magz4

SPREADING THE WORD: THE ROLE OF SURROGATES IN CHARISMATIC LEADERSHIP PROCESSES BENJAMIN M. GALVIN University of Washington, Bothell PRASAD BALKUNDI University at Buffalo DAVID A. WALDMAN Arizona State University By integrating a social networks perspective with charismatic leadership theory, we attempt to explain how charismatic perceptions of leaders spread among distant followers through third-party individuals (surrogates). We propose a mediated model considering how individuals engage in surrogate behavior (promoting the leader, defending the leader, and modeling followership) that increases charismatic percep- tions among distant followers. We consider the behaviors associated with this role, as well as antecedents and moderators that may explain motivation to engage in the role and effectiveness of the role in influencing followers. Charismatic leadershipis a term often used to characterize extraordinary forms of influence, and it is frequently associated with leaders who are perceived as exceptional, gifted, and even heroic (Bass, 1990; Conger, 1989). Scholars com- monly conceptually such leadership as a rela- tionship between leaders and subordinates involving “a unique bonding,” emotional attach- ment, respect, and trust (Avolio & Yammarino, 2002: xvii). Subordinates who view a leader as charismatic will tend to see the leader and his or her vision as core to their own identity and will be willing to sacrifice so as to benefit the collective and achieve the goals and vision of the leader (Howell & Shamir, 2005).

Implicit leadership and cognitive categoriza- tion theories explain that individuals create leader prototypes in their mental schema and evaluate behaviors and actions of leaders ac- cording to those prototypes (Lord & Brown, 2001; Lord & Maher, 1991; Shamir, 1992). It follows thatcharismatic leadership is based on highly favor- able perceived attributes and personal charac- teristics of a leader, such as confidence, power, success, and influence (Burns, 1978; Jacobsen & House, 2001). Individuals in leadership positions have the ability to enhance follower perceptions of their charisma by engaging in behaviors (e.g., symbolic acts, speeches/statements, and direct interactions) that amplify such attributions in accordance with follower schemas and proto- types (Bass, 1985, 1990). Such influence is based on direct contact between leaders and subordi- nates. We define direct contact as interactions via rich channels of communication, which al- low for a maximum amount of information to be transmitted through dialogue between the two parties (Daft, 2008; Lengel & Daft, 1988). For ex- ample, face-to-face communication provides a high degree of channel richness since the two parties get instantaneous feedback from each other when they interact. In addition to the ac- tual words exchanged, the facial expressions and the vocal intonations provide contextual in- formation not available through other mediums of communication, such as a written memo (Daft, 2008; Lengel & Daft, 1988).

However, many leaders do not have direct contact with all of the subordinates under their authority. This lack of direct contact between We thank Jeff LePine and three anonymous reviewers for the significant time and effort they contributed to ensure that this article met its potential. We also thank Anne Tsui for her guidance and encouragement during the writing of early drafts of the manuscript. We further acknowledge Mar- tin Kilduff for his insightful comments regarding an early draft.

Academy of Management Review 2010, Vol. 35, No. 3, 477–494. 477 Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or email articles for individual use only. leaders and subordinates in organizations gen- erally can be attributed to time constraints, geo- graphic distance, or simply the distance created by the size or extensive hierarchy of organiza- tions. Without substantial direct contact be- tween a leader and subordinates, there may be a void of leader behaviors and actions for sub- ordinates to evaluate and compare against pro- totypes in their mental schema. A lack of posi- tive information about the leader will likely make it difficult for subordinates to develop charismatic perceptions (Bass, 1990). Yet it is not uncommon for subordinates who have had little direct contact with a leader to perceive the dis- tant leader as charismatic. This raises the ques- tion of why some subordinates, who have had little direct contact with their leaders, perceive their leaders as charismatic.

One explanation may be that some individu- als inherently romanticize leaders and tend to associate greatness and charisma with individ- uals in positions of power (Meindl, Ehrlich, & Dukerich, 1985). Another explanation may be that some leaders successfully engage in efforts that minimize the distance between themselves and subordinates. For instance, leaders may hold company-wide events to build support for their directives and initiatives. They may even disseminate their bios and photos in an attempt to personalize themselves and help subordi- nates see them in a positive light. Further, they may share video addresses and written state- ments via email, webpages, and blogs. In short, leaders may provide positive information about themselves, which followers can then use to form their own perceptions by making compari- sons against schema and prototypes of effective leadership.

Attempts to garner charismatic perceptions such as those discussed above may be some- what successful in overcoming distance be- tween leaders and distant followers. However, such attempts represent only a portion of the information that distant followers consider when developing their perceptions of leaders.

Indeed, distant followers will also glean infor- mation about leaders from other individuals with whom they directly interact in the organi- zation. Individuals who act as secondhand sources of information may, in many cases, be seen as more reliable and trustworthy by distant followers than such formal sources as company websites and staged appearances from leaders.These secondhand sources of information about leaders have the potential to be especially in- fluential in determining distant followers’ per- ceptions of leaders—potentially even more so than firsthand interactions (Bowler & Brass, 2006; Gilovich, 1987).

To date, the role of these influential individu- als in the charismatic leadership process has lacked clarity. Theories of charismatic leader- ship generally have focused on leader behav- iors or social processes that occur between lead- ers and followers—that is, relationships. Klein and House (1995) and Gardner and Avolio (1998) suggested that followers may play an important role in charismatic leadership processes. Build- ing on this idea, Waldman and Yammarino (1999) proposed that leader storytelling and sym- bolic behavior increase the likelihood that fol- lowers will spread stories and sagas about the leader that may lead to perceptions of leader charisma.

Along related lines, Balkundi and Kilduff (2005) noted the importance of a networks per- spective in understanding how intermediaries from the leader’s network may dampen or en- hance leader influence. This perspective is in line with Bono and Anderson’s (2005) findings that managers who are perceived as transfor- mational tend to hold more central positions in the informal social networks of the organization, as do their direct reports. Accordingly, there is the possibility that managers who are perceived as transformational may exert some influence over distant subordinates through intermediar- ies, including their direct reports.

By drawing on charismatic leadership theory (Bass, 1985) and a social networks perspective (Balkundi & Harrison, 2006), we advance new ideas regarding how third-party individuals may increase levels of charismatic perceptions or attributions for a leader among distant sub- ordinates. Our work expands on previous theory and research suggesting that charismatic attri- butions of a leader develop among subordinates through social contagion processes among in- formal networks (e.g., Gardner & Avolio, 1998; Meindl et al., 1985; Pastor, Meindl, & Mayo, 2002).

Specifically, our theory provides an understand- ing of the behaviors these individuals engage in that enhance perceptions of leader charisma.

Below, we refer to these individuals assurro- gates.An understanding of this largely un- known role and associated behavior can pro- 478July Academy of Management Review vide new insight into a process that may be, in great part, responsible for leaders’ gaining in- fluence and being seen as charismatic by their distant followers.

By adopting a social networks approach, we clarify how surrogates in key network positions are especially motivated and influential in this role. In doing so we seek to respond to the largely unanswered call for more theory to ex- plain how networks and social processes in or- ganizations are related to leadership and its perception (Balkundi & Kilduff, 2005; Brass, Ga- laskiewicz, Greve, & Tsai, 2004). It is to the issue of how distant leaders can be aided by surro- gates that we now turn our attention.

A MEDIATED MODEL OF CHARISMATIC LEADERSHIP To date, little effort has been made to show the connection between perceptions of leaders and the social processes that occur among dis- tant followers in formal (i.e., formally specified work relationships) and informal (i.e., discre- tionary relationships) networks (Ibarra, 1993). In Figure 1 we propose a mediated model that in- volves behavior, performed by individuals we call “surrogates,” that enhances perceptions of leader charisma among distant followers in the surrogates’ formal and informal networks. This process of influencing attitudes and emotions is similar in nature to Bandura’s (1977) sociallearning theory, in that a socially mediated pro- cess occurs in which individuals develop their perceptions (of leaders) based on their observa- tions of the attitudes and behaviors of others (surrogates). We now proceed to describe the nature of surrogates and surrogacy as a behav- ioral construct.

