Earthly power doth then show likest God's
When mercy seasons justice (Shakespeare, 1597/1997, p. 311).
A catastrophe inherently creates an urgent challenge for leadership. Here, catastrophe is used to describe a major crisis realized suddenly and so serious that the leader's organization, or a significant part of it, may not achieve fundamental goals or even survive. Confederate capture of Fort Sumter challenged Abraham Lincoln. Japan 's attack on Pearl Harbor challenged Franklin Roosevelt. Argentina 's attack on the Falkland Islands challenged Margaret Thatcher.
Vision statements, as a standard tool leaders use to focus organization members on achieving a desired goal, may be especially significant when a catastrophe has occurred. A useful research design might view the catastrophe as an intervention and measure its effect by comparing vision statements the same leader articulated before and after the catastrophe. Psychological theory predicts there will be measurable differences in those texts (Satterfield, 1998; Suedfeld & Bluck, 1988; Suedfeld & Tetlock, 1977; Wallace & Suedfeld, 1988; Winter & Stewart, 1977).
Overt changes in the content of vision statements over time frequently are analyzed publicly in the news media. A number of theories posit the existence and measurability of latent characteristics such as moral reasoning stage, moral reasoning orientation, explanatory style, and rumination. A common theme of such theories is that examination of latent characteristics reveals important truths that otherwise might be overlooked and yields helpful insights into human and organization behaviors (Post, Walker , & Winter, 2003). Theory-based examination of latent characteristics may use structured content analysis to appropriately sample thought content that leaders produced under particular conditions (Smith, 1992).
Moral Reasoning Orientation
This study focused on organization-level analysis of one particular latent characteristic of the vision statement: its moral reasoning orientation. Moral reasoning orientation is an important lens to use because it describes a worldview that frames thinking about moral conflicts and what factors deserve priority when resolving them.
Gilligan drew attention to moral reasoning orientation in 1982. Piaget (1976, 1997) and Kohlberg (Colby & Kohlberg, 1987) had based their research into moral reasoning development stage on an assumption that moral reasoning orientation frames conflicts in terms of justice. The considerations identified with justice included laws, standards, rules, principles, obligations, duty, commitment, and fairness. Gilligan (1993) discovered a group of women she believed used a different moral reasoning orientation: caring. The considerations identified with caring included alleviating burdens and suffering, fostering the welfare of another, maintaining and restoring personal relationships, interdependence, and avoiding conflict.
Gilligan (1993) demonstrated that which orientation is used, caring or justice, can lead to different decisions and outcomes. Gilligan used as an example a classic situation devised by Kohlberg that is known as the Heinz dilemma. Heinz' wife has a terminal illness but a recently developed drug might save her. The manufacturer's price for the drug is cost of manufacture plus a 900% profit, and that total is double the money Heinz was able to raise using his best efforts. The manufacturer refuses to reduce his profit or let Heinz pay later. When framed as a justice issue, this dilemma can give priority to life over property and give Heinz justification to steal the unaffordable drug to save the life of his wife. Oppositely, when framed as a caring issue, this dilemma can eschew theft and require seeking some way to buy the drug, some way to pay the unaffordable price, because the circumstance of Heinz stealing, being caught, and going to jail would create harm for the wife by making it impossible for Heinz to care for her over the long term.
Lyons (1982) designed a procedure to identify moral considerations and categorize them as justice or care (Gilligan, 1988; Gilligan & Attanucci, 1988). Johnston (1988) demonstrated that adolescents of either gender could access either orientation, caring or justice, and could switch between them. This study assumed that leaders are no different. They, too, can access either caring or justice and switch between them. Which do real leaders' vision statements express when they face a real catastrophe?
Definitions and Considerations
To qualify as a vision, a statement must (a) be accessible to substantially all organization members or their designated representatives, (b) present an image of the future organization, and (c) present an improvement more attractive than other perceived alternatives. Although other criteria have been offered, they often prove unhelpful or sometimes wrong. For example, “specific,” “realistic,” and “credible” are not valid criteria because a useful vision statement may be quite the opposite: fuzzy, faith-based, values-based, belief-based, radical, bold, and unconventional (Bass, 1998; Schein, 1992).
Level of Analysis
Since leaders' statements may be influenced by other people, level of analysis is an important consideration. This study focused on catastrophes and vision statements related to 7 pivotal national leaders who clearly have advisors and do not act alone (Basler, 1953; Bush, 2003; Campbell, 2003; Frum, 2003; Gelderman, 1997; Hughes, 2004). It is difficult to discern precisely whether the advisors' influence is large (Kennan & Hadley, 1986) or small (National Archives and Records Administration, 2001). Presidents have been so deeply involved in the preparation of their vision statements that observers opine that advisors cannot sneak words into the president's mouth. The usual presidential speech is prepared by 5 to 20 people and that extensive group input means the speeches do not voice just personal ideas of the president but instead represent at least the advisory group and some say the nation and its people (Medhurst, 2003). Eight other scholars provided evidence supporting that view in the same volume (Ritter & Medhurst, 2003). Frum reported personal experience as one of George W. Bush's speechwriters and described a process designed intentionally to allow many people to contribute ideas and words that are integrated into the coherent vision voiced by the president.
Pivotal leaders are formal leaders whose choices appear to set direction for their entire organization. The literature recognizes that throughout history pivotal leaders who won widespread fame relied on a team of supporters (O'Toole, Galbraith, & Lawler, 2003). Under that condition, if analysis were attempted at the individual level to predict an individual leader's behavior or to interpret an individual leader's personality, the influence of advisors and consultants would be a confounding factor degrading interpretations and inferences. A fair criticism would be, who is being analyzed—the leader or advisors and speechwriters? However, if analysis is focused at the organization level and assumes the organization's leadership is composed of a formal leader working together with advisors and consultants, then the multiple influences on vision statements are not a confounding factor but part of the integrated phenomenon under study.
