of a chain, then we can imagine political and economic enablers of state failure as those (relatively large) parts of the gigantic chain that firstly break down, and the state failure as a total breakdown of the whole chain, with perhaps only isolated sets of local links, representing emotional ties between closest family and friends surviving at the micro level.

As yet, we lack detailed empirical data supporting the claim that the breakdown of the state is associated with a lack of emotion-generating rituals. However, there is some evidence from Afghanistan, one of the most studied failing states, testifying to a widespread lack of emotional energy among the population. For example, the 2002 nationwide survey “found that 42 percent of Afghans suffered from post-traumatic stress disorder and 68 percent exhibited signs of major depression” (Badkhen, 2012, p. 34).

Another study (MacDonald, Manfreda, Jackson, & Picardi, 2010, p. 38) focused on two of Afghanistan’s provinces—Helmand and Kandahar—where a sample of respondents were asked which of the listed emotions they feel most strongly on a regular basis. The results show that anger is by far the most prevalent emotion, followed by sadness/depression, frustration, and fear—all negative emotions. Taking into account that these negative emotions can be associated with conflict (anger) or avoidance of others (sadness/depression, fear), it is plausible to expect that they are also associated with reduced occurrence of interaction rituals contributing to predictable and cooperative interactions underlying social order. It is also worth noting that strong positive emotions, which can foster cooperation, have been reported by only 6% of the respondents in Helmand, and 16% in Kandahar.

Communities of Feeling

How to restore the deranged order in failing states? How to replenish the depleted supplies of emotional energy? If we return to the ritual theories of emotion, the answer would be—by restoration of emotion-generating rituals, particularly those involving not only closest relatives, but also members of different clans, groups, or neighborhoods. But how can such rituals be restored in the absence of a functioning state and its ritual-facilitating infrastructure?

Here it may be helpful to refer to Berezin’s (2002, p. 39) notion of “communities of feeling,” which “serve to intensify emotional identification with the polity.” Communities of feeling “bring individuals together in a bounded, usually public, space for a discrete time period to express emotional energy” (Berezin, 2002, p. 39). As we understand this notion, communities of feeling are those meso-level units to which direct emotional attachments may develop—beyond a dozen or two strong attachments that a human individual is able to maintain at the level of immediate interpersonal ties with closest relatives and friends. Turner (2007, p. 195) sees meso-level units as critical in facilitating the use of positive emotional energy for a macro-level social change. We may also conceive of communities of feeling as consisting of mid-range interaction ritual chains, possibly emerging from the aforementioned isolated sets of local links, which survived the collapse of the gigantic chain.

However, an important problem that Berezin (2002, p. 39) identifies is that “[e]ffective emotional politics appropriates rather than generates emotional energy,” implying, as we have already noted, that the state and its political system is primarily a facilitator, rather than generator of socially integrative emotions. “Emotional politics” are therefore parasitic, rather than generative of the communities of feeling, and in the absence of already existing communities of feeling, such politics usually do not stand a chance of success. Moreover, when politics appropriates the emotional energy of already existing communities of feeling and their rituals, it most often uses this energy to pitch groups one against the other. In such cases emotional energy of negative emotions, such as anger and hatred toward “others,” that is, those who do not follow rituals of the same kind (non-Aryans, non-Muslims, non-Westerners, etc.), becomes a “glue” that bonds the community together.

Fortunately, there seem to exist some generic forms of ritual which are universally acceptable and conducive to social order, such as exchange and power and status rituals that we shall discuss later. Here we would like to draw attention to yet another variety of seemingly universally acceptable rituals known as “reconciliation events.”

Research on correlations between reconciliation events and restoration of civil and international order (Long & Brecke, 2003) provides some indications of how politics itself may help generate communities of feeling through reconciliation events, which include direct physical contact or proximity between opposing sides; a public ceremony that relays the event to the wider national society; and ritualistic or symbolic behavior indicating the resolution of previous disputes and prospective more amicable mutual relations (Long & Brecke, 2003, p. 6). Long and Brecke present considerable evidence testifying to the importance of reconciliation events in bringing civil conflicts to an end. They also argue that “forgiveness and reconciliation enable members of a society to maintain stability and mutually beneficial affiliation with each other” (Long & Brecke, 2003, p. 30; see also Nadler & Shnabel, 2008). Since in the aftermath of violent conflicts “others” tend to be viewed as enemies, efforts at reconciliation may, in addition to emotion work, also require cognitive effort toward extension of existing social categories and inclusion of outgroup members within a common “We” category (Jarymowicz & Bar-Tal, 2006).

Long and Brecke (2003, pp. 30–31) attribute beneficial effects of the process of reconciliation to “an emotionally cued change to a specific problem-solving mechanism” taking place within our brains, which “helps us restore relations in our societal group.” We see reconciliation events as one possible way of jolting people out of their daily routine of perpetuating normlessness toward social interactions that are more effective in bringing about positive emotions fostering cooperation.

Change Agents

Under the conditions of deranged order and persistent normlessness, who are those that can start emotion-generating rituals, such as reconciliation events, that can lead toward the formation of communities of feeling?