The Nature of Surrogates “Surrogate” is derived from the Latin termsur- rogatus, the past participle ofsurrogare,which means to substitute (American Heritage Dictio- nary). It is generally used to designate two types of individuals who stand in the place of, or sub- stitute for, another person in a social or family role (American Heritage Dictionary). First, it is used to describe a formal role, where “one ap- pointed to act in the place of another” (Merriam- Webster Dictionary) does so in an official man- ner. Second, it is used more informally to designate “one that serves as a substitute” (Mer- riam-Webster Dictionary) for another person. For instance, a formally appointed surrogate court may distribute assets for a deceased individual because the individual cannot distribute them him/herself. However, other examples of the use of the term may include both individuals who are intentionally sent to act in the role and in- dividuals who represent another of their own accord. We will return to the issue of intention- ality below.

FIGURE 1 Model of Surrogate Behavior Positive social exchange relationship with the leader Perceptions of leader charisma among distant subordinates Surrogate’s network proximit y to leader Surrogate behavior Promoting the leader Defending the leader Modeling followership Strength of existing subordinate perceptions of the leader Perceptions of leader charisma Surrogate’s network position Prestige Brokerage Core/periphery Surrogate behavior Moderators of the in uence surrogate behavior Moderators of antecedents of surrogate behavior Antecedents of surrogate behavior Outcomes of surrogate behaviors of 2010479 Galvin, Balkundi, and Waldman In the political realm the termpolitical surro- gateis used to describe well-known or influen- tial individuals who build support and credibil- ity for politicians (Sack, 1989). These individuals promote the politician; defend his or her posi- tions, actions, and/or reputation; and model how to be a supportive follower. Political surrogates may be intentionally sent by politicians or, al- ternatively, may simply view the politician in a positive light and consequently support the pol- itician of their own volition. During the 2008 presidential campaign in the United States, for instance, surrogates of Barack Obama, such as Oprah Winfrey, promoted Barack Obama among specific interest groups that may have been hes- itant to support him (e.g., female voters; Hallo- ran, 2007). Surrogates such as Clifford Alex- ander, former Secretary of the Army, defended the ability and credibility of Barack Obama when his (military-relevant) credentials were challenged, thus helping him to be perceived as a capable future Commander-in-Chief (Montan- aro, 2008). In addition, Hillary Clinton, who was Barack Obama’s primary competitor in the Dem- ocratic primary, eventually acted as a surrogate for him. Specifically, she modeled for others an example of committed followership to Barack Obama by campaigning for him and by publicly announcing her vote for him (Fitzgerald, 2008).

We employ the termsurrogatein a manner similar to the above usages to describe a role that individuals fill by actively engaging in non- coercive impression management behavior that facilitates a positive image of a leader (cf. Bo- lino, Kacmar, Turnley, & Gilstrap, 2008). This behavior—surrogate behavior—may take the form of promoting the leader, defending the leader, and/or providing a model of followership for others. Leaders at all levels of organizations who have followers with whom they have little contact because of time constraints, hierarchical/or- ganizational distance, or geographic distance may benefit when individuals engage in this role. Therefore, when we refer to the leader, we are not necessarily referring to the highest-level leader (e.g., CEO) of the organization. That is, leaders at lower levels (e.g., divisional manag- ers) who have distant followers may also re- ceive benefits when individuals engage in sur- rogate behavior.

Individuals who engage in the surrogate role may intentionally be sent by the leader, or they may act of their own volition. For example, aleader may intentionally suggest to would-be surrogates that the leader’s vision should be spread and characterized in a favorable light to distant followers. Alternatively, an individual may engage in surrogate behavior without prompting from the leader. Direct contact with the leader is not a prerequisite for engaging in surrogate behavior. Surrogates’ level of direct contact with the leader can range from high (e.g., an assistant manager who meets with the leader daily to discuss strategy) to zero (e.g., a distant subordinate who has never seen the leader in person). We do not limit the surrogate role to the direct reports of leaders. In the dis- cussion that follows we present a model of sur- rogate behavior and its effects on distant follow- ers’ perceptions of leader charisma.

Surrogate Behavior As a Construct As shown in Figure 1, surrogate behavior is central to an understanding of the surrogacy phenomenon in relation to charismatic leader- ship. Surrogate behavior is a latent construct that may be manifested through promoting the leader, defending the leader, and modeling fol- lowership. Surrogate behavior may be observed through one or all of these behaviors. These three behavioral dimensions may be engaged in simultaneously or independently, depending on the opportunities to engage in the behavior. Al- though these three behavioral dimensions are not dependent on one another, they manifest from a common cause (i.e., surrogacy) and likely complement one another. Thus, we conceptual- ize the behavioral dimensions as reflective of surrogate behavior, and as such, the behavioral categories are reflective and not formative in nature (MacKenzie, Podsakoff, & Jarvis, 2005).

The characteristic behaviors of the surrogate role mirror behaviors that have been tradition- ally associated with assertive and defensive im- age presentation/construction in the impression management literature (Goffman, 1959; Roberts, 2005). However, surrogate behavior focuses on managing others’ impressions of the leader, rather than of the self. Surrogate behavior fits within the broad category of interpersonal citi- zenship behavior, or OCBI (Bowler & Brass, 2006; Organ, 1997). This category captures citizenship behaviors that specifically benefit an individual—in this case the leader—and, in the long run, have the potential to positively influence the organi- 480July Academy of Management Review zation (LePine, Erez, & Johnson, 2002; Williams & Anderson, 1991).

There is no specific temporal or behavioral threshold for one to be considered as acting, or having acted, as a surrogate. For example, an individual who has engaged in characteristic behavior during a meeting could be considered as having acted as a surrogate, as well as an individual who consistently engages in charac- teristic behavior over time. In addition, there is no intentionality threshold to be considered a surrogate. That is, it is possible that in some cases an individual will engage in the behavior without the cognizant intention of generating charismatic leadership perceptions. Rather, the behavior may be uncalculated in nature. For instance, an individual may simply tell positive stories about the leader, with no intention of increasing charismatic perceptions of that indi- vidual among those to whom he or she is telling the stories.

As mentioned previously, surrogate behavior is manifested through promoting the leader, de- fending the leader, and modeling followership.

Leader promotion publicizes the accomplish- ments, positive qualities, and traits of the leader (cf. self-promotion and enhancement; Bolino et al., 2008; Roberts, 2005). Leader promotion largely consists of making positive statements and engaging in positive storytelling about the leader. Stories provide a forum for sharing the symbolic behaviors of the leader (Waldman & Yammarino, 1999) and are easily remembered and passed through the networks (Hatch, Kos- tera, & Kozminski, 2006). They are likely to in- clude instances of the leader’s overcoming chal- lenges or acting for the greater good of the organization, or high levels of leader perfor- mance. Storytelling can spread examples of the leader’s positive actions in supporting and working with an employee, or it can publicize accounts of the leader’s actions that have had a positive impact on the overall organization or that teach important lessons (Hatch et al., 2006).

Another aspect of leader promotion is making positive direct statements about the leader.

Such behavior includes making positive assess- ments of the leader’s performance and inten- tions, making positive statements about the leader’s personality and ability, and sharing the leader’s vision (Gardner & Avolio, 1998).

Defense behaviors by surrogates provide ex- planations, justifications, and excuses forleader actions and behaviors that may have been or may be interpreted in a negative man- ner by others (cf. defensive impression manage- ment; Bolino et al., 2008; Schlenker, 1980). De- fense behaviors represent attempts by the surrogate to mitigate negative feelings that may emerge among distant followers as the leader makes decisions or takes actions that might lack widespread support, such as decisions involv- ing significant change. For example, in the case of a layoff, followers may harshly criticize the leader and the decision.

Defense of the leader may also include pro- viding detailed justifications for leader actions or the leader’s perspective. In instances where there is high contact between leader and surro- gate, it might involve sensegiving on the part of the surrogate, who provides background infor- mation or detailed information regarding the process by which a decision was made. It may also include describing challenges the leader must react to (e.g., a budget cut or dropping revenue), or even attempts to get distant follow- ers to take a bigger-picture perspective so as to create an increased understanding regarding the leader’s actions.

Furthermore, defense behaviors may involve downplaying or even concealing negative infor- mation about the leader, such as leader defi- ciencies or shortcomings, from distant followers (Goffman, 1959). In sum, while leader promotion illuminates positive information about the leader, defense behaviors dampen the potential effects of damaging information (accurate or not) about the leader.

