Reconsidering the Higher-Order Legitimacy of French and Raven's Bases of Social Power
in the Information Age

Owen Ambur, University of Maryland University College, July 15, 2000

French and Raven have identified seven types of power exhibited by people interacting socially. (Andrews & Baird)  Social power may be defined as "the capacity to exert influence over others." (Larison)  Power and authority are not synonymous. Power does not necessarily imply a commonly accepted authority ("right") to exert influence over others.  Moreover, the legitimacy of power is relative to the time, place, and circumstances in which it is wielded. This paper questions the higher-order legitimacy of several of French and Raven's seven forms of power in the information age.

As distinguished by French and Raven, the seven types of social power are: 1) reward, 2) coercive, 3) referent, 4) legitimate, 5) expert, 6) informational, and 7) connectional.

Reward power results from the ability to provide positive reinforcement for desired behavior.  Conversely, coercive power reflects the potential to inflict punishment.  In a sense these are not so much two different types of power as they are opposite ends of a continuum.  The common and essential element for both reward and punishment is that they are controlled by the "superior" person and are conferred upon subordinates based upon relationships that are less than perfectly aligned with their behaviors.  In other words, if the "inferior" individuals are perfectly able to control the results of their actions, the role of the superior is superfluous. The consequences speak for themselves and merely constitute the results - degrees of accomplishment or failure - rather than rewards or punishments.

Referent power is a function of the respect and esteem accorded to an individual by virtue of personal attributes with which others identify.  By contrast, legitimate power is based upon authority recognized in accordance with position in an organizational structure.  Referent power is person-oriented, while legitimate power is depersonalized.  Larison observes that referent power does not require action by the "superior" individual.  Instead, the referring individual voluntarily molds him or herself to the referent person.

Paradoxically, unless it is well supported by other forms, legitimate power lacks higher-order legitimacy.  Lack of such legitimacy is why organizational hierarchies are often ignored and bear relatively little relationship to the supply chains by which value is actually delivered.  The rumor mongering and griping about the "bosses" that are endemic in bureaucratic organizations are symptomatic of the illegitimacy of so-called legitimate power.  Employees simply fail to volunteer referent power to those occupying superior positions in the organizational hierarchies.  Lower-order, bureaucratic legitimacy is powerless to do anything about it, since referent power cannot be enforced through punishment nor bought with "rewards".

The legitimacy of power structures embedded in old-style bureaucracies is threatened with irrelevancy by virtue of the growing ease with which information can be shared outside the confines of those artificial and outmoded constructs.  However, to the degree that they are insulated from the impact of or are more potent than other forms of power, legitimate powers may prevail for extended periods of time, even though their most efficacious usages may be merely to support themselves.  Moreover, notwithstanding the fact that legitimate power is depersonalized, the individual human beings who are able to accrue it are personally motivated to maintain their share of it through application of other forms of power that they possess.

Expert power is a form of referent power resulting from recognized expertise while, as defined by French and Raven, informational power is a variation of legitimate power stemming from the ability to control the availability and accuracy of information. Larison observes that expert power is grounded in substantial differences in the knowledge of two individuals on particular subjects. Expert power is limited to the topic of expertise and, thus, is more delimited than referent power. However, it should be noted that expertise may be the most important form of referent power in the information age.

The evolution of information technology (IT), most notably the revolution that is the Internet and particularly the World Wide Web, is rendering moot French and Raven's definition of informational power, as more and more knowledge becomes "common".  In the reality of the cyberage, the information power wielded by old-style hierarchies is becoming restricted to information about the organization itself, information that it is either incapable or unwilling to share freely, honestly, and efficiently with others - a circumstance the portends poorly for the longevity of those bureaucracies.  At the same time, the explosion of information technology and thus the availability of common knowledge threatens to render as commodities much of the personal expertise that has been previously highly valued.

For organizations, the clear and widely aired implication is that hierarchies must be "flattened" and made more responsive to internal as well as external stakeholders.  In other words, the legitimacy of organizational power structures must be reevaluated.  For individuals, the implications are somewhat more complex.  For example, as increasing quanta of knowledge and expertise are commoditized, embedded into technology, and more widely and freely distributed, other, more qualitative forms of referent power may become more highly valued.  At the same time, the individual expertise that is highly valued will become more narrowly focused, ever expanding the boundaries of knowledge in highly specialized areas.

