EXPERIMENTAL FINDINGS
What do we learn from experimental research that is useful for understanding or coping with social conflict? There is no way to summarize the vast corpus of experimental findings in the brief space available here, so only a few highlights will be mentioned.
Social dilemmas
Social dilemmas are dyadic or group settings in which one is tempted to be non-cooperative at the expense of the other(s), but if the other(s) are also non-cooperative, one is worse off than if everyone cooperated (see Weber and Messick, 2004). In short, they are variants of the prisoners’ dilemma. Resource dilemmas of the kind described earlier are one form of social dilemma. Arms races and other kinds of security dilemmas are another.
Experimental research reveals a number of ways to enhance the likelihood of cooperation in such situations. One approach is to encourage communication among the parties in the dilemma (Deutsch, 1973; Kerr and Kaufman-Gilliland, 1994). Communication fosters coordination, which is sometimes the main problem. A second approach is to encourage interdependence among the parties, such that each party perceives that the other party(ies) can provide it considerable rewards in exchange for cooperation (Pruitt, 1967). A third approach is for one or more parties to employ a tit-for-tat strategy, in which they begin by cooperating and then match the other parties’ cooperation or noncooperation (Axelrod, 1984; Komorita and Esser, 1975). Tit-for-tat provides rewards for cooperation and punishment for noncooperation. In larger groups, tit-for-tat only works if a number of parties enact it at the same time (Komorita, et al. 1992). A fourth approach is to encourage attraction (Yamagishi and Sato, 1986), perceived similarity (McNeel and Reid, 1975), or perceived common group identity (Brewer and Kramer, 1986) among the parties. These emotions and perceptions foster generosity and trust that others will reciprocate one’s cooperative actions.
A fifth approach is to break a large group into smaller parts since, as was mentioned earlier, there is more cooperation in smallergroups (Allison and Messick, 1985; Yamagishi, 1992).When trust is low and the parties are locked into a vicious circle of non-cooperation, it is sometimes possible for one of them to break out of this circle and take a dramatic unilateral conciliatory initiative. An example of such an initiative is Egyptian PresidentAnwar Sadat’s flight to Jerusalem in 1977, which came soon after a war between Egypt and Israel and paved the way to peace between these countries. Laboratory studies have shown that this tactic is most successful when the actor announces a series of initiatives ahead of time (Lindskold and Aronoff, 1980) and carries them out as announced (Lindskold, et al. 1976). There is also evidence that unilateral initiatives are especially effective when the actor is of equal or greater strength than the target (Lindskold and Aronoff, 1980).
Negotiation
Negotiation is a genteel form of conflict that transforms issues into words. Words allow the development of solutions to complex problems, though there is no guarantee of success. Most of the experimental research on negotiation has looked at the two- party case.
Getting the Otherparty to concede
Some of this research yields advice about how to get the other party to concede. The more the other concedes, the more likely it is that agreement will be reached and the larger will be one's outcomes. Setting high aspirations, making an ambitious initial offer, and conceding slowly at first tend to diminish the other's expectations and encourage the other to concede (Huber and Neale, 1986). These effects are especially large if one makes the very first offer, before the other has a chance to state his or her demands (Galinsky and Mussweiler, 2001). However, there is usually a limit to how far one can pull the other toward one's preferred solution, hence more substantial concessions may eventually be needed to ensure that agreement is reached (Esser and Komorita, 1975).
This slow-fast pattern elicits more concessions from the other party than a fast-slow pattern, in which one concedes a lot at first and then slows down (Benton, et al. 1972). Indeed, substantial early concessions tend to give the other party false hopes that can lead to failure to reach agreement (Bartos, 1974).Positional commitments, in which one indicates that one has reached the limit of concession making, are also useful in pulling the other party in one's direction (Chertkoff and Baird, 1971). But this is also a hazardous tactic because there is usually a limit to how far the other can concede. Hence, experienced negotiators tend to delay positional commitments until they have a good idea about the other party's limit (Kelley, 1966). Another approach to persuading the other to concede is to exhibit anger, which tends to make the other think that one cannot concede very far (van Kleef, et al. 2004).
Finding integrative agreements
In most negotiations, there is the possibility of reaching integrative agreements - of adopting creative options that yield good outcomes to both sides. But such agreements are often hard to find in comparison to simple compromises, in which both parties move toward each other on an obvious dimension. Hence, they require effortful problem solving. Integrative agreements provide a number of benefits in comparison to compromises: they are more popular with the parties, more likely to be complied with, and more beneficial to the relationship between the parties (Pruitt and Carnevale, 1993). In addition, when aspirations are high on both sides, they are the only way to reach agreement (Pruitt and Lewis, 1975).
How to increase the likelihood of integrative agreements? One way is for one or both sides to take a firm but concerned approach to the negotiation, adopting high aspirations for oneself but at the same time seeking an agreement that will satisfy the other party's needs (De Dreu, et al. 2000). Managers can foster such a negotiation style in their subordinates by holding them accountable for the outcome of the negotiation while encouraging them to develop a positive relationship with the other negotiator (Ben Yoav et al.