Finally, surrogate behavior may also involve modeling followership. Modeling followership is a visible representation for others of an ap- propriate response to or interaction with the leader (Bandura, 1977; Goffman, 1959). Modeling helps distant followers develop their percep- tions of the leader based on observations of the attitudes and behavior of the surrogates (Ban- dura, 1977). It provides social cues for these dis- tant followers, facilitating the leader’s appear- ance as influential, friendly, and worthy of following. Modeling has the potential to take place in both formal and informal settings, as well as when the leader is or is not present.

Furthermore, modeling occurs as individuals publicly pledge their participation and commit- ment to a leader’s initiative or vision. Thus, it can take place when an individual responds 2010481 Galvin, Balkundi, and Waldman positively to a leader’s request without compul- sion. It can also take place when an individual demonstrates his or her willingness to contrib- ute time or effort on behalf of the leader. Mod- eling can further occur when a leader has the chance to interact directly with distant followers with whom he or she has had little interaction in the past. Modeling may also take the form of such behaviors as a friendly reaction to the leader’s greeting in a crowded elevator, or it may involve making small talk with the leader before a meeting in the presence of others who do not know the leader well. Such responses portray the leader as friendly and well liked by others and help other distant followers feel at ease around the leader.

Proposition 1: Promoting the leader, defending the leader, and modeling followership are behaviors that reflect surrogacy.

Antecedents to Surrogate Behavior In order to understand the surrogate phenom- enon better, it is important to recognize potential antecedents to such behavior. Below we articu- late these antecedents and their relationships to surrogate behavior, as well as a likely modera- tor of these relationships.

Perceptions of leader charisma.Individuals who view a leader as charismatic will tend to intertwine their own identities with the leader’s identity, thus aligning their values, interests, and goals with those of the leader (Gardner & Avolio, 1998). The outcome of this type of align- ment is that the individuals’ motives will shift from self-interest to support for the leader and the vision or cause the leader promotes (Gard- ner & Avolio, 1998). Individuals who perceive the leader as charismatic will tend to engage in behaviors they perceive will benefit the leader and the leader’s vision, and they will be moti- vated to continue to build support for the leader through their identification with the leader’s vi- sion and through their emotional attachment to the leader (Howell & Shamir, 2005). These indi- viduals may have substantial contact with the leader on a daily basis, or they may have lim- ited contact with the leader (Baker & Dutton, 2007). It is even possible that they may have had no direct contact with the leader at all and, instead, have gained their perceptions of theleader from more indirect or secondhand sources. Such sources could include documents written by the leader, or even communication with other individuals who have had direct con- tact with the leader.

Engagement in surrogate behavior may re- flect a natural outgrowth of perceiving a leader as charismatic. It is likely that some individuals will engage in surrogate behavior without the intention of doing so or the cognizant knowledge (on the part of either the leader or the surrogate) that their behavior is building charismatic per- ceptions of the leader. These individuals may simply be enamored with the leader and not realize that they are promoting and defending the leader and, thus, acting as committed fol- lowers. Such individuals can be compared to highly satisfied customers who spread the word about the good service and food at a particular restaurant. Truett Cathy, the founder of the Chick-fil-A restaurant chain, has referred to these types of loyal customers as cheerleaders (Cathy, 2002). Moreover, individuals may engage in surrogate behavior because of a genuine de- sire to ensure that others perceive the leader accurately, who, according to their own percep- tions, is charismatic.

Proposition 2: Perceptions of leader charisma are positively related to an individual’s participation in surrogate behavior.

Perceptions of a positive social exchange.In- dividuals who perceive a positive social ex- change relationship with the leader will also tend to engage in surrogate behavior. We use the term socialexchange relationshiphere since it broadly encompasses exchange relationships that involve direct contact (e.g., leader-member exchange), as well as more abstract exchanges that do not require high levels of contact (e.g., perceived leader support, fulfilled psychologi- cal contract with the leader, complementary agendas). It is likely that individuals will not see exchange relationships in exactly the same manner as the leader, since perceptions of ex- change relationships tend to differ between in- dividuals in an exchange relationship (Gerstner & Day, 1997). This is especially likely for individ- uals who have little direct interaction with the leader, and, thus, the social exchange is based on less tangible benefits and exchanges. Indi- viduals will be motivated by their perceptions of 482July Academy of Management Review the relationship, rather than the actual ex- change that takes place.

Exchange relationships can be characterized using a continuum ranging from generalized reciprocity to negative reciprocity, with bal- anced reciprocity in the middle (Sahlins, 1972; Sparrowe & Liden, 1997). Generalized reciprocity is altruistic in nature, and not focused on the specific contributions or obligations of the other individual in the relationship (Sparrowe & Li- den, 1997). Balanced reciprocity is more focused on the exchange of equivalent benefits or favors and the assurance of a level of fairness and equality of contributions. Positive exchanges are likely to be based on either generalized or balanced reciprocity. In contrast, negative ex- changes tend to be based on self-interest and on taking advantage of the other in the exchange (Sahlins, 1972; Sparrowe & Liden, 1997). We focus our discussion here on follower perceptions of positive exchanges based on generalized and balanced reciprocity, since we are interested in how follower perceptions of a positive exchange relationship result in the individual’s engaging in behavior that benefits the leader. Later dis- cussion will touch on the implications of a neg- ative exchange relationship.

Individuals who perceive that they are in a positive exchange relationship may feel a de- sire and/or sense of obligation to reciprocate the leader’s influence, which they believe they have benefited from in terms of preferential treatment (e.g., obtaining perquisites), policy changes, ad- vice, and resources (Sparrowe & Liden, 2005). In order to reciprocate, individuals may provide high contributions and commitment (Sparrowe & Liden, 2005; Uhl-Bien & Maslyn, 2003) and “the expression of public support for the goals and the personal character” of the leader (Dienesch & Liden, 1986: 625). These public expressions of support are likely to include promoting the leader through positive stories and making af- firmative statements, defending the leader, and modeling followership— behaviors that we as- sociate with the surrogate role. Individuals who perceive they are in a positive exchange rela- tionship based on balanced reciprocity are likely to engage in surrogate behavior out of a sense of obligation or with the intent to glean additional benefits, rather than from altruistic desires. In contrast, individuals who perceive that they are in a positive exchange relationship that is based on generalized reciprocity will en-gage in the above behavior from an altruistic desire to support the leader.

Relationships based on generalized reciproc- ity, particularly those with direct contact be- tween the individual and the leader, are likely to include sponsorship from the leader (Spar- rowe & Liden, 1997, 2005). Sponsorship is char- acterized by the leader’s introducing the indi- vidual into his or her networks of relationships with influential others, resulting in shared ties with the leader (Sparrowe & Liden, 1997, 2005).

Individuals who receive sponsorship from a leader as part of a generalized exchange rela- tionship are likely to engage in surrogate be- havior. This exchange simultaneously provides benefits to the leader and the individual, and it can be seen as a form of positive exchange.

Sponsorship from the leader provides the indi- vidual with enhanced status and reputation in the organization (Bono & Anderson, 2005; Kilduff & Krackhardt, 1994), particularly if the leader is influential, and it may be another mechanism “whereby individuals are integrated into orga- nizations as influential players” (Sparrowe & Liden, 2005: 506).

Individuals who perceive that they are in a positive exchange relationship based on bal- anced reciprocity are more likely to engage in surrogate behavior with the intent to garner benefits. These benefits may include special fa- vors from the leader (e.g., a promotion or raise) or affiliation with the leader, which can increase one’s influence and status in the organization (Cialdini & Richardson, 1980). As part of positive exchange relationships, particularly those based on balanced reciprocity, individuals may be asked by the leader to engage in surrogate behaviors and to represent the leader positively among distant subordinates. This is most likely to occur when a leader perceives that he or she is in a positive exchange relationship with an individual who is an opinion leader among dis- tant followers, and there is a sense of trust and/or level of comfort in asking for specific favors (Burt, 1999). However, in order for the lead- er’s request to result in the individual’s engag- ing in surrogate behavior, the individual who is asked to engage in the behavior will need to perceive the exchange relationship in a simi- larly positive light.

This scenario represents an example of a leader who intentionally uses individuals as sources of influence to get things done or to 2010483 Galvin, Balkundi, and Waldman facilitate action (Brass & Krackhardt, 1999; Mintz- berg, 1973). In certain cases leaders may even have an individual transferred to a certain part of the organization to enhance their own influ- ence among certain networks of subordinates.

While surrogate behavior is focused on increas- ing charismatic perceptions of the leader, the leader may also ask the individual to report back and provide information regarding how things are going and “crucial information nec- essary” to successfully complete a project or ini- tiative (Sparrowe & Liden, 1997: 524). Individuals may even confidentially report troublemakers or discuss problems that distant followers face.