Analysis of leadership effected by multiple people, dispersed or shared leadership, found organized expression beginning in the 1990s, although elements existed earlier in the literature (Bryman, 1996; Pearce & Conger, 2003). One conceptual strength is recognition that even the most pivotal leader—perhaps especially the most pivotal leader—is unlikely to act alone.
An unresolved issue relates to two practical aspects of distribution of power: accountability and priority. With respect to accountability, Western culture tends to hold a single person accountable for the actions of an organization no matter how leadership is shared (O'Toole et al., 2003). With respect to priority, Locke (2003) opined that shared leadership should be very fragile, if it even exists, because people at the top will disagree and decision making will be paralyzed if no one has the final say.
As early as 1974, Stogdill addressed the distribution and sharing of leadership power. But a focus on power may miss the point. Shared leadership is dynamic, an interactive influence process, whereby people lead each other to achieve the organization's goals (Pearce & Conger, 2003). That definition is silent on power; it is focused on interaction.
This study focused only on human-caused catastrophes. Situational influences are widely recognized to be mediators of human behavior and this study assumes it is important to distinguish between catastrophes caused by nature and those caused by humans. While nature does cause catastrophes, human-caused catastrophes seem to evoke qualitatively different feelings. Nature seems not to be held culpable of wrongdoing in the way people who cause a catastrophe are.
Not every research method is suited to catastrophe research. Direct observation is possible (Festinger, Riecken, & Schachter, 1956) but difficult because researchers are unlikely to have adequate access to the leader under emergency conditions. Repeating catastrophes to facilitate researcher access would be unconscionable. Covert simulations were attempted in the past; researchers sought high reliability by using high-impact deceptions (Amato, 1983; Aronson, 1992; Darley & Latané, 1968; Latané & Darley, 1968; Milgram, 1969), but today those are unlikely to be approved by institutional research ethics boards. Direct observation, interviews, surveys, and overt simulations all face criticism because results are influenced by reactivity between the researcher and the participant. What people say and do when they know they are research participants may not be what they actually do when living through a real catastrophe.
Unobtrusive methods appear to be a good choice for catastrophe research (Webb, Campbell, Schwartz, & Sechrest, 2000). For this study, content analysis of archival records offered opportunity to obtain data untainted by interaction between the investigator and the leader. This method also permits comparison of natural events across time and cultures (Oettingen, 1995; Peterson, 1992; Schulman, Castellon, & Seligman, 1989; Seligman, 1998; Webb et al., 2000). However, content analysis does not allow the control over variables that is available under laboratory conditions, and it can be affected by investigator bias or fuzzy interpretations. For those reasons, several procedural controls have been recommended, including adhering to strict criteria to identify passages extracted for analysis, using multiple raters to score the passages, keeping those raters blind to each other's scoring and to study outcomes, presenting passages to raters in random order, measuring intercoder reliability, and seeking triangulating evidence (Smith, Feld, & Franz, 1992; Webb et al.; Winter, 2003).
When real leaders face a real catastrophe, what do their vision statements express, caring or justice? This study examined the vision statements of 7 pivotal national leaders to determine how a real catastrophe affected those visions by using content analysis to detect changes in moral reasoning orientation, a latent characteristic, that allow inferences about the effects of real-world catastrophes on those visions and the worldview used to frame thinking about moral conflicts and what factors deserve priority when resolving them. Three visions articulated before a catastrophe were compared to three articulated afterward for consistency or changes in moral reasoning orientation. The same vision statements were scored twice by different rater panels, once holistically with forced choice between caring and justice, and once at proposition level with caring and justice scored separately.
Because Lyons (1988) and Johnston (1988) found people using both caring and justice orientations simultaneously, in this study both justice and caring orientations were expected likely to appear. That seemed especially likely because, for these leaders operating at the national level, the vision statements were likely to incorporate multiple advisors' ideas (Gelderman, 1997; Ritter & Medhurst, 2003; Suedfeld, Guttieri, & Tetlock, 2003; Winter, 2003). Since conditions of stress affect the influence various people exert in a group (Janis, 1982), changes in moral reasoning orientation in reaction to catastrophe also were expected. No prior research was located with empirical data predicting the direction of that change. If the hypothesis to be tested is based on the Piaget (1976, 1997) and Kohlberg (Colby & Kohlberg, 1987) assumption, moral reasoning orientation will be expected to frame conflicts in terms of justice.
The hypothesis tested is that these leaders' vision statements before a catastrophe will show both justice and caring moral reasoning orientations and that their vision statements after a catastrophe will show change and the direction will be toward more justice.
Selecting Leaders and Texts
The 7 leaders were selected by identifying human-caused catastrophes with vision statements leaders articulated before and after the catastrophe (Table 1). A practical constraint was that archival vision statements may exist only for national leaders in the public sector. They may not exist for other leaders for multiple reasons. Perhaps those leaders' vision statements were not publicly voiced or documented, or perhaps post-catastrophe vision statements were avoided for legal reasons.
Leaders and Catastrophes Selected for Study
George W. Bush
|| Terrorists attacked the United States
|| September 11, 2001
| Margaret Thatcher
|| Argentina attacked the Falkland Islands
|| April 2, 1982
| Golda Meir
|| Egypt and Syria attacked Israel
|| October 6, 1973
| John F. Kennedy
|| Soviet Union deployed offensive missiles and aircraft to Cuba
|| October 16, 1962
| Franklin Roosevelt
|| Japan attacked Pearl Harbor
|| December 7, 1941
| Woodrow Wilson
|| Germany waged unrestricted submarine warfare
|| April 12, 1917
| Abraham Lincoln
|| Confederates captured Fort Sumter
|| April 14, 1861
For each leader, an inventory identified prepared speech texts before and after the catastrophe that are available from official or authoritative sources. While a reality is that published texts exist in various forms that do not always match exactly, such as pre-delivery news releases, speaking texts, as-delivered texts, and transcripts, the published texts from official or authoritative sources bear a rebuttable presumption that they are substantially correct and a foundation for the leader's reputation.