Finally, connectional power reflects the influence that leaders possess as a result of whom they know and the support they engender from others as a result (i.e., the bandwagon effect).  Connectional power is also a variation of referent power.  However, like legitimate power, it is depersonalized in the sense that it reflects attributes of others with whom the individual is associated, rather than attributes that are directly inherent to the person him or herself.  As the face and force of organizational constructs change, so too must the connectional power vectors.

A commonly recognized example is the impact upon lobbying firms from the changing of the guard that occurs when the party in power loses the majority in an election.  A less direct but equally informative example - which also illustrates the overlap of connectional and informational power - is the substitution of hyperlinks and search engines on the Internet for the intermediaries and gatekeepers whose personal and positional powers previously provided the only effective keys to information. In many circles, whom one knows is still far more important than what one knows.  However, those power vectors too are being pressed by IT.

While those who have come to occupy previously recognized positions of authority may feel threatened by these changes, it is important to recognize that legitimacy relates only to commonly accepted practice and implies no higher-order degree of truth, logic, honor, or justice. As Larison puts it, legitimate power stems from internalized values supporting the belief that the manager has a right to exert power and the individual has an obligation to accept it.

For example, a king may legitimately exercise dominion over his subjects simply because heredity has long been the accepted means by which such power is conferred.  Likewise, steep hierarchies and bureaucratic organizational structures bestow legitimate, reward, coercive, informational, and connectional power upon individuals regardless of the relevance of their referent or expert powers in net present value chains. (pun intended)  In short, legitimacy implies self-justification, i.e., the means justify themselves without regard to the merits of the ends (results).

Larison notes the importance of legitimacy to effective leadership.  However, he also draws a critical distinction: "Power and authority come from the person being influenced - not the person in the more powerful position.  If the follower chooses to not follow them, they are no longer leaders.  Leadership is really followership." (emphases added)

He also notes that no one is entirely bereft of influence but "some people have more net influence than others and hence ... more power."  Whether he intended it as such or not, his use of the word "net" has a dual meaning in this context.  First and most obviously, it relates to the power of individual's influence to overcome competing influences.  Second, it highlights that influence is exerted within a network of influences, including the informational power of the Internet (the Net).

Hayden (2000) excludes connectional power from his analysis and reiterates that among the other six bases of social power, some are positional and some are personal.  Legitimate, reward, and coercive powers are based upon an individual's position in an organization, while expert, informational, and referent powers are based upon personal attributes.  Insightfully, he avers: "Those who possess only positional power are bureaucrats.  Those who possess only personal power are leaders.  Those who possess all six are [good] managers." (emphases added)

Hayden suggests individuals have two types of power upon becoming a manager - the legitimate power of the position and the expert power that led to their appointment.  In addition, reward and coercive power generally accompany legitimate power.  Thus, in Hayden's conceptualization, managers may have four of the six powers by default.

However, he acknowledges that managers may not in fact have expert power - a tacit admission of the seventh power (connectional), which is excluded from his analysis, as well as the rapid degradation of expertise in a dynamic, information-based economy.  Moreover, managers may be given the benefit of the doubt on informational power and expertise, meaning that such power is theirs to lose, based upon their performance as perceived by others.

The broadest type of power - referent power - must be earned through expanded expert power sustained by a high degree of professional knowledge and integrity.  Credibility on one topic may carry over into other areas of expertise and it can be fostered via the effective use of informational power.  Hayden also notes that legitimate power often encompasses the power of access to information.

Hayden quotes John P. Kotter: "Americans, as a rule, are not very comfortable with power or with its dynamics... And this misunderstanding is becoming increasingly burdensome because in today's large and complex organizations the effective performance of most managerial jobs requires one to be skilled at the acquisition and use of power."

Toward that end, Hayden suggests the keys to wielding power successfully are to: 1) be sensitive to the sources of power and be consistent with the expectations of others; 2) recognize the costs, risks, and benefits of the different bases of power and draw on whichever is appropriate to the situation and person; 3) appreciate that each power base has merit; 4) possess career goals that allow development and use of each type of power; 5) act maturely and exercise self-control, avoiding impulses and egotistical displays; and 6) understand that power is necessary to get things done.  He concludes that effective managers use all six social bases of power.  If it had been included in his list, undoubtedly he would argue they should use connectional power too.