Proposition 3: Perceptions of a positive exchange relationship (with the leader) based on generalized or bal- anced reciprocity are positively re- lated to an individual’s engaging in surrogate behavior.

Moderating effects of network proximity to the leader.Individuals who are closer to the leader in a network sense (i.e., network proximity) are more likely to be exposed to signals (both direct and ambient) from the leader (e.g., organization- wide speeches, team meetings, and one-on-one discussions) that are aimed at engaging those individuals in surrogate behavior. Individuals who perceive the leader to be charismatic and/or who are in a positive exchange relation- ship with the leader are likely to feel that they should respond to these calls to action. That is, they will likely feel a sense of responsibility to respond to the leader. They will tend to believe that because of their high level of contact with the leader and shared network ties, their level of engagement in surrogate behavior is more ap- parent to the leader. The increased level of ac- countability and transparency associated with network proximity is likely to strengthen the re- lationship between perceptions of leader cha- risma and surrogate behavior and between per- ceptions of a positive exchange relationship and surrogate behavior.

Proposition 4a: An individual’s net- work proximity to the leader will mod- erate the relationship between per- ceptions of leader charisma and participation in surrogate behavior.

Specifically, there will be a stronger positive relationship between percep-tions of leader charisma and partici- pation in surrogate behavior when the individual is proximal to the leader in the leader’s network.

Proposition 4b: An individual’s net- work proximity to the leader will mod- erate the relationship between a pos- itive exchange relationship and participation in surrogate behavior.

Specifically, there will be a stronger positive relationship between a posi- tive exchange relationship and partic- ipation in surrogate behavior when the individual is proximal to the leader in the leader’s network.

Surrogate Behavior and Follower Perceptions Surrogate behavior is likely to influence dis- tant followers’ perceptions of the leader. Such behavior is influential because of the process outlined in Bandura’s (1977) social learning the- ory: individuals develop perceptions based on their observations of the attitudes and behav- iors of others. Observations of the views and attitudes of others toward the leader provide a form of social proof that individuals can rely on in forming their own perceptions of the leader (Cialdini, 1993).

Surrogate behavior helps guide the process distant followers engage in to form their percep- tions of the leader by providing opinions, infor- mation, and behavior that portray the leader in a positive light (Cialdini, 1993; Weick, 1995). Dis- tant followers are likely to base their percep- tions of a leader’s charisma on simplified views of whether or not the leader matches their pro- totypes of what a charismatic leader should be (Shamir, 1995). Surrogate behavior helps align distant follower perceptions of the leader with attributes generally associated with charis- matic leadership, such as confidence, success, and influence (Burns, 1978; Jacobsen & House, 2001).

Leader promotion provides positive stories and statements about the leader that portray the leader in a charismatic manner. Positive stories can result in a personalization of the leader that helps individuals in the organization feel that they know the leader and have a connection to his or her vision (Hatch et al., 2006). Stories, such as those surrounding Southwest’s Herb Kelle- 484July Academy of Management Review her’s support of his employees and his response to disrespectful customers to fly with someone else (Elliott, 2004), help distant subordinates make positive attributions about the leader.

Similarly, positive statements about the leader from individuals filling the surrogate role have the potential to be influential in the formation of distant follower views of the leader (Gardner & Avolio, 1998; Gilovich, 1987). They may influence distant follower perceptions of the leader’s per- sonality, values, ability, vision, performance, and other areas that are relevant to perceptions of charismatic leadership. When an individual hears another individual describe the leader in charismatic terms, such as gifted or inspiration- al, these endorsements are likely to influence the listener’s perceptions and to increase the likelihood the listener will see the leader as charismatic. Similarly, when an individual con- veys the leader’s message and vision to distant followers in an inspirational manner, distant followers will tend to view the leader as charis- matic (Gardner & Avolio, 1998).

However, negative information, stories, per- ceptions, and interpretations of the leader and his or her actions have the potential to reduce distant follower perceptions of leader charisma.

Thus, when surrogates engage in defending the leader, they will clarify misconceptions and dampen potentially harmful information that might damage the leader’s reputation. Through the act of defending the leader, surrogates can increase distant follower perceptions of leader charisma. By providing explanations for contro- versial leader actions, there is a reduced likeli- hood that distant followers will perceive the leader in a negative light. Surrogates, by de- fending the leader, create understanding and provide information regarding decisions or pol- icies that the leader has enacted, with the result that distant followers will be more likely to trust the leader and, thus, see the leader as charis- matic.

Modeling followership also increases distant followers’ perceptions of leader charisma. When a surrogate models followership by standing in support of the leader, his or her behaviors are likely to positively influence the way that others perceive that leader. Individuals will see the leader as more influential and worthy of follow- ing when others whom they respect follow the leader. When an individual publicly demon- strates trust for the leader, distant followers willbe more trusting of the leader and less skeptical of his or her initiatives, thus increasing the lead- er’s legitimacy. These interactions set the norm for charisma as the basis of influence, rather than influence based on coercion or extrinsic rewards (Bass, 1990).

Additionally, when a surrogate offers a friendly response or reaction to a greeting from the leader in public, the leader is portrayed as friendly and conscientious. This creates positive perceptions among more distant followers that the leader is approachable and interested in all followers (close and distant). Modeling follower- ship by the individual filling the surrogate role provides a way for a leader to break down per- ceptions of distance between that leader and distant followers, thus setting norms that facili- tate charismatic attributions. In total, these ar- guments suggest the following.

Proposition 5: Surrogate behavior (promoting the leader, defending the leader, and modeling followership) is positively related to distant follower perceptions of leader charisma.

The Moderating Effect of Network Position Although surrogate behavior will likely result in increased charismatic perceptions of the leader on the part of distant followers, it is im- portant to understand factors that may moderate this relationship. Individuals who are in key network positions will be better able to make stories and ideas about a leader contagious (Burt, 1999; Mayo & Pastor, 2007). That is, they will be better able to access, disperse, and con- trol information throughout the network. Below we outline these relevant network positions.

Prestige.The first network position of rele- vance to surrogacy is network prestige—the ex- tent to which an individual is sought out for advice and friendship by others. Network re- searchers refer to this as in-degree, which is the number of nominations received by an individ- ual (Wasserman & Faust, 1994). This network measure assesses prestige, capturing the extent to which one is popular and visible in the net- work (Wasserman & Faust, 1994). Individuals who are well liked and consistently sought out for advice by others will have positive attribu- tions made about them. These individuals tend to be respected sources of information in their 2010485 Galvin, Balkundi, and Waldman informal social networks (Balkundi, Barsness, & Michael, 2009; French & Raven, 1959).

Prestige in informal networks allows surro- gates to influence the opinions of and attribu- tions made by distant followers (e.g., Newman, 2005). Surrogates who have high prestige in sub- ordinate networks will have a high level of di- rect contact with distant subordinates who seek them out for friendship and advice, which can lead to greater opportunities to engage in surro- gate behavior (e.g., Mullen, Johnson, & Salas, 1991). The more a surrogate is sought out for advice or is liked by followers distant from the leader, the greater the surrogate’s ability to talk about and transmit charismatic attributions of the leader (Mayo & Pastor, 2007). In other words, surrogate behavior can be especially effective when the individual engaging in the behavior is someone who is perceived as prestigious, be- cause the surrogate is already seen as a credi- ble and reliable source of information (Lam & Schaubroeck, 2000; Mayer, Davis, & Schoorman, 1995; Thomas & Griffin, 1989).

Proposition 6a: An individual’s pres- tige in subordinate networks moder- ates the relationship between partici- pation in surrogate behavior and charismatic perceptions of the leader on the part of distant followers. Spe- cifically, a high level of prestige will enhance the relationship, whereas a low level of prestige will weaken the relationship.

Brokerage.Brokerage between the leader and distant subordinates is also relevant to surro- gate effectiveness. Brokerage exists when an individual is socially tied to two unconnected others (e.g., the leader and distant followers). A brokerage position gives surrogates access to information about the leader and the opportu- nity to convey this unique information to others (Burt, 2004). Surrogates in high-brokerage posi- tions in subordinate networks are likely to emerge as established sources of information about the leader. They may also be opinion leaders in the subordinate networks, since they tend to control and coordinate information shar- ing (Burt, 1999; Mullen et al., 1991). Having both access to the leader and control of information in the networks facilitates the effectiveness of surrogate behavior because it increases a sur-rogate’s ability to influence opinions and atti- tudes about the leader (Mayo & Pastor, 2007).