Examination of the texts listed on each inventory identified those meeting the criteria for a vision statement: accessible to substantially all organization members or their designated representatives, providing an image of the future organization, and providing an improvement more attractive than other perceived alternatives. Counting from the catastrophic event, the last three vision statements before the catastrophe and the first three after it were identified. Care was used to avoid subjective editing by rules of type, like whether the statements expressed national direction or policy, or by rules related to topic, like whether they addressed defense or welfare, war preparedness, or education improvements. This study examined each leader's pattern of actual vision statements of all types before and after a catastrophe. Each of the 7 leaders was represented by 6 vision statements, for a total of 42 visions in all.
A content analysis study of moral reasoning orientation first requires extraction of scorable statements that strictly meet specified criteria. Following precedent that one researcher do all extraction because the process is tedious and requires meticulous examination of a large volume of text (Peterson, Schulman, Castellon, & Seligman, 1992), each vision statement was examined at least four times for extractable passages. For moral reasoning orientation, Lyons (1982) established extraction criteria. To be extractable, a text must contain: (a) a real-life moral dilemma, (b) an asserted solution, and (c) an explanation of the dilemma or an evaluation of that dilemma. Dilemma can further be split into two components: situation and conflict.
Extraction identified 115 scorable passages that were used for both Study 1 and Study 2. What varied between the studies was the unit of coding. For Study 1, the 115 extracts were used intact. They contained all three elements: a dilemma, an asserted solution, and an evaluation or explanation. For Study 2, the 115 extracts were divided into 470 stand-alone propositions. Usually these were one element: situation, or conflict, or solution, or evaluation. But sometimes one proposition could be subdivided into several stand-alone sub-propositions and occasionally solution and evaluation were inextricably intertwined.
For presentation to raters, the 115 Study 1 extracts and the 470 Study 2 propositions were randomly sequenced using numbers drawn from a table of random numbers. Raters were instructed to score each extract by comparing content to a scoring aid showing considerations of caring and justice moral reasoning orientations (Table 2).
For Study 1, raters performed forced choice scoring with five options: justice (exclusively or overwhelmingly), justice dominates but caring appears, caring dominates but justice appears, caring (exclusively or overwhelmingly), or no justice and no caring. For analysis, those ratings were converted to numbers: + 2, + 1, - 1, - 2, or 0, respectively. After scoring, the extracts were grouped by vision statement and their mean became the vision statement's score (Figure 1).
Considerations of Care and Justice
|Considerations of Care
||Considerations of Justice
Alleviate burden/suffering, Foster welfare of another, Maintain / restore interpersonal relationships, Avoid interpersonal conflict, Interdependence, Situation over principle, Effects on others, Etc.
||Laws, Standards / rules / principles, Obligation / duty / commitment, Fairness, Others have their own contexts, Principle over situation, Effect on self, Etc.
For Study 2, raters scored each proposition twice, once for justice and once for caring. In both cases raters had three options: clearly present, uncertain, and clearly not present. For analysis, those ratings were converted to 2, 1, and 0, respectively; justice scores were assigned a positive sign and caring scores were assigned a negative sign. After scoring, the propositions were grouped by vision statement and their means, one for justice and one for caring, became the vision statement's scores (Figure 2).
Figure 1. Moral reasoning orientation scores of 42 vision statements based on holistic scoring by raters who examined 115 scorable extracts in Study 1.
For both studies, students at California Lutheran University were employed as raters. They were primarily undergraduates but two Study 1 raters were in masters programs. They received orientation training and scored a number of practice texts to compare their scoring to “expert” scores. The raters were blind to each other's scoring and to study outcomes. They were not told who said each scorable item, but inescapable content clues sometimes may have allowed them to guess. [For example, who said, “Yesterday, December 7, 1941 —a date which will live in infamy . . .”? (Roosevelt, 1941/1950a).] Studies often delete such clues, but they were so vital to these vision statements that deletion was impossible here.
Figure 2. Moral reasoning orientation scores of 42 vision statements based on proposition scoring by raters who examined 470 scorable items in Study 2.
For Study 1, a panel of nine raters convened to score the 115 extracts in a one-session classroom environment. Each statement was scored holistically and independently by three of the raters. For Study 2, a panel of six raters scored the 470 propositions twice, once for the presence of caring and separately for the presence of justice. Each proposition was scored by all six raters, who worked independently at times and places of their own choosing over a period of several weeks.
Intercoder reliability was good. For Study 1, Cronbach's alpha (Cronbach, 1951), a = .90 when raters scored holistically and, for Study 2, a = .95 for caring and a = .97 for justice when raters scored them independently. While no measure of interrater agreement has won universal support (Jones, Johnson, Butler, & Main, 1983), including alternatives such as percent agreement, Pearson's r , and Cohen's kappa (Lombard, Snyder-Duch, & Bracken, 2004; Uebersax, 2002), Cronbach's alpha is an appropriate measure because an alpha approaching 1 indicates all raters applied scoring standards consistently and scores differed because raters' opinions differed (Jones et al.; SPSS, 1999). Cronbach's alpha also was adopted for this study for consistency with a larger, on-going study, of which this research is a part, and consistency with similar research previously completed by other investigators (Ballard & Suedfeld, 1988; Schulman et al., 1989; Zullow, Oettingen, Peterson, & Seligman, 1988; Zullow & Seligman, 1990). Cronbach's alpha effectively measures how raters' scores co-vary (the ratio of inter-item covariance to the average item variance) and mathematically is equivalent to the average of all split-half estimates of reliability.
Cronbach's alpha does not measure rater accuracy or the reliability of raters' judgments. Questions of that sort must be answered by examining relationships between ratings and external events, the role of triangulating evidence (Jones et al., 1983).
Moral reasoning orientation was found in 95.2% of the vision statements. Of the 42 vision statements, only 2 had no content meeting the criteria to be extracted for moral reasoning orientation. They were John Kennedy's fifth vision statement, which lacked a real-life moral dilemma, and Golda Meir's fourth vision statement, which lacked an asserted solution.