1984). Another approach is to seek information about the other party's priorities and the interests underlying his or her positions, and to provide such information about one's own priorities and interests (Thompson, 1991). Such information allows one or both parties to locate mutually beneficial options. A third approach is to seek post-settlement settlements - the parties continue negotiating after they have reached agreement, with the understanding that the original agreement holds if nothing better is found. This procedure often yields a more integrative agreement in the second round of negotiation than in the first (Bazerman, et al. 1987).Features of the negotiation setting also affect the likelihood of finding integrative agreements. High time pressure tends to defeat integrative bargaining because it takes time and effort to find mutually beneficial options (Yukl, et al., 1976). Face-to-face negotiation tends to produce more problem solving than message-only negotiation (McGinn and Croson, 2004).
Third-party intervention
Third parties are often important in preventing and coping with conflict. For example, research on aggression reveals many ways in which third parties can reduce the likelihood that frustrated or insulted people will retaliate against those who annoyed them. The horn-honking experiment described earlier suggests that third parties can encourage emotions that compete with anger, for example, by amusing the frustrated individual (Baron, 1976). They can remove angry people from settings that would remind them of aggression (for example, a televised prize fight) (Berkowitz and LePage, 1967) and from aggressive individuals who would otherwise serve as models (Bandura, 1973). Other conditions that encourage frustrated people to aggress include unpleasantly hot or cold surroundings (Baron and Bell, 1975; Berkowitz, et al. 1981), autonomic arousal (for example, by exercise) (Zillmann, et al. 1972), and the necessity of a quick response (Yovetich and Rusbult, 1994).
Third parties can attempt to remove such conditions. Another approach is to try to persuade people that the annoyance they experienced was unintentional or justified and hence that retaliation is inappropriate (Ferguson and Rule, 1983).When conflicts are mild, third parties can help to reduce tensions by encouraging pleasant interaction between the disputants (Pruitt and Kim, 2004). If the disputants are groups and the contact is between individual group members, intergroup tensions will only be reduced if each individual sees the other as a typical member of the other's group (Wilder, 1984). These generalizations do not extend to severe conflicts, where interaction between the disputants tends to be worse than useless, often generating angry and insulting interchanges that intensify the conflict (Rubin, 1980; Sherifand Sherif, 1969).
Third parties can also help disputants to develop superordinate goals - common aims that require cooperation between the groups (Sherif and Sherif, 1969), such as defeating a common enemy. Having and working on such goals enhances unity between the parties in numerous ways (Pruitt and Kim, 2004). However, if superordinate goals are not achieved (for example, if the common enemy wins), unity is likely to disintegrate and the prior conflict will reassert itself (Worchel and Norvell, 1980).
Furthermore, third parties can attempt to diminish ingroup identity in members of groups that are in conflict. This can be done by enhancing identity with a larger entity that embraces both groups, for example, the nation to which both belong (Korostelina, 2005). Another way to diminish ingroup identity is to construct ‘cross-cutting' groups that contain members of both of the conflicting parties (Vanbeselaere, 1991).
Mediation involves third-party efforts to help disputants reach their own agreements. There is a large literature on mediation and how to do it, but relatively little of this literature is based on experimental research (Herrman, 2006; Kressel, et al.
1989; Pruitt and Kim, 2004). We know from experimentation that mediator bias in favor of the other disputant reduces receptivity to mediation (Welton and Pruitt, 1987). However, disputants are less troubled by a mediator's closeness to the other side before the start of mediation than by a mediator's support of the other side's position during mediation (Wittmer, et al. 1991). As mentioned earlier, mediation tends to be more successful if the disputants understand that failure to reach agreement will lead to binding arbitration (Conlon, et al. 2002; McGillicuddy, et al. 1987).Some experimental literature provides advice about mediator behavior. For example, mediators should try to put the disputants in a good mood, as this makes them more cooperative and more creative (Carnevale and Isen, 1986; Forgas, 1998). Mediators should also try to reframe the task and the issues. If disputants see their task as dealing with a ‘partner' or ‘arranging an exchange', they are more likely to reach a mutually favorable agreement than if they see it as dealing with a ‘competitor’ or arranging a ‘business transaction’ (Batson and Moran, 1999; Burnham, et al. 2000). Reframing the options under consideration as various forms of ‘gain’ rather than various forms of ‘loss’ also makes it easier for disputants to make concessions (Bazerman, et al. 1985). Another useful tactic is for the mediator to present as his or her own proposal a position that is acceptable to the other party but would be rejected if put forward by that party because of reactive devaluation (Ross and Stillinger, 1991).
Mediation need not always involve a human being. A number of computer programs have been developed to advise negotiators who are having trouble reaching agreement (Pruitt and Carnevale, 1993). One of these programs, the electronic mediator, asks the parties a number of questions about the dispute and then offers some advice. The developers of this program (Druckman, et al. 2004) have evaluated it in three experiments involving a simulated international negotiation. More agreements and more integrative agreements were reached when the program was accessed by the two sides jointly than when it was accessed by the two sides separately; and the latter condition produced more agreements than were reached in the absence of the program. Clearly, the electronic mediator was a success.
These are only a few of the many experimental findings on social conflict. For a fuller account, see Pruitt (1998).