Opportunities to broker may decay over time if distant followers perceive that a surrogate is no longer a reliable source of information. Ac- cordingly, they may bypass the surrogate, thus reducing the surrogate’s level of brokerage (Burt, 2007). In low-brokerage settings, distant followers may circumvent the surrogate as the basis of information about the leader. Further, these distant followers may observe distortions between the messages they receive from the surrogate and those they glean from alternate sources. Surrogates who minimize perceptions of information distortion will tend to be per- ceived as credible, thus maintaining the broker- age role and the ability to generate and sustain perceptions of leader charisma through surro- gate behavior. Overall, these arguments sug- gest the following.

Proposition 6b: An individual’s level of brokerage in the subordinate networks moderates the relationship between participation in surrogate behavior and distant follower percep- tions of leader charisma. Specifically, a high level of brokerage will en- hance the relationship, whereas a low level of brokerage will weaken the re- lationship.

Core periphery.In the social networks litera- ture the concept of core periphery, a type of network centrality, captures the extent to which a node (i.e., individual) is connected to a dense set of nodes (i.e., individuals) in a social network (Borgatti & Everett, 1999). It assesses the ex- tended influence of a surrogate who is tied to the few central members in a dense network (e.g., a clique of friends). Being tied to these dense clusters allows surrogates to partake of the benefits associated with mutual friendships, ingroup status, and shared identities, such as perceived integrity and liking (Ashforth & Mael, 1989; Heider, 1958). Surrogates who are tied to these dense clusters of relationships will be per- ceived as sharing the norms, principles, and assumptions of the group, allowing for percep- tions of trustworthiness among the distant sub- ordinates (Mayer et al., 1995).

This ingroup status is likely to be associated with influence within the network, even among those with whom direct ties do not exist. In con- 486July Academy of Management Review trast, a surrogate who is not connected to dense cliques in subordinate networks (i.e., is peripheral in the networks of subordinates) may struggle to influence perceptions of the leader among distant followers. Thus, unlike prestige and brokerage, the core periphery po- sition captures influence beyond the surro- gate’s direct ties.

Proposition 6c: An individual’s core periphery position moderates the rela- tionship between participation in sur- rogate behavior and distant follower perceptions of leader charisma. Spe- cifically, a core position will enhance the relationship, whereas a peripheral position in the subordinate network will weaken the relationship.

The Moderating Effect of Existing Perceptions of the Leader Surrogate behavior does not guarantee that a leader will be perceived as highly charismatic by distant followers. In addition to the surro- gate’s network position, preexisting distant fol- lower perceptions of a leader can come into play in determining how influential surrogate behav- ior is (e.g., Pastor, Mayo, & Shamir, 2007; Shamir, Pillai, Bligh, & Uhl-Bien, 2007). When individuals are already convinced and certain of their per- ceptions of the distant leader, surrogate behav- ior is less likely to be influential in determining subordinate perceptions of the leader. Followers are likely to compare the surrogate’s ideas about the leader with their own preexisting per- ceptions in developing subsequent or modified perceptions of the leader (Mayo & Pastor, 2007).

In other words, although surrogate behavior can be highly influential, such influence is tempered when subordinates have crystallized their per- ceptions of a leader.

Proposition 7: Distant followers’ exist- ing perceptions of the leader act as a moderator of the relationship between surrogate behavior and subsequent distant follower perceptions of leader charisma. Specifically, well-estab- lished perceptions of the leader will weaken the relationship.DISCUSSION An examination of the surrogate role expands our understanding of charismatic leadership, followership, and networks. Specifically, it facil- itates an understanding of how a leader and followers (whom we label surrogates) may work together (intentionally or not) to transmit posi- tive information about the leader. The subse- quent effect may be a heightening of percep- tions of charismatic leadership on the part of distant followers, which, in turn, can help estab- lish and maintain widespread leader support, power, and influence. This perspective sheds new light on the complexity of charismatic lead- ership and underscores the idea that charis- matic leadership cannot be fully understood by looking solely at the leader’s behavior.

In short, our concepts and ideas highlight and clarify the role of third-party individuals and the social contagion processes within informal net- works that influence perceptions of leader cha- risma (Gardner & Avolio, 1998; Meindl et al., 1985; Pastor et al., 2002). Specifically, our theory provides an understanding of the behaviors sur- rogates engage in that enhance perceptions of leader charisma. It also provides insight into the motivation to engage in a surrogate role, as well as factors that may influence surrogate effec- tiveness. Ultimately, our theory adds clarity to how networks and social processes in organiza- tions are related to the widespread formation of charismatic leadership perceptions.

Theoretical Extensions This integration of social networks with surro- gate behavior pushes social network theory on two different fronts. First, it helps us address the question “What are the mechanisms by which social networks influence leadership pro- cesses?” Our model proposes that specific sur- rogate behavior exhibited by central network members may be a basis of distant subordi- nates’ attitudes toward a leader. Second, this article builds on a growing body of literature that suggests that both social networks and in- dividual-level attributes need to be taken into account when studying individuals (e.g., Mehra, Kilduff, & Brass, 2001; Toegel, Anand, & Kilduff, 2007). An interactional approach of this nature provides a more balanced view of individuals as not being oversocialized by a network struc- 2010487 Galvin, Balkundi, and Waldman ture that exclusively determines outcomes, re- gardless of individuals (e.g., Mayhew, 1980), nor undersocialized, whereby only an individual’s actions matter in determining outcomes (Granovetter, 1985). With that in mind, we now examine some potential extensions to our theory in terms of person, situation, time, external per- ceptions, and self-interest.

Person.Future research might explore a deeper understanding of what motivates indi- viduals to engage in the surrogate role, as well as which types of personalities are especially relevant to such a role. It may also be important to determine which distant follower personali- ties are likely to be susceptible to surrogate influence. For instance, are followers with a weak self-concept, low self-esteem, and low self-efficacy especially influenced by surro- gates? It may be that personality factors play a similar role in determining a follower’s suscep- tibility to surrogate influence in a manner com- parable to what has been more generally de- scribed in predicting susceptibility to charismatic leaders (e.g., Gardner & Avolio, 1998; Howell & Shamir, 2005).

Situation.Surrogates may be most likely to emerge (intentionally or unintentionally) under circumstances especially conducive to charis- matic leadership, such as high uncertainty or change (Shamir & Howell, 1999; Waldman, Ramirez, House, & Puranam, 2001). Surrogates also may be likely to emerge or have more in- fluence during crisis conditions, such as layoffs or tragedies affecting the organization (e.g., a major industrial accident). Leaders facing these conditions may especially feel the need for sup- port from distant followers in order to achieve their goals or purposes and, hence, be more likely to engage surrogates. Likewise, in such circumstances distant followers may seek to better understand the leader and his or her mo- tives and tendencies and, thus, be likely to seek out the thoughts and opinions of surrogates.

Surrogates may also be especially effective in situations where dense networks exist. In dense networks with multiple surrogates, a social con- tagion process is likely to occur (Meindl, 1990).

As individuals in the distant follower networks are motivated by their perceptions of leader charisma and the inspirational ideas and vi- sions that have been shared through the surro- gates, additional people will likely then share these ideas and visions. Although not specifi-cally characterized in network terms, similar processes have been conceived in terms of the ability to “blossom to impact more and more followers” (Gardner & Avolio, 1998: 52) and “spread like a contagious disease” (Gardner & Avolio, 1998: 51).

Time.How does time influence this process?

Over time, individuals engaging in surrogate behavior may continuously reassess their rela- tionship with the leader. In the case where sur- rogates have been sent intentionally, perhaps leaders identify and then train and maintain these individuals. Moreover, leaders may reas- sess the efforts of surrogates and take measures to deal with underperforming surrogates. Over time, as trust is garnered, leaders may engage individuals in more important efforts aimed at increasing their influence in the organization (Brower, Schoorman, & Tan, 2000). Leaders may also need to find the appropriate balance be- tween depending on surrogates and making more attempts at face-to-face contact with dis- tant followers. Furthermore, surrogates who are sent (intentionally) by the leader may tend to be more consistent and effective over time, in com- parison to those who engage in surrogacy of their own volition. How leader intentionality in- fluences surrogate effectiveness still needs to be explored.