However, assertions that Americans should be more receptive to the application of power in large, bureaucratic organizations should not be accepted uncritically, regardless of whatever personal or positional powers Kotter and Hayden may claim for themselves.  Instead, the far greater risk may be that Americans, indeed citizens of all nations, are already much too amenable to the application positional powers, whereby the means are deemed to justify the ends.

With reference to so-called legitimate authority and other forms of power run amok, consider Suedfeld's (1999) succinct summarization of the results of research on why people participate in the pursuit of patently unjustifiable aggressive acts against others:

Milgram's research has shown the prevalence ... of obedience to what appears to be legitimate authority; Kelman, French and Raven, and others have analyzed how a leader's perceived expertise, attractiveness, power, and arguments can make trust and obedience more likely. Asch and Schachter demonstrated that even under unthreatening laboratory conditions, conformity to peer-group norms and pressures makes it difficult to dissent when those around us are fully convinced that they are right... groupthink, a combination of ingroup pride, conformity, and leader-worship, may lead to the unthinking approval of decisions that are both immoral and disastrous. Kelman and Hamilton's model of how moral inhibitions against violence are the approval of an authority figure...seems valid for what they call "sanctioned massacres" ... Ervin Staub has traced the step-by-step indoctrination and training - the "continuum of destruction" - that turns an innocent military conscript into a torturer, as well as the conditions that can turn an artistically advanced, scientifically sophisticated nation into a culture of mass murder.

The focus of Suedfeld's remarks is the cause and historical context of the Holocaust.  In a more positive and forward-looking vein, addressing human requirements for living aloft, NASA (2000, ch8-2) observes that prolonged separation from Earth may undermine the bases of legitimate power.  They cite Haythorn as follows:

Leadership and other behavioral patterns are clearly determined to some degree by role expectations and behavioral prescriptions of larger segments of organizations and societies than are fully represented in the small group. When groups are isolated from contact with (such larger segments) these prescriptions and expectations cannot be as frequently and strongly reinforced as they normally are .... Leadership under such circumstances is unable to rely as strongly on formal role relationships and must depend more on the individual capabilities of the men to whom leadership is assigned.

NASA (ch8-5) notes further that hierarchical control requires that individuals at one hierarchical level have the power to influence individuals at subsequently lower levels.  They suggest that each of the bases of power defined by French and Raven may be undermined during a space mission of extended duration, and that the norms supporting legitimate power may not be sufficiently reinforced in space.  In particular, they say:

The satisfactions and sanctions that underlie reward and coercive power on Earth may prove unavailable or unreliable in space. On extended-duration missions, the decline of unused technical skills, cross-training, and the mounting likelihood of appearing foolish may undermine expert power. Finally, interaction under conditions of isolation, confinement, and risk may decrease interpersonal attraction and erode referent power... It is expected that crews will become increasingly autonomous as they involve more people, last longer, and travel farther from Earth. If an acceptable level of Earth-based or centralized control is to remain possible, communications problems must be solved, and new means must be identified for setting bounds on the behavior of crewmembers.

Thus, it seems that self-justified, so-called legitimate and centralized power is threatened by physical space (distance) and elapse of time, as well as by the immediate and growing presence of cyberspace.  NASA suggests that further research is needed to reinforce the bases for legitimate power.  However, a more pertinent topic for study might be the very bases of the higher-order legitimacy of so-called legitimate power itself.  Extended travel to outer space may provide a unique perspective from which to view this issue, but for most of us - including anyone who has a serious interest in the future of our socio-organizational constructs - cyberspace provides a far more relevant and timely focus for such discourse.

In the information age, there is good cause to believe that the greatest good for the greatest number of human beings would be well served by a fundamental reexamination of the higher-order legitimacy of so-called legitimate power and its associated powers of position.  Chansler (1997) reports "scores of studies ... suggest that the use of expert power by leaders is most acceptable and effective with followers and that coercive power is least likely to result in follower compliance."