The notion of change agents is used to describe those individuals who first start to deviate from an established pattern of behavior. Although the concept has been originally developed within the framework of emergent norms theory (Ellickson, 2001), which presumes primarily cognitive actors, there is a general logical equivalence between the emergent norms theory and the more affectively oriented dominant emotions theory (Lofland, 1985, p. 39). Such equivalence justifies using the notion of change agents when considering changes of collectively dominant emotions. The emotional belief that some novel pattern of behavior—for example, forgiving others their past deeds—is “good,” “right,” or in some other way “desirable” can be regarded as a “norm” that an emotional change agent attempts to promote.

As one of the reviewers of this article proposed, change agents themselves may be products of the social emotional order. Individuals at the center of ritual ceremonies are those who get the most out of emotional energy (Collins, 2004). Perhaps, therefore, in failed states one should look for potential change agents among those at the center of surviving local rituals at the micro level. They are most likely the ones who can create new interaction ritual chains eventually leading to the restoration of order on a large scale.

An example of a change agent in action is provided by Mohamed Aden, a Somali-American who returned to Somalia after finishing studies in the US and transformed a small region of Somalia called Adado into an enclave of peace—a community of positive feeling of a sort (Gettleman, Kamber, Abdulle, & Singh, 2009).

In their endeavors to invent novel rituals that would generate positive emotions, change agents may rely on local cultures and their various devices, but use them in innovative ways. This can be justified by referring to dramaturgical theories of emotion, which point out that culture is not only a constraint on individual behavior, but also provides individual agency with “material” to use in their strategic action (Turner & Stets, 2005, p. 26). “Carriers”—means by which styles of thinking and doing are passed from generation to generation, and “story lines”—narrative constructs that weave through societal history and carry forward certain aspects of collective identity (Moghaddam, 2008, pp. 891–894) are examples of such “material” that may be handy for prospective change agents.

An example of an innovative use of a local cultural tradition is fatwā against terrorism, issued by influential Muslim scholar Tahir ul-Qadri. Using a traditional form of expressing religious opinion on issues concerning Islamic law, this Pakistani-born cleric has set out “a point-by-point theological rebuttal of every argument used by al-Qaeda inspired recruiters” (Casciani, 2010). Here we also see that, although ritual can be repetitive and conservatizing, it also provides occasions for change (Collins, 2004, p. 43).

Power and Status Considerations

Change agents may be more successful in their endeavors if they have greater power and status and/or endorsement from higher level, possibly international, civil, or religious authorities. Theories of emotions focusing on the role of power and status underscore that when individuals have or gain power or status, they experience positive emotions, whereas when they lose power or status, they experience negative emotions (Kemper, 1978). Starting from this general assumption, the theories attempt to determine particular conditions of power and status interactional dynamics conducive to the arousal and flow of emotions, as well as the types of emotions being instigated under various conditions (Thamm, 2004).

The main implication of these theories for attempts to restore order in failing states would be that conferral of status to and empowerment of prospective change agents would probably increase their emotional energy and help them reincite interaction rituals generating integrative emotions.

Another interesting topic for consideration, from the viewpoint of applying the sociology of emotions to problems of social order, would be what happens, in emotional terms, when higher ranked individuals have increasingly been seen as undeserving of their status by lower ranked members of the hierarchy. Some clues can be found in the study of impact of dependence and legitimacy of power on emotional reactions in a simulated organizational setting by Johnson, Ford, and Kaufman (2000). They found that subordinates report the highest levels of anger and resentment in response to an act by a superior which they regard as inappropriate, when they are in the least dependent position, and the lowest levels when they are in a highly and equally dependent relationship with the superior. They also report more resentment when the superior is not endorsed—that is, supported or approved—by their peers. Somewhat unexpectedly, authorization of the superior from higher ranked superiors appears to influence neither the emotional reactions, nor the likelihood of expressing emotions of the subordinates. Subordinates thus seem to care more for bottom–up than for top–down legitimacy of their superiors.

This seems to be good news for all those change agents trying to initiate bottom–up change. However, it must be stressed that these results have been obtained in laboratory settings. Viewed from the nation-building perspective, it might be interesting to try to verify these findings with state officials instead of students as study participants. As power/status hierarchies within the state administration are generally stronger than within the academic community, some of these results may not hold for state officials. It is hard to imagine, for example, that the authorization of a superior from higher ranked state officials would not affect the likelihood of expression or—more to the point—repression of negative emotions by subordinates in cases of the superior’s inappropriate acts.

Exchange, Emotion, and the Restoration of Order

Change agents might also capitalize on findings from the exchange literature, where repeated equitable market exchanges between individuals and groups are seen as interaction rituals in themselves, promoting positive feelings and perceptions of predictability. The nature and intensity of emotions experienced by individuals may vary with the type of exchange, the types of structures in which exchanges occur, the relative power and dependence of actors on each other, the expectations for resources, the standards of justice that apply to the exchange, and the attributions that actors make for success or failure in receiving payoffs (Turner & Stets, 2006, p. 41).