External perceptions.Surrogacy may also be relevant to understanding perceptions of CEOs by boards of directors and outsiders. Board members who are tied to the CEO may play surrogate roles by influencing perceptions of the CEO among new or distant board members. Fur- thermore, they may play a similar role with out- siders, such as potential investors, the media, analysts, and other companies (Fanelli, Mi- sangyi, & Tosi, 2009). Thus, surrogates may be an influential factor in determining external CEO reputation. Similarly, there may be individ- uals who are external to the organization who act in surrogate-like roles for a CEO, thus im- proving the CEO’s reputation and legitimacy among external stakeholders (Fanelli & Mi- sangyi, 2006).

Self-interest.Finally, individuals in leader- ship roles may intentionally use surrogates in a responsible manner, thus serving the interests of both the organization and stakeholders, in- cluding employees. Indeed, much of our theory would suggest that leaders can use surrogates for the purpose of spreading a positive vision.

488July Academy of Management Review Alternatively, leaders may pursue less mutually beneficial outcomes in a more dark and decep- tive manner, largely out of self-interest. The former is commonly referred to as socialized charisma, whereas the latter has been charac- terized as personalized charisma (House & Howell, 1992).

When intentionally used in a personalized or manipulative manner, surrogates could be used to cover up a leader’s unethical, abusive, or de- ceptive behaviors, which may ultimately have a negative impact on the organization and exter- nal stakeholders (Fanelli & Misangyi, 2006). For example, the personalized leader may try to curry favor with surrogates in an attempt to get them to relay positive attributions to those in the surrogates’ networks. Indeed, there is some rea- son to believe that personalized charismatic leaders may be more likely to engage surro- gates as a means of building their images (Gardner & Avolio, 1998). They may also encour- age surrogates to engage in less genuine be- haviors, such as helping to spread inaccurate or “spun” stories that enhance follower percep- tions of leader charisma.

In a political vein, leaders may seek to tap influential surrogates in diverse networks who can provide credibility with powerful interest groups. Such instances may be especially rele- vant for new leaders who seek to win over influ- ential individuals in diverse roles who can then provide the leader with legitimacy throughout the organization. For example, new leaders might attempt to use surrogates as they seek to win support from the network of a passed-over peer, who may not be initially supportive of the new leader. However, when engaging in such political processes, the personalized charis- matic leader may intentionally enlist surrogates to spread negative stories or perceptions of one or more individuals who are competing for in- fluence in the organization in an effort to weaken those individuals or diminish their in- fluence. Such a leader may even engage surro- gates in encouraging rebellion or insurgency in order to destabilize an organization and pro- mote the leader’s individual interests. Issues of this nature raise important questions regarding the surrogate role as it pertains to ethical lead- ership and leader authenticity (Avolio, Gardner, Walumbwa, Luthans, & May, 2004; Brown & Trevin˜ o, 2006).We should further note that while an under- standing of the surrogate aspect of followership may lend insight into the amplification of influ- ence and power for particular leaders, it also uncovers the potential for follower behaviors that undermine leader influence. It is possible for individuals to engage in behaviors that mir- ror the surrogate behavior but take on a nega- tive tone. That is, while individuals may have the ability to help increase the influence of a leader or power structure in the organization, they may also have the potential to subvert it.

However, when aware of these situations, lead- ers would likely act quickly to negate the influ- ence of individuals who engage in negative sur- rogate-like behaviors in order to preserve their own influence and reputation. It may also be possible for individuals acting in a surrogate role to undermine leader influence unintention- ally through incompetence or unintended conse- quences (e.g., unintentionally portraying the leader in an offensive light to an interest group).

Empirical Testing There are several possibilities for testing the ideas proposed in this paper. Empirical tests of this model can focus on the formal leader or on the surrogate and then extend to more complex issues. Social network analyses provide multi- ple measurement options. In this section we elaborate on some potential directions for test- ing our model by first focusing on the leader as the unit of analysis, and we then shift our focus to surrogates and the measurement of surrogate behavior.

To understand a leader’s connectedness with potential surrogates, the researcher can calcu- late the eigenvector centrality of the leader in the informal networks. This network measure captures the extent to which a network member is connected to powerful others (i.e., potential surrogates; Wasserman & Faust, 1994). Unlike the more intuitive measure of in-degree central- ity, which assesses the direct number of ties, eigenvector centrality takes into account both direct and indirect ties of network members to the leader. As suggested above, by being con- nected to powerful others, a leader can augment his or her social capital and influence by tap- ping into the resources and information that are accessible to these surrogates. In support of this idea, previous research has found that leaders 2010489 Galvin, Balkundi, and Waldman with high eigenvector centrality are seen as more charismatic by their subordinates and tend to have more productive teams (Balkundi, Kilduff, & Harrison, 2009). Thus, if the unit of analysis is the leader and the leader’s access to surrogates, then eigenvector centrality would be a relevant measure.

To assess leader intentionality, a researcher may ask the leader to identify those individuals he or she perceives to be influential in the orga- nization. Social network analysis can be used to measure the leader’s understanding and accu- racy of the social networks in the organization.

Researchers can administer a survey to explore questions regarding leader intentionality and engagement in particular behaviors. Surveys can also assess intentionality and types of leader interactions and communication with the individuals the leader perceives to be influen- tial. In addition to providing information regard- ing intentionality, this information can be used to identify the actions that leaders exhibit to engage followers in surrogate behaviors. Future research can explore the types of influence tac- tics that leaders use with their surrogates and how these tactics may vary across leaders.

There are multiple surrogate-relevant, net- work-based measures that can be used to iden- tify surrogates, independent of whether these individuals have contact with the leader. In pre- vious research on opinion leaders in social set- tings, researchers have used in-degree as a measure of influence (Rogers, 2003). In-degree— the number of people coming to the surrogate for advice or friendship—would capture the surro- gate’s popularity and influence with other sub- ordinates. However, to capture the brokerage role played by the surrogate, betweenness cen- trality would be a better measure, since it cap- tures the extent to which the surrogate is on the shortest path between other subordinates (Was- serman & Faust, 1994). Thus, a high between- ness centrality of the leader’s alters (i.e., mem- bers who are connected to the leader) would suggest the presence of surrogates surrounding the leader.

A surrogate’s social power can be amplified if he or she is connected to other subordinates, who themselves are also well connected with other subordinates or to dense clusters of sub- ordinates. To identify whether surrogates are connected to other powerful network members (and to other surrogates), researchers can useeigenvector centrality or the core periphery measure to capture the social power of the sur- rogates’ direct and indirect contacts to powerful actors in the network (other than the leader). In sum, if the unit of analysis is the surrogate, in-degree, betweenness, core periphery, and eigenvector centrality would all be relevant operationalizations. Because these four mea- sures are continuous and not categorical in nature (unless one chooses to dichotomize them), they provide fine-grained information about the varying levels of surrogacy.

In order to test whether these individuals act for or against the leader, a researcher can first identify possible surrogates using the different operationalizations mentioned above, and then determine if these individuals like, dislike, or are indifferent to the leader. Based on whether the surrogates nominate the leader in the advice or friendship network, the researcher can iden- tify the potential of positive surrogacy (Lincoln & Miller, 1979). However, if the leader is nominated by the surrogates in the dislike network, then the survey captures the potential for negative surrogacy, as described above. Thus, an individ- ual’s level of network centrality, based on the four measures above, along with the aforemen- tioned network information, captures the level of positive or negative influence that the individ- ual can potentially have on perceptions of the leader among distant followers.

We should further note that if the focus is exclusively on those surrogates with high con- tact with the leader, there are different ways of identifying them. The first step would be to iden- tify surrogates in the organizational network us- ing the centrality measures described above.

This would be a list of all of the potential surro- gates in the organization. Second, from this list the researcher can find the subset of surrogates who are directly connected to the leader. Alter- natively, if the researcher can get network data from the leader, he or she can identify members who are directly connected to the leader (the leader’s alters) and then develop different cen- trality measures for this subset of employees.

Either approach would identify the surrogates who are directly tied to the leader. The re- searcher can then compare this subset of surro- gates to other individuals and see if there are any attitudinal differences, which are similar to those exhibited by the members of the leader’s ingroup and outgroup.