By definition, in any line-authority hierarchy there are more subordinates than supervisors, meaning: a) that power is concentrated among relatively few people, and b) most people are relatively disempowered.  Thus, it is not surprising that Klein (1998) found correlation lacking between supervisory power bases and subordinate satisfaction with work.  Since employees are disempowered by supervision, why should it matter which form of power the supervisor uses, and why should we be surprised if employees draw no satisfaction from it?

Consistent with other studies, Klein did find a relationship between satisfaction with supervision and the use of expert and referent power bases.  In other words, to the degree they are doomed to subservience, employees are happier with superiors who possess expert knowledge and who are commonly recognized for it and/or for other personally attractive attributes (i.e., referent power.)  Even then, however, Klein found no relationship between employee satisfaction with the work itself and the forms of power exerted by their leaders.  She viewed that as a cause for concern and further study.

Indeed, the pursuit of greater knowledge is a very worthy cause but a more appropriate objective is just the opposite of Klein's proposition, which is to look for ways to bolster employee work satisfaction derived from the followership.  Instead, the charge is well captured in the old anarchists slogan: "Question authority!"  In somewhat less confrontational terms, it is well stated in Savage's (1996) admonition to "push back" until the truth is known.  In more idealistic terms, it is articulated in Martin Luther King's famous truism: "The truth will set us free" ... free from enslavement to bureaucratic positional powers - "powers-that-be" first and foremost for their own self-preservation.

The quest for truth is unending.  It is the essence of the human condition.  To suggest that legitimate power is devoid of truth may be exaggeration but, at best, legitimacy is an expedient short-cut - one that may have been necessary in earlier times but seems significantly outmoded and vastly over-rated in the information age.

There is little doubt that: a) the higher-order cause of truth is well served through the free and open application of personal informational, expert, and referent powers, and b) lower-order legitimate positional powers are commonly used to thwart the free and open application of higher-order personal powers.

People do not need to prove their worth as human beings, at least they shouldn't be expected to do so in a land where "all [persons] are [deemed to have been] created equal."  Instead, the organizational and positional artifacts that groups of people have collectively imposed upon themselves and others should be rigorously and continuously subjected to reality checks, as to their efficacy in supporting and upholding the dignity and creativity embodied in each and every individual human being.  In the final analysis, it is highly doubtful that the higher-order legitimacy of many existing pseudo-legitimate power structures can be sustained.  How long they and their legitimacy can last in the information age remains to be seen.

Technology and IT in particular may be morally neutral, taking on the intents ascribed by the human beings who wield it.  However, technology empowers people, thereby embodying the potential to reshape the socio-organizational artifacts through which human beings have traditionally resolved their personal and social shortcomings.  To the degree that those with positional power are permitted to magnify and lord it over others through the application of IT, the fate of the masses may be in grave jeopardy.

However, since lower-order legitimacy relies upon tacit followership, it bears the seeds of its own destruction.  A higher-order form of legitimacy is the likely result - ushering in a golden age of personal rather than positional power.


Andrews, P.H., & Baird, J.E. (2000). Communication for Business and the Professions. 7th Edition. New York, NY: McGraw Hill. pp. 371 - 373

Hayden, William D. The Power Theory of Leadership. Kent State University, Stark Campus. Downloaded from the Web on July 1, 2000, at:

Klein, Carin Joyce. (1998). French and Raven's Bases of Social Power in a Not-for-Profit Health Care Facility: Perceptions and Satisfaction. Dissertation Abstracts. 59, no. 07A: 2605.

Larison, Robert. (1997, October 5). Course Syllabus, BA 321, Principles of Management, Eastern Oregon University School of Education and Business Programs. Downloaded from the Web on July 1, 2000, at

NASA. LIVING ALOFT: Human Requirements for Extended Spaceflight. Downloaded from the Web on July 4, 2000: and

Savage, Charles M. (1996) 5th Generation Management: Co-Creating Through Virtual Enterprising, Dynamic Teaming, and Knowledge Networking. Boston, MA: Butterworth-Heinemann. p. 248.

Suedfeld, Peter. (1999, May 20). The Past as the Crossroads of the Present and the Future: Reverberations of the Holocaust. The University of British Columbia. Presidential Address at the Annual Convention of the Canadian Psychological Association. Downloaded from the Web on July 1, 2000, at