Whether Study 1 scores might be random was tested using a one-sample t test that compared the mean score of the actual single sample population to the mean score of a hypothesized random population and showed they are statistically different at the .01 level (p < .0005). The consolidated mean of Study 1 scores was 0.75, with a standard deviation of 0.82 and a 95% confidence interval ranging from 0.60 to 0.90. Overall, justice orientation dominated the 40 visions. However, raters scored caring dominant in 7 visions, a variation significant at the .01 level (p = .008). Those visions' consolidated mean was -0.62, with a standard deviation of 0.42.
Comparing Visions 3 and 4 for each leader, the visions immediately before and after the catastrophe, showed a norm and a deviation. The norm: Five leaders (71.4%) showed an increase in justice orientation after a catastrophe (Meir excluded because Vision 4 was unscorable). A paired comparison t test, used to compare scores before and after the catastrophe, showed the change was significant at the .05 level (p = .032). The deviation: Margaret Thatcher showed an increase toward caring orientation, a change statistically significant because it was 0.37 and greater than the 95% confidence interval for the 7-leader dataset of 0.30.
Comparing Visions 1-3 to Visions 4-6, all visions before and all visions after the catastrophe, similarly showed a norm, a deviation, and an anomaly. The norm: Six leaders (85.7%) showed an increase in justice orientation after a catastrophe. A paired-comparison t test, comparing scores before and after the catastrophe, showed the change was significant at the .01 level (p = .004). The deviation: Woodrow Wilson showed an increase toward caring orientation of 1.19, statistically significant because it is greater than the 95% confidence interval for the 7-leader dataset of 0.82. The anomaly: George W. Bush changed dramatically from caring to justice. Comparing Bush's Visions 1-3 to his Visions 4-6 showed change toward justice significant at the .05 level (p = .036). From Vision 3 to 4 alone, his moral reasoning orientation increased toward justice by 0.5, statistically significant because it is greater than the 99% confidence interval for the 7-leader dataset of 0.43.
One possible explanation of the norms and deviations is that they indicate framing of vision statements to garner followers' support. No direct test of the explanation is available in the data. Triangulating evidence from Frum is that advisors urged George Bush to change his vision statements after Vision 4 when a poll showed only moderate trust. A poll after Vision 5 showed high trust (Frum, 2003).
For both studies, the number of items (extracts or propositions) scored for each vision was tallied so a paired-comparison t test could be used to evaluate whether variability in vision scores was due largely to variability in the number of scorable items composing each vision. Restated, were the scores for visions containing many scorable items significantly different from the scores for visions containing only a few scorable items? The paired-comparison t test determined the statistical difference. Using p less than or equal to .05 as the standard, this test of Study 1 data showed variability in scores was not due to variability in the number of scorable items. Of the 40 scorable vision statements, 13 had fewer than two extractions, 15 had more than two, and the 12 with exactly two extractions could be split chronologically to add the first 7 to the low extractions group and the last 5 to the high extractions group. The paired-comparison t test for that split showed p = .873, so the difference between the low and high extraction groups was not statistically significant.
Generally, these American presidents showed considerable caring orientation even when justice dominated their visions. The two non-American prime ministers, Meir and Thatcher, showed higher justice orientation. One possible explanation is that the Presidents' environment called for a “presentation” style of speaking but Meir and Thatcher both spoke primarily in a parliamentary forum, perhaps using a discussion or debating style.
Another explanation might be image setting. Meir and Thatcher might have preferred the justice orientation in belief that it better created an image of invulnerability and powerful command. Such belief could be based on political experience, on antipathy toward gender stereotypes, or on advice from consultants. Triangulating information supporting the image-setting explanation included Bush's anomaly, his switch from caring to justice after 9/11 when Frum (2003) said advisors urged a change in his image, and biographer Campbell's (2003) comment that before Argentina attacked the Falkland Islands Thatcher already had taught the public to view her as the Iron Lady resolutely defending British interests and pride.
One-sample t tests were conducted to determine if the Study 2 means, for caring and for justice, based on 470 propositions might be random. Means for the actual single sample population were compared to mean scores for a hypothesized random population; they were statistically different at the .01 level (p < .0005). For caring orientation, the consolidated mean was -1.28, with a standard deviation of 0.21, and a 95% confidence interval ranging from -1.22 to -1.35. For justice orientation, the consolidated mean was 1.28, with a standard deviation of 0.13, and a 95% confidence interval ranging from 1.24 to 1.32.
Overall, Study 2 data showed no statistically significant change when Vision 3 was compared to Vision 4 or when Visions 1-3 were compared to Visions 4-6. Dominant score was calculated for each vision by subtracting the caring score from the justice score. Whether the dominant scores might be random was tested using a one-sample t test that compared the mean of the dominant scores to the mean score of a hypothesized random population. This test showed the means are not statistically different (p = .414) and therefore the dominant scores appeared random. Likewise, a paired-comparison t test, comparing the dominant scores before and after the catastrophe, showed no statistically significant change when Vision 3 was compared to Vision 4 (p = .077), or when Visions 1-3 were compared to Visions 4-6 (p = .321).
However, three specific changes in the Study 2 data were statistically significant at the .05 level. Lincoln and Roosevelt showed decreases in the number of propositions scored caring from Vision 3 to Vision 4 (p = .022); corresponding, Study 1 holistic scoring reported justice orientation prevailed and became stronger. In contrast, Bush showed an increase in the number of propositions scored caring from Vision 3 to Vision 4, a change of 0.18 that exceeded the 95% confidence interval (0.13). For Bush, Study 1 holistic scoring reported caring prevailed and became weaker, implying the holistic scorers based their judgment on something other than the quantity of caring propositions.
Study 2 found a correlation of .34 between caring and justice scores. Study 2 dominant scores showed a correlation to Study 1 scores of .51. Effect size, as a power analysis to quantify the impact of a particular intervention, showed the catastrophe caused a large effect (d = 1.02) in Study 1 when Vision 3 was compared to Vision 4. The Study 2 dominant scores for the same visions showed a medium effect (d = 0.71).