490July Academy of Management Review One of the underlying assumptions in the us- age of these network measures is that individu- als who occupy these positions exhibit surro- gacy behaviors. To confirm this assumption, such research would need to be complemented by survey and/or observation data assessing en- gagement in surrogate behavior. Thus, it may also be necessary to develop and validate a measure of such behavior, which could poten- tially be self-reported or assessed by others. The measure would likely include multiple items for each of the dimensions mentioned earlier. If a self-report approach were adopted, items for promoting the leader could include “I tell posi- tive stories about the leader to others” and “I make positive statements about the leader in my conversations with others at work.” Items for defending the leader might include “I offer pos- sible explanations for the leader’s behavior when others challenge the leader’s actions” and “I provide excuses for the leader when people criticize his or her decisions.” Items for modeling followership could include “I demonstrate sup- port for the leader through my actions” and “When I interact with the leader in the presence of others, I assist the leader in appearing friendly and well liked.” The wording of these items could be adapted if individuals were be- ing assessed by others. Initial analyses could involve exploratory factor analyses, followed by confirmatory analyses. The goal would be to determine the uniqueness of the dimensions, as well as the possibility of a higher-level latent factor of surrogate behavior as conceptualized above.

Concluding Thoughts In conclusion, an understanding of the surro- gate role gives us important insights into the role of individuals besides the leader in influ- encing distant follower perceptions of the leader. Perhaps more important, it helps us un- derstand how leaders may use, or at least ben- efit from, networks, social processes, and flows of information in organizations. Specifically, a consideration of surrogacy provides insights into how networks and social processes pertain to a leader’s ability to influence the organization through the dispersion of information. The pro- cesses described here are much more subtle and pluralistic than the more dictatorial and mono- lithic strategy described by Burns (1978), inwhich a core of devoted followers overwhelms any resistance that might stand in opposition to the leader.

Nevertheless, we have described potentially powerful influences on the organization and the perceptions of followers. This understanding may provide greater clarity regarding why indi- viduals who have not had direct interaction with a leader may develop the same strong and en- during perceptions of the leader as those indi- viduals who have had substantial interaction.

At a more macro level, this understanding may provide insights into how large-scale move- ments that are associated with individual lead- ers are initiated and gain a critical mass before they spread throughout organizations, societies, and cultures.

REFERENCES Ashforth, B. E., & Mael, F. 1989. Social identity theory and the organization.Academy of Management Review,141: 20 – 39.

Avolio, B. J., Gardner, W. L., Walumbwa, F. O., Luthans, F., & May, D. R. 2004. Unlocking the mask: A look at the pro- cess by which authentic leaders impact follower atti- tudes and behaviors.Leadership Quarterly,15: 801– 823.

Avolio, B. J., & Yammarino, F. J. (Eds.). 2002.Transformational and charismatic leadership: The road ahead,vol 2. Am- sterdam: JAI-Elsevier Science.

Baker, W., & Dutton, J. E. 2007. Enabling positive social cap- ital in organizations. In J. E. Dutton & B. R. Ragins (Eds.), Exploring positive relationships at work: 325–345. Mah- wah, NJ: Lawrence Erlbaum Associates.

Balkundi, P., Barsness, Z. I., & Michael, J. H. 2009. Unlocking the influence of leadership network structures on team conflict and viability.Small Group Research,40: 301–322.

Balkundi, P., & Harrison, D. 2006. Ties, leaders, and time in teams: Strong inference about network structure’s ef- fects on team viability and performance.Academy of Management Journal,49: 49 – 68.

Balkundi, P., & Kilduff, M. 2005. The ties that lead: A social network approach to leadership.Leadership Quarterly, 16: 941–961.

Balkundi, P., Kilduff, M., & Harrison, D. 2009.Constructing charisma: A social network approach to team leadership and performance.Working paper, SUNY at Buffalo School of Management, Buffalo, NY.

Bandura, A. 1977.Social learning theory.Englewood Cliffs, NJ: Prentice-Hall.

Bass, B. M. 1985.Leadership and performance beyond expec- tations.New York: Free Press.

Bass, B. M. 1990.Bass & Stogdill’s handbook of leadership:

Theory, research, and managerial applications(3rd ed.).

New York: Free Press. 2010491 Galvin, Balkundi, and Waldman Bolino, M. C., Kacmar, K. M., Turnley, W. H., & Gilstrap, J. B.

2008. A multi-level review of impression management motives and behaviors.Journal of Management,34:

1080 –1109.

Bono, J. E., & Anderson, M. H. 2005. The advice and influence networks of transformational leaders.Journal of Applied Psychology,90: 1306 –1314.

Borgatti, S. P., & Everett, M. G. 1999. Models of core/periphery structures.Social Networks,21: 375–395.

Bowler, W. M., & Brass, D. J. 2006. Relational correlates of interpersonal citizenship behavior: A social network perspective.Journal of Applied Psychology,91: 70 – 82.

Brass, D. J., Galaskiewicz, J., Greve, H. P., & Tsai, W. 2004.

Taking stock of networks and organizations: A multi- level perspective.Academy of Management Journal,47:

795– 817.

Brass, D. J., & Krackhardt, D. 1999. The social capital of twenty-first-century leaders. In J. G. Hunt, G. E. Dodge, & L. Wong (Eds.),Out-of-the-box leadership: Transforming the twenty-first-century army and other top-performing organizations:179 –194. Stamford, CT: JAI Press.

Brower, H. H., Schoorman, F. D., & Tan, H. H. 2000. A model of relational leadership: The integration of trust and lead- er-member exchange.Leadership Quarterly,11: 227–250.

Brown, M. E., & Treviño, L. K. 2006. Ethical leadership: A review and future directions.Leadership Quarterly,17:

595– 616.

Burns, J. M. 1978.Leadership.New York: Harper & Row.

Burt, R. S. 1999. The social capital of opinion leaders.Annals of the American Academy of Political and Social Sci- ence,566: 37–54.

Burt, R. S. 2004. Structural holes and good ideas.American Journal of Sociology,110: 349 –399.

Burt, R. S. 2007. Secondhand brokerage: Evidence on the importance of local structure for managers, bankers, and analysts.Academy of Management Journal,50: 119 – 148.

Cathy, S. T. 2002.Eat mor chikin: Inspire more people.Deca- tur, GA: Looking Glass Books.

Cialdini, R. B. 1993.Influence: Science and practice(3rd ed.).

New York: Harper Collins.

Cialdini, R. B., & Richardson, K. D. 1980. Two indirect tactics of image management: Basking and blasting.Journal of Personality and Social Psychology,3: 406 – 415.

Conger, J. A. 1989.The charismatic leader: Behind the mys- tique of exceptional leadership.San Francisco: Jossey- Bass.

Daft, R. L. 2008.The leadership experience(4th ed.). Mason, OH: Thompson South-Western.

Dienesch, R. M, & Liden, R. C. 1986. Leader-member ex- change model of leadership: A critique and further de- velopment.Academy of Management Review,11: 618 – 634.

Elliott, C. 2004. Airlines blacklist fliers, some merely annoy- ing.NYTimes.com, http://www.nytimes.com/2004/12/14/ business/14blacklist.html, December 14.Fanelli, A., & Misangyi, V. F. 2006. Bringing out charisma:

CEO charisma and external stakeholders.Academy of Management Review,31: 1049 –1061.

Fanelli, A., Misangyi, V. F., & Tosi, H. L. 2009. In charisma we trust: The effects of CEO charismatic visions on securi- ties analysts.Organization Science.20: 1011–1033.

Fitzgerald, J. 2008. Hillary Clinton casts vote for Obama.

ABCNews.go.com,http://abcnews.go.com/Politics/ wireStory?id 6178401.

French, J., & Raven, B. H. 1959. The bases of social power. In D. Cartwright (Ed),Studies in social power:150 –167. Ann Arbor: University of Michigan Institute for Social Re- search.

Gardner, W. L., & Avolio, B. 1998. The charisma relationship:

A dramaturgical perspective.Academy of Management Review,23: 32–58.

Gerstner, C. R., & Day, D. V. 1997. Meta-analytic review of leader-member exchange theory: Correlates and con- struct issues.Journal of Applied Psychology,82: 827– 844.

Gilovich, T. 1987. Secondhand information and social judg- ment.Journal of Experimental Social Psychology,23:

59 –74.

Goffman, E. 1959.The presentation of self in everyday life.

Garden City, NY: Doubleday Anchor Books.

Granovetter, M. S. 1985. Economic action, social structure and embeddedness.American Journal of Sociology,91:

481–510.