The micro relation between exchange and positive emotions may shed a new light on the often noticed macro relation between economic development, conceptualized as large-scale expansion of micro exchanges on one side and social order on the other. Exchange frequency, which may be regarded as a proxy for economic development, positively influences predictability and positive emotion, and the latter positively affects group cohesion (Lawler, Thye, & Yoon, 2000). Since group cohesion (and predictability as well) positively influences cooperativeness, the exchange frequency has beneficial effects on both predictability and cooperation—the two basic ingredients of social order.

This is, in fact, very close to saying that solid chains of a particular type of interaction rituals (exchanges) contribute to the solidity of the overall gigantic chain (representing social order).

Studies of fairness and justice in exchange processes indicate that economic development with distribution of benefits according to merits should be more likely to contribute to social order than development attended by excessive social stratification based on criteria other than merit (clientelistic, nepotistic, etc.). This advice, in Afghanistan for instance, roughly translates into “encouraging a more responsive and transparent state, promoting more merit-based appointment mechanisms, building social capacity along the lines of the National Solidarity Programme, and addressing abuses of power that look inequitable to the population” (Thompson, 2010, p. 3).

Meritocratic distribution, however, does not imply absence of any conflict. Since the level of own outcomes affects people’s perception of fairness (Hegtvedt & Killian, 1999), low performers may regard meritocratic distribution as unfair. Our argument is simply that, if equitable distribution is more attuned to the emotions of most of the people (that is, if at least medium and high performers regard it as fair), then the total level of conflict under meritocratic distribution will be less than the levels of conflict under distributions that rely on other criteria.

Underrewards of a given amount seem to affect the evaluation of justice more strongly than overrewards of an equal amount (Hegtvedt, 1990). In other words, although people feel bad when underrewarded, they may still feel good when overrewarded. The transition from, say, nepotistic to meritocratic distribution may get complicated because established expectations of those overrewarded may prevent them from seeing their rewards as unfair, and make them unwilling to refrain from their claims. Reluctance of the overrewarded to relinquish privileges may also be expected within populations characterized by excessive gender inequalities, as well as inequalities between different ethnic, racial, or religious groups.

An important implication of the affect theory of social exchange is that successes in productive exchanges—that is, group-oriented coordination tasks in which actors seek to produce a valued result through their joint collaboration (Lawler et al., 2000, p. 619)—are likely to promote micro social orders (Lawler, 2001), the establishment of communities of feeling, and the restoration of social order. Lack of a productive exchange component might, for example, help explain why the so-called Quick Impact Projects in Afghanistan “mainly achieve short-term stability benefits that are restricted to the immediate transactional exchange, … but they do not reflect the type of transformation necessary to generate trust and reduce suspicion over the long-term” (Thompson, 2010, pp. 13–14).

Conclusions

The most promising way to restore social order seems to be the “bottom–up” way. Since emotion can be regarded as “that part of the social relationship in which the subject of the relationship … is in some way changed, and, in being so changed, is disposed to change the relationship itself” (Barbalet, 2002, p. 4), changing emotions of individual subjects often changes their relationships with others, which provides a social foundation for changing wider social structures. In such a way, the role of certain local actors as change agents may be encouraged, thereby beginning a possible cascade of micro changes at the level of individual interactions, leading to an all-pervasive social change at the macro level.

Change agents may start new or link to already existing interaction rituals into longer, further-reaching chains. The more emotional change agents there are, the greater the chances that positive affective attachments would reach out of small and closed cliques, toward the emergent communities of feeling, and across the wider society.

As the change agents employ interaction rituals, various cultural devices, means of exchange, and power and status conferral in the course of their social interactions, positive, socially integrative emotions—which tend to circulate in interpersonal encounters, families, and small groups—are gradually channeled toward larger social meso and macro structures, thereby increasing individuals’ commitments to formation, maintenance, or constructive change of these larger structures. Such commitments become the core of emotional beliefs about the place of individual members of a society within a wider social order. A psychological social contract becomes established and the usual socialization mechanisms contribute to its maintenance and reproduction.

The processes that we describe are inherently complex and frail. Therefore, nation-building projects aiming to restore social order in failing states cannot be reduced to “mechanical” polity-building efforts. We do not claim that political procedures and institutions are irrelevant for the establishment, maintenance, and reproduction of social order. Our contention is that efforts in rebuilding political institutions in failing states are less likely to succeed if not coupled with a wider concern for the restoration of emotionally fulfilling rituals that would connote a sense of meaningful participation in political institutions and social life in general.

Naturally, the picture that we have begun to paint in this article is not yet complete. Further research will need to add more detail along the lines drawn here. In particular, we need to know more about emotions experienced by inhabitants of failing states. Data sources should not be limited to self-reports of local actors. The scope of research should be widened in order to include various forms of public discourse, cultural products, and other available material. A closer interaction between the fields of international relations, policy studies, political sociology, and sociology of emotions might be beneficial to all these fields. The relatively young field of the sociology of emotions is not mature enough to offer a comprehensive account of social order, nor is it likely that any of the other scientific fields could do it alone.