Catastrophe's Impact on Leaders' Caring and Justice
How did these seven human-caused catastrophes, when viewed as “interventions,” affect the moral reasoning orientation of these 7 leaders' vision statements? The data generally supported the tested hypothesis that these leaders' vision statements before a catastrophe would show both justice and caring moral reasoning orientations and that their vision statements after a catastrophe would show change and the direction would be toward more justice. But the data also showed deviations that warrant explanation.
Study 1 showed that Gilligan's (1993) linear model, with caring at one pole and justice at the opposite pole, emerged as useful for vision statement analysis if raters have opportunity to holistically score the extracts and to judge what points deserve emphasis.
Raters' Study 1 results are consistent with overt content of the texts. With respect to text raters scored caring, for example, Lincoln's (1861/1953) first vision statement was made at an early stop on his train trip to be inaugurated in Washington, DC. He praised the political parties for uniting to greet him in Cincinnati. He hoped threatening national difficulties would simply pass away. He spoke of intent to leave citizens alone and to treat people as equals with good hearts. Those are concepts of the caring orientation. Woodrow Wilson's (1917/1983c) sixth speech, also scored caring, was unusual in focusing on Flag Day, a holiday Wilson first proclaimed for national celebration. Wilson already had committed the United States to fight in World War I and in this vision he spoke of the nation's young men about to carry the flag into battle. Wilson said the nation was forced into war, but instead of arguing for revenge and justice he argued for honor, freedom, self-government, and making the world safe for all people who live upon it. Those are caring concepts. Franklin Roosevelt's (1941/1950b) sixth vision celebrated the 150th anniversary of the ratification of the Bill of Rights. His nation was fighting World War II to preserve the rights to life, liberty, and the pursuit of happiness, which are caring concepts.
Similarly, overt content showed justice themes in visions raters scored justice. Wilson 's (1917/1983a) first vision expressed outrage at the injustice of German submarines sinking the steamer Sussex, an attack that killed several U.S. citizens, and demanded that Germany abide by the rules of international law and the universal dictates of humanity, which are justice concepts. Wilson 's (1917/1983b) second vision sought formal authority and power from Congress to remain a legally neutral nation but to supply merchant ships with arms for their self-defense, all justice concepts. Roosevelt's (1941/1950a) fourth speech, following Japan 's attack on Pearl Harbor, presented a justice view that protested Japan 's premeditated invasion and committed the United States to prevail in absolute victory so the same form of treachery never again would endanger the nation.
Frum (2003) provided evidence triangulating the change from caring to justice orientation that raters saw in George Bush's vision statements. Bush's (2001a) first three vision statements were part of a series of brief, pre-recorded Saturday morning radio addresses to the nation. These advocated social welfare and education reforms. The fourth vision statement immediately followed and addressed the 9/11 terror attacks, praised responses to the emergency, and promised reaction. Frum attributed the fourth statement principally to Bush's senior communication advisor, Karen Hughes, saying she discarded and replaced words others had drafted. Frum also criticized the caring, compassionate tone of that fourth vision statement and wished it had been a war speech, justice oriented, and recognizing Americans were victims of a massacre and were ready to go to war against everyone connected to the attack. Frum said such criticism caused Hughes to back away from drafting Bush's (2001b) Vision 5, the September 20 address to a joint session of Congress and to the nation, which was the first of his clearly justice-oriented visions. Michael Gerson chaired the drafting of that speech. The change in moral reasoning orientation coincided with an increase in popular support. A poll after 9/11 showed about half the country felt Bush could cope with the attacks. A poll after the September 20 address showed what Frum characterized as a level of trust never before achieved by any leader in American history. Bush's sixth vision statement, two days later, focused on overcoming the nation's economic plunge and Frum did not report who influenced it.
What emerged from Study 2 with clarity is confirmation that these raters saw both caring and justice in every vision statement when they were required to score those characteristics separately and examine every component of each vision extract. That scoring procedure prevented raters from applying qualitative human judgment about what deserved emphasis and instead applied a quantitative measure, mechanical counting of the number of elements in each category. Although speakers and writers sometimes say they emphasized a point by addressing it at length, the lack of high correlation between Study 1 and Study 2 results showed differences in sensemaking. The Study 1 results, which had some support from triangulating evidence, indicated these raters were able to use holistic judgments to make sense of visions and that using mechanical length or frequency, a procedure sometimes used in previous research (Lyons, 1988), did not always yield corresponding outcomes, a reality others also have observed (Jones et al., 1983; Lombard et al., 2004; Winter & Stewart, 1977).
While early research positioned justice and caring as if they were opposite ends of a single, linear scale (Johnston, 1988; Lyons, 1982), a meta-analysis by Jaffee and Hyde (2000) found some research measured caring and justice independently, which would allow moral reasoning orientation to be scaled in two dimensions. Liddell (1990) did measure them independently, concluded her instrument measured two different constructs, but did not report the scores plotted on a two-dimensional chart. This study explored two-dimensional plots but found them unproductive. Raters' holistic judgments as to which orientation prevailed were more useful. This outcome is consistent with Jaffee and Hyde's opinion that a useful direction for research would be to focus on how individuals integrate justice and caring and determine which is the right basis for action.
Generally, the vision statements of these leaders, when facing these catastrophes, changed toward increased justice orientation. The two exceptions, the visions expressed by Woodrow Wilson and Margaret Thatcher, may be explainable. Both remained predominantly justice oriented, but Wilson 's framed going to war in Europe as a service to mankind and Thatcher's framed going to war in the Falkland Islands as a service to Falkland Islanders, people who deserved self-determination. For both Wilson and Thatcher, the issue was not framed as defense of the homeland from overt attack (the issue for Lincoln, Roosevelt, Kennedy, Meir, and Bush), which would be a justice cause, but enhancement of freedom for other, distant people, a caring cause.