Halloran, L. 2007. Oprah lends her magic to Obama’s run:

How the “O” factor lures cash and—maybe—women’s votes.USNews.com,http://www.usnews.com/articles/ news/politics/2007/11/30/oprah-lends-her-magic-to- obama.html/, November 20.

Hatch, M. J., Kostera, M., & Kozminski, A. K. 2006. The three faces of leadership: Manager, artist, priest.Organiza- tional Dynamics,35(1): 49 – 68.

Heider, F. 1958.The psychology of interpersonal relations.

New York: Wiley.

House, R. J., & Howell, J. M. 1992. Personality and charismatic leadership.Leadership Quarterly,3: 81–108.

Howell, J. M., & Shamir, B. 2005. The role of followers in the charismatic leadership process: Relationships and their consequences.Academy of Management Review,30: 96 – 112.

Ibarra, H. 1993. Personal networks of women and minorities in management: A conceptual framework.Academy of Management Review,18: 56 – 87.

Jacobsen, C., & House, R. J. 2001. Dynamics of charismatic leadership: A process theory, simulation model, and tests.Leadership Quarterly,12: 75–112.

Kilduff, M., & Krackhardt, D. 1994. Bringing the individual back in: A structural analysis of the internal market for reputation in organizations.Academy of Management Journal,37: 87–108.

Klein, K. J., & House, R. J. 1995. On fire: Charismatic leader- ship and levels of analysis.Leadership Quarterly,6:

183–198. 492July Academy of Management Review Lam, S. K. S., & Schaubroeck, J. 2000. A field experiment testing frontline opinion leaders as change agents.Jour- nal of Applied Psychology,85: 987–995.

Lengel, R. H., & Daft, R. L. 1988. The selection of communica- tion media as an executive skill.Academy of Manage- ment Executive,2(3): 225–232.

LePine, J. A., Erez, A., & Johnson, D. E. 2002. The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis.Journal of Applied Psychology,87: 52– 65.

Lincoln, J. R., & Miller, J. 1979. Work and friendship ties in organizations: A comparative analysis of relation net- works.Administrative Science Quarterly,24: 181–199.

Lord, R. G., & Brown, D. J. 2001. Leadership, values, and subordinate self-concepts.Leadership Quarterly,12:

133–152.

Lord, R. G., & Maher, K. J. 1991.Leadership and information processing: Linking perceptions and performance.Bos- ton: Unwin Hyman.

MacKenzie, S. B., Podsakoff, P. M., & Jarvis, C. B. 2005. The problem of measurement model misspecification in be- havioral and organizational research and some recom- mended solutions.Journal of Applied Psychology,90:

710 –730.

Mayer, R. C., Davis, J. H., & Schoorman, F. D. 1995. An inte- grative model of organizational trust.Academy of Man- agement Review,3: 709 –734.

Mayhew, B. 1980. Structuralism versus individualism. Part 1:

Shadowboxing in the dark.Social Forces,59: 335–375.

Mayo, M., & Pastor, J. C. 2007. Leadership embedded in social networks. In B. Shamir, R. Pillai, M. C. Bligh, & M. Uhl- Bien (Eds.),Follower-centered perspectives on leader- ship: A tribute to the memory of James R. Meindl:93–113.

Greenwich, CT: Information Age.

Mehra, A., Kilduff, M., & Brass, D. J. 2001. The social networks of high and low self-monitors: Implications for work- place performance.Administrative Science Quarterly, 46: 121–146.

Meindl, J. R. 1990. On leadership: An alternative to the con- ventional wisdom.Research in Organizational Behav- ior,12: 159 –203.

Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. 1985. The ro- mance of leadership.Administrative Science Quarterly, 30: 78 –102.

Mintzberg, H. 1973.The nature of managerial work.New York: Harper & Row.

Montanaro, D. 2008. Obama surrogates defend candidate.

msnbc.msn.com,http://firstread.msnbc.msn.com/ archive/2008/03/10/751685.aspx/, March 10.

Mullen, B., Johnson, C., & Salas, E. 1991. Effects of communi- cation network structure: Components of positional cen- trality.Social Networks,13: 169 –185.

Newman, M. E. J. 2005. A measure of betweenness centrality based on random walks.Social Networks,27: 39 –54.

Organ, D. W. 1997. Organizational citizenship behavior: It’sconstruct clean-up time.Human Performance,10(2) : 85– 97.

Pastor, J. C., Mayo, M., & Shamir, B. 2007. Adding fuel to the fire: The impact of followers’ arousal on ratings of cha- risma.Journal of Applied Psychology,92: 1584 –1596.

Pastor, J. C., Meindl, J., & Mayo, M. 2002. A network effects model of charisma attributions.Academy of Manage- ment Journal,45: 410 – 420.

Roberts, L. M. 2005. Changing faces: Professional image con- struction in diverse organizational settings.Academy of Management Review,30: 685–711.

Rogers, E. M. 2003.Diffusion of innovations(5th ed.). New York: Free Press.

Sack, K. 1989. Danger of political surrogates is seen in Mason controversy.New York Times,October 2: B6.

Sahlins, M. 1972.Stone age economics.Chicago: Aldine Atherton.

Schlenker, B. R. 1980.Impression management: The self- concept, social identity, and interpersonal relations.

Monterey, CA: Brooks/Cole.

Shamir, B. 1992. Attribution of influence and charisma to the leader: The romance of leadership revisited.Journal of Applied Social Psychology,22: 386 – 407.

Shamir, B. 1995. Social distance and charisma: Theoretical notes and an exploratory study.Leadership Quarterly,6:

19 – 47.

Shamir, B., & Howell, J. M. 1999. Organizational and contex- tual influences on the emergence and effectiveness of charismatic leadership.Leadership Quarterly,10: 257– 283.

Shamir, B., Pillai, R., Bligh, M. C., & Uhl-Bien, M. (Eds.). 2007.

Follower-centered perspectives on leadership: A tribute to the memory of James R. Meindl.Greenwich, CT: Infor- mation Age.

Sparrowe, R. T., & Liden, R. C. 1997. Process and structure in leader-member exchange.Academy of Management Re- view,22: 522–552.

Sparrowe, R. T., & Liden, R. C. 2005. Two routes to influence:

Integrating leader-member exchange and social net- works perspective.Administrative Science Quarterly, 50: 505–535.

Thomas, J. G., & Griffin, R. W. 1989. The power of social information in the workplace.Organizational Dynamics, 18(2): 63–75.

Toegel, G., Anand, N., & Kilduff, M. 2007. Emotion helpers: The role of high positive affectivity and high self-monitoring managers.Personnel Psychology,60: 337–365.

Uhl-Bien, M., & Maslyn, J. M. 2003. Reciprocity in manager- subordinate relationships: Components, configurations, and outcomes.Journal of Management,29: 511–532.

Waldman, D. A., Ramirez, G. G., House, R. J., & Puranam, P.

2001. Does leadership matter? CEO leadership at- tributes and profitability under conditions of perceived environmental uncertainty.Academy of Management Journal,44: 134 –143.

Waldman, D. A., & Yammarino, F. 1999. CEO charismatic 2010493 Galvin, Balkundi, and Waldman leadership: Levels-of-management and levels-of- analysis effects.Academy of Management Review,24:

266 –285.

Wasserman, S., & Faust, K. 1994.Social network analysis:

Methods and applications.Cambridge, Cambridge Uni- versity Press.Weick, K. E. 1995.Sensemaking in organizations.Thousand Oaks, CA: Sage.

Williams, L. J., & Anderson, S. E. 1991. Job satisfaction and organizational commitments as predictors of organiza- tional citizenship and in-role behaviors.Journal of Man- agement,17: 601– 617.

Benjamin M. Galvin([email protected]) is an assistant professor in the business pro- gram at the University of Washington, Bothell. He received his Ph.D. from Arizona State University. His research interests include social processes, personality, and identity as they relate to leadership.

Prasad Balkundi([email protected]) is an assistant professor of management at the University at Buffalo. He received his Ph.D. in business administration from The Pennsylvania State University. His research interests include social networks and leadership in teams.

David A. Waldman([email protected]) is a professor of management in the W. P.

Carey School of Business at Arizona State University. He received his Ph.D. from Colorado State University. His research interests focus largely on leadership issues, especially at strategic levels. 494July Academy of Management Review Copyright of Academy of Management Review is the property of Academy of Management and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.