Justice versus Caring Moral Reasoning Orientations
When Gilligan (1993) documented the existence of caring orientation, she asked if Kohlberg's system for scoring moral reasoning stage was flawed by its reliance solely on justice orientation and its exclusion of caring orientation. Kohlberg argued his system was not flawed because even if dilemmas were presented as justice issues people could choose to resolve them by using caring orientation (Colby, Kohlberg, & Kauffman, 1987). But this study's data reinforce Gilligan's question. The Study 2 data on 40 visions indicated these 7 leaders always used both caring and justice considerations and measuring just one of the perspectives did not yield results that matched the Study 1 holistic judgments. Moreover, the Study 1 data showed that Lincoln, Wilson, Roosevelt, and Bush all articulated visions that raters scored as predominantly caring. Thus, caring emerged as both a frequent characteristic and an influential one, supporting Gilligan's position that reliance solely on justice orientation could yield flawed results.
Although 2 of the 7 leaders were women, Meir and Thatcher, the organization level of analysis used for this study means the outcomes contribute nothing to the idea originally raised by Gilligan (1993) that women may prefer the caring and men may prefer the justice orientation. The conservative view adopted here is that these organization-level vision statements were dynamically influenced by multiple advisors as part of a shared leadership process, so both men and women may have had significant input that shaped the vision statements.
Two of the leader statements that qualified as visions contained no text meeting the criteria to be extracted for moral reasoning orientation scoring. The first was Kennedy's fifth vision, which followed so closely after his fourth vision that it might be seen as an auxiliary supplementing and assuming the listener still remembered the fourth vision. In Vision 4, Kennedy (1962a) announced discovery of offensive missile sites being prepared on Cuba and intention to stop that. In Vision 5, Kennedy (1962b) announced evidence that the missile bases were being dismantled and he presented no moral dilemma. The second vision lacking text qualified for extraction for moral reasoning orientation scoring was Meir's (1973) fourth vision. It was a broadcast to her nation that announced an attack on Israel by Egypt and Syria , asserting that defense forces were beating back the assault but presented no explanation or resolution of the dilemma, possibly assuming that listeners' patriotism would provide them all the explanation they required.
Results clearly showed these raters, when allowed opportunity to judge holistically as in Study 1, appeared to interpret what vision elements warranted emphasis even when the vision statement gave other elements more space, time, or frequency of mention. Do organization members interpret their leaders' visions the same way? It is a testable research question, and the answer could influence leaders who, today, sometimes say they believed they emphasized a point by speaking on it at length or speaking about it frequently.
It is interesting that these raters found in Study 2 that all 40 of the scorable visions included both justice and caring elements. This is consistent with previous research that found both genders could access both justice and caring and switch between them and found people using both orientations simultaneously. It also is consistent with the theory that shared leadership is a dynamic, interactive influence process allowing multiple people to lead each other to achieve the organization's goals. That concept assumes multiple people will express multiple viewpoints that will influence others in the leadership group and be accommodated in visions articulated by the formal leader. Frum's observation of Karen Hughes' withdrawal from influencing Bush's vision statements at the same time this study measured a change in those visions from caring to justice moral reasoning orientation suggests the shared leadership dynamic was real in that situation and also that it potentially has important effect on leadership outcomes. Left for future research is the question of when is it beneficial for a vision statement to be caring and when is it better oriented toward justice?
A unifying perspective from organizational systems theory recognizes levers as a mechanism allowing a small effort by the leader to achieve a large effect on the organization. A vision statement can be seen as a standard tool that leaders intend to have exactly that effect. When visions change toward more caring or more justice, the reason might be reactive or proactive. A reactive rationale is that the situation influenced how the leaders framed their vision statements. A proactive rationale, drawing on the concept of systems levers, is that the leaders sought to influence the situation by choosing how to frame their vision statements, using them as a lever with large effect on their organization.
The data here confirmed the hypothesis that both caring and justice moral reasoning orientations appeared in these vision statements by these leaders before a catastrophe. The data confirmed that change occurred after a catastrophe and showed the change usually was toward increased justice with several clear exceptions that showed change toward increased caring. These measurements perhaps reveal a leadership lever in action.
About the Author
Dr. Carl R. Oliver received his Ph.D. from the School of Human and Organization Development, Fielding Graduate University. His research interests include positive organization development and development of ethics in complex adaptive systems.
Amato, P. R. (1983). Helping behavior in urban and rural environments: Field studies based on a taxonomic organization of helping episodes. Journal of Personality and Social Psychology, 45(3), 571-586.
Aronson, E. (1992). The return of the repressed: Dissonance theory makes a comeback. Psychological Inquiry, 3(4), 303-311.
Ballard, E. J., & Suedfeld, P. (1988). Performance ratings of Canadian prime ministers: Individual and situational factors. Political Psychology, 9(2), 291-302.
Basler, R. P. (1953). The collected works of Abraham Lincoln (Vols. 4-5). New Brunswick, NJ: Rutgers University Press.
Bass, B. M. (1998). Transformational leadership: Industrial, military, and educational impact. Mahwah, NJ: Lawrence Erlbaum Associates.
Bryman, A. (1996). Leadership in organizations. In S. R. Clegg, C. Hardy, & W. R. Nord, Handbook of organization studies (pp. 276-292). Thousand Oaks, CA: Sage.
Bush, G. W. (2001a, September 11). Statement by the president in his address to the nation. Retrieved October 27, 2002 , from http://www.whitehouse.gov/news
Bush, G. W. (2001b, September 20). Address to a joint session of Congress and the American people. Retrieved October 27, 2002 , from http://www.whitehouse .gov/news/releases/2001/09
Bush, G. W. (2003). President Bush discusses top priorities for the U.S. : Press conference of the president. Retrieved November 29, 2003 , from http://www.whitehouse.gov/news
Campbell, J. (2003). Margaret Thatcher: Vol. 2. The iron lady. London: Jonathan Cape.
Colby, A., & Kohlberg, L. (1987). The measurement of moral judgment: Theoretical foundations and research validation (Vol. 1). New York : Cambridge University Press.
Colby, A., Kohlberg, L., & Kauffman, K. (1987). Theoretical introduction to the measurement of moral judgment. In A. Colby & L. Kohlberg, The measurement of moral judgment: Theoretical foundations and research validation (Vol. 1, pp. 1-61). New York: Cambridge University Press.
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334.
Darley, J. M., & Latané, B. (1968). Bystander intervention in emergencies: Diffusion of responsibility. Journal of Personality and Social Psychology, 8(4), 377-383.
Festinger, L., Riecken, H. W., & Schachter, S. (1956). When prophecy fails. Minneapolis: University of Minnesota Press.
Frum, D. (2003). The right man: The surprise presidency of George W. Bush. New York: Random House.
Gelderman, C. (1997). All the presidents' words: The bully pulpit and the creation of the virtual presidency. New York: Walker .
Gilligan, C. (1988). Preface. In C. Gillian, J. V. Ward, & J. M. Taylor, Mapping the moral domain (pp. i-v). Cambridge, MA: Center for the Study of Gender, Education and Human Development, Harvard University Graduate School of Education.
Gilligan, C. (1993). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.
Gilligan, C., & Attanucci, J. (1988). Two moral orientations. In C. Gillian, J. V. Ward, & J. M. Taylor, Mapping the moral domain (pp. 73-86). Cambridge, MA: Center for the Study of Gender, Education and Human Development, Harvard University Graduate School of Education.
Hughes, K. (2004). Ten minutes from normal. New York: Viking.
Jaffee, S., & Hyde, J. S. (2000). Gender differences in moral orientation: A meta-analysis. Psychological Bulletin, 126(5), 703-726.
Janis, I. L. (1982). Groupthink (2nd ed.). Boston: Houghton Mifflin.
Jones, A. P., Johnson, L. A., Butler , M. C., & Main , D. S. (1983). Apples and oranges: An empirical comparison of commonly used indices of interrater agreement. Academy of Management Journal, 26(3), 507-519.
Johnston, D. K. (1988). Adolescents' solutions to dilemmas in fables: Two moral orientations – Two problem solving strategies. In C. Gilligan, J. V. Ward, & J. M. Taylor (Eds.), Mapping the moral domain (pp. 49-71). Cambridge, MA: Harvard University .
Kennan, N., & Hadley, M. (1986). The creation of political leaders in the context of American politics in the 1970s and 1980s. In C. F. Graumann & S. Moscovici (Eds.), Changing conceptions of leadership (pp. 145-169). New York : Springer Verlag.
Kennedy, J. F. (1962a, October 22). Radio and television report to the American people on the Soviet arms buildup in Cuba. In Office of the Federal Register, Public papers of the presidents of the United States : John F. Kennedy (Vol. 1962, pp. 806-809). Washington, DC: U.S. Government Printing Office.
Kennedy, J. F. (1962b, November 2). Radio and television remarks on the dismantling of Soviet missile bases in Cuba. In Office of the Federal Register, Public papers of the presidents of the United States: John F. Kennedy (Vol. 1962, p. 821). Washington, DC: U.S. Government Printing Office.
Latané, B., & Darley, J. M. (1968). Group inhibition of bystander intervention in emergencies. Journal of Personality and Social Psychology, 10(3), 215-221.
Liddell, D. L. (1990). Measure of moral orientation: Construction of an objective instrument measuring care and justice, with an investigation of gender differences. Dissertation Abstracts International 51(11), 5626B. (UMI No. AAT9109093)
Lincoln, A. (1953). Speech at Cincinnati, Ohio. In R. P. Basler (Ed.), The collected works of Abraham Lincoln (Vol. 4, pp. 197-200). New Brunswick, NJ: Rutgers University Press. (Original work published February 12, 1861)
Locke, E. A. (2003). Leadership: Starting at the top. In C. L. Pearce & J. A. Conger (Eds.), Shared leadership (pp. 271-284). Thousand Oaks CA: Sage.
Lombard , M., Snyder-Duch, J., & Bracken, C. C. (2004). Practical resources for assessing and reporting intercoder reliability in content analysis research projects. Retrieved April 19, 2004, from Temple University Web site: http://www.temple.edu/mmc/reliability/
Lyons, N. P. (1982). Conceptions of self and morality and modes of moral choice: Identifying justice and care in judgments of actual moral dilemmas. Unpublished doctoral dissertation, Harvard University, Cambridge, MA .
Lyons, N. P. (1988). Two perspectives: On self, relationships, and morality. In C. Gilligan, J. V. Ward, & J. M. Taylor (Eds.), Mapping the moral domain (pp. 21-48). Cambridge, MA: Harvard University .
Medhurst, M. J. (2003). Presidential speechwriting: Ten myths that plague modern scholarship. In K. Ritter & M. J. Medhurst (Eds.), Presidential speechwriting: From the new deal to the Reagan revolution and beyond (pp. 3-19). College Station, TX: Texas A&M University Press.
Meir, G. (1973, October 6). Broadcast to the nation [Radio broadcast]. Retrieved March 31, 2003, from http://www.mfa.gov.il
Milgram, S. (1969). Obedience to authority: An experimental view. New York: Harper & Row.
National Archives and Records Administration. (2001). Our heritage in documents: FDR's “day of infamy” speech: Crafting a call to arms. Prologue: Quarterly of the National Archives and Records Administration, 33(4). Retrieved March 30, 2002 , from http://www.nara .gov/publications/prologue/infamy.html. Website changed to http://www. archives.gov /publications/prologue/winter_2001_crafting_day_of_infamy_speech.html
Oettingen, G. (1995). Explanatory styles in the context of culture. In G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style (pp. 209-224). Hillsdale, NJ: Lawrence Erlbaum Associates.
O'Toole, J., Galbraith, J., & Lawler, E. E., III. (2003). The promise and pitfalls of shared leadership: When two (or more) heads are better than one. In C. L. Pearce & J. A. Conger (Eds.), Shared leadership (pp. 250-267). Thousand Oaks, CA: Sage.
Pearce, C. L., & Conger, J. A. (2003). All those years ago: The historical underpinnings of shared leadership. In C. L. Pearce & J. A. Conger, (Eds.), Shared leadership (pp. 1-18). Thousand Oaks, CA: Sage.
Peterson, C. (1992). Explanatory style. In C. P. Smith (Ed.), Motivation and personality: Handbook of thematic content analysis (pp. 376-382). New York: Cambridge University Press.
Peterson, C., Schulman, P., Castellon, C., & Seligman, M. E. P. (1992). The explanatory style scoring manual. In C. P. Smith (Ed.), Motivation and personality: Handbook of thematic content analysis (pp. 383-400). New York: Cambridge University Press.
Piaget, J. (1976). The grasp of consciousness: Action and concept in the young child. Cambridge, MA: Harvard University Press.
Piaget, J. (1997). The moral judgment of the child. New York: Free Press.
Post, J., Walker, S., & Winter, D. (2003). Profiling political leaders: An introduction. In J. Post (Ed.), The psychological assessment of political leaders (pp. 1-7). Ann Arbor: University of Michigan Press.
Ritter, K., & Medhurst, M. J. (Eds.). (2003). Presidential speechwriting: From the new deal to the Reagan revolution and beyond. College Station, TX: Texas A&M University Press.
Roosevelt, F. D. (1950a). Address to the Congress asking that a state of war be declared between the United States and Japan. In S. I. Rosenman (Ed.), The public papers and addresses of Franklin D. Roosevelt: 1941 vol. (pp. 514-516). New York: Harper & Brothers. (Original work published December 8, 1941)
Roosevelt, F. D. (1950b). Radio address commemorating the 150th anniversary of the ratification of the Bill of Rights. In S. I. Rosenman (Ed.), The public papers and addresses of Franklin D. Roosevelt: 1941 vol. (pp. 554-556). New York: Harper & Brothers. (Original work published December 15, 1941)
Satterfield, J. M. (1998). Cognitive-affective states predict military and political aggression and risk taking: A content analysis of Churchill, Hitler, Roosevelt, and Stalin. Journal of Conflict Resolution, 42(6), 667-690.
Schein, E. H. (1992). Organizational culture and leadership (2nd ed.). San Francisco: Jossey-Bass.
Schulman, P., Castellon, C., & Seligman, M. E. P. (1989). Assessing explanatory style: The content analysis of verbatim explanations and the attributional style questionnaire. Behaviour Research and Therapy, 27(5), 505-512.
Seligman, M. E. P. (1998). Learned optimism: How to change your mind and your life. New York: Pocket Books.
Shakespeare, W. (1997). The merchant of Venice . In G. B. Evans, The Riverside Shakespeare (2nd ed.): The complete works (pp. 288-319). Boston: Houghton Mifflin. (Original work published circa 1597)
Smith, C. P. (Ed.). (1992). Motivation and personality: Handbook of thematic content analysis. New York: Cambridge University Press.
Smith, C. P., Feld, S. C., & Franz, C. E. (1992). Methodological considerations: Steps in research employing content analysis systems. In C. P. Smith (Ed.), Motivation and personality: Handbook of thematic content analysis (pp. 515-536). New York: Cambridge University Press.
SPSS. (1999). SPSS base 10.0 applications guide. Chicago, IL: SPSS, Inc.
Stogdill, R. M. (1974). Handbook of leadership: A survey of theory and research. New York: Free Press.
Suedfeld, P., & Bluck, S. (1988). Changes in integrative complexity prior to surprise attacks. Journal of Conflict Resolution, 32(4), 626-635.
Suedfeld, P., Guttieri, K., & Tetlock, P. E. (2003). Assessing integrative complexity at a distance: Archival analyses of thinking and decision making. In J. M. Post (Ed.), The psychological assessment of political leaders (pp. 246-270). Ann Arbor, MI: University of Michigan Press.
Suedfeld, P., & Tetlock, P. E. (1977). Integrative complexity of communications in international crises. Journal of Conflict Resolution, 21(1), 169-184.
Uebersax, J. (2002). Kappa coefficients. Retrieved March 25, 2004, from http://ourworld.compuserve.com/homepages/jsuebersax/kappa.htm
Wallace, M. D., & Suedfeld, P. (1988). Leadership performance in crisis: The longevity-complexity link. International Studies Quarterly, 32(4), 439-451.
Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. (2000). Unobtrusive measures (Rev. ed.). Thousand Oaks, CA: Sage.
Wilson , W. (1983a). An address to a joint session of Congress. In A. S. Link (Ed.), The papers of Woodrow Wilson (Vol. 41, pp. 108-112). Princeton, NJ: Princeton University Press. (Original work published February 13, 1917)
Wilson , W. (1983b). An address to a joint session of Congress. In A. S. Link (Ed.), The papers of Woodrow Wilson (Vol. 41, pp. 283-287). Princeton, NJ: Princeton University Press. (Original work published February 26, 1917)
Wilson , W. (1983c). A Flag Day address. In A. S. Link (Ed.), The papers of Woodrow Wilson (Vol. 42, pp. 498-504). Princeton, NJ: Princeton University Press. (Original work published June 14, 1917)
Winter, D. G. (2003). Measuring the motives of political actors at a distance. In J. M. Post (Ed.), The psychological assessment of political leaders (pp. 153-177). Ann Arbor, MI: University of Michigan Press.
Winter, D. G., & Stewart, A. J. (1977). Content analysis as a technique for assessing political leaders. In M. G. Hermann (Ed.), A psychological examination of political leaders (pp. 27-61). New York: Free Press.
Zullow, H. M., Oettingen, G., Peterson, C., & Seligman, M. E. P. (1988). Pessimistic explanatory style in the historical record: CAVing LBJ, presidential candidates, and East virus West Berlin. American Psychologist, 43(9), 673-682.
Zullow, H. M., & Seligman, M. E. P. (1990). Pessimistic rumination predicts defeat of presidential candidates, 1900-1984. Psychological Inquiry, 1(1), 52-61.