USF Libraries
USF Digital Collections

Making sense of complex system failure


Material Information

Making sense of complex system failure the case of 911
Physical Description:
Cooper, Sandra M
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
September 11 2001
System failure
System resilience
Dissertations, Academic -- Communication -- Doctoral -- USF   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


ABSTRACT: The National Commission on Terrorist Attacks attributed the September 11 attacks on the U.S. homeland to the terrorists' exploitation of "deep institutional failings." These findings are similar to the conclusions of the Presidential Commission investigating the 1986 Challenger accident and the Columbia Accident Investigative Board (2003). Generally Commissions aim to provide the fullest possible account of events contributing to the catastrophe under investigation and to identify lessons learned, but avoid specifying responsibility and accountability. For this reason, various commission reports have been criticized for being abstract and shallow. These criticisms make a valid point. How commissions make sense of failures has real consequences in terms of preventing reoccurrences. If these accounts do not satisfactorily address the question, How did this happen?, clear prophylactic measures for the future also remain unspecified.^ ^This dissertation calls into question the usefulness of current constructions of system failure that focus solely on the abstract role of the institution or system in creating the conditions for failure. For the purpose of acquiring insight into our current narratives of system failure and accountability, the 9/11 Public Hearing Transcripts are analyzed. This research is a qualitative textual analysis of excerpts from the Transcripts related to both pre-9/11 intelligence failures and accountability. Using Weick's view of sensemaking to gain a better understanding of our current constructions of system failure/resilience and accountability, this research identifies the dominant constructions of pre-9/11 intelligence failures and accountability that are documented in the 9/11 Public Hearing Transcripts and the sensemaking resources that reinforce and solidify these constructions. Verbatim excerpts from the 9/11 documents are included to support claims.^ ^The theory of autopoiesis, a form of systems theory, is introduced as an alternative resource for constructing narratives on system-environment relationships and accountability. Leadership practices that foster system resilience and individual accountability for system-wide performance are presented.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Sandra M. Cooper.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 181 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001935403
oclc - 226034297
usfldc doi - E14-SFE0002015
usfldc handle - e14.2015
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001935403
003 fts
005 20080423095741.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080423s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002015
1 100
Cooper, Sandra M.
0 245
Making sense of complex system failure :
b the case of 9/11
h [electronic resource] /
by Sandra M. Cooper.
[Tampa, Fla] :
University of South Florida,
3 520
ABSTRACT: The National Commission on Terrorist Attacks attributed the September 11 attacks on the U.S. homeland to the terrorists' exploitation of "deep institutional failings." These findings are similar to the conclusions of the Presidential Commission investigating the 1986 Challenger accident and the Columbia Accident Investigative Board (2003). Generally Commissions aim to provide the fullest possible account of events contributing to the catastrophe under investigation and to identify lessons learned, but avoid specifying responsibility and accountability. For this reason, various commission reports have been criticized for being abstract and shallow. These criticisms make a valid point. How commissions make sense of failures has real consequences in terms of preventing reoccurrences. If these accounts do not satisfactorily address the question, How did this happen?, clear prophylactic measures for the future also remain unspecified.^ ^This dissertation calls into question the usefulness of current constructions of system failure that focus solely on the abstract role of the institution or system in creating the conditions for failure. For the purpose of acquiring insight into our current narratives of system failure and accountability, the 9/11 Public Hearing Transcripts are analyzed. This research is a qualitative textual analysis of excerpts from the Transcripts related to both pre-9/11 intelligence failures and accountability. Using Weick's view of sensemaking to gain a better understanding of our current constructions of system failure/resilience and accountability, this research identifies the dominant constructions of pre-9/11 intelligence failures and accountability that are documented in the 9/11 Public Hearing Transcripts and the sensemaking resources that reinforce and solidify these constructions. Verbatim excerpts from the 9/11 documents are included to support claims.^ ^The theory of autopoiesis, a form of systems theory, is introduced as an alternative resource for constructing narratives on system-environment relationships and accountability. Leadership practices that foster system resilience and individual accountability for system-wide performance are presented.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 181 pages.
Includes vita.
Adviser: Eric Eisenberg, Ph.D.
September 11 2001.
System failure.
System resilience.
Dissertations, Academic
x Communication
t USF Electronic Theses and Dissertations.
4 856


Making Sense of Complex System Failure: The Case of 9/11 by Sandra M. Cooper A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Communication College of Arts and Sciences University of South Florida Major Professor: Eric Eisenberg, Ph.D. Kenneth Cissna, Ph.D. Frederick Steier, Ph.D. Marilyn Myerson, Ph.D. Date of Approval: April 3, 2007 Keywords: accountability, autopoiesis, Se ptember 11, 2001, sensemaking, system failure, system resilience Copyright 2007, Sandra M. Cooper


iTable of Contents List of Tables iv List of Figures v Abstract vi Chapter One Making Sense of Complex System Failure 1 Systems Thinking 4 Understanding 21st Century Organizational Issues 5 The Theory and Language of Systems Thinking 8 Autopoietic Systems 9 System/Environment Interconnections: Meaning and Resonance 12 Requisite Variety 15 Assessing and Influencin g System Resilience 17 Structures of Accountability 19 Individual and Collective Performance 20 Interpretive Choice 21 Enactment Choice 21 Individual and Colle ctive Accountability 22 A New Image of System Resilience and Accountability 23 Organizational Theory 26 Organizational Rese arch and Practice 32 Chapter Two The Case of 9/11 3 34 Sensemaking 35 Sensemaking is Grounded in Identity Construction 36 Sensemaking is Retrospective 37 Sensemaking Enacts Se nsible Environments 37 Sensemaking is Social 39 Sensemaking is Ongoing 40 Sensemaking is Focused on and by Extracted Cues 40 Sensemaking is Driven by Plausibility rather than Accuracy 41 Discussion and Summary 42 Research Protocols 43 Excerpts from the 9/11 Public Hearing Transcripts 43 Analysis 44 Overview of Similar Studies 46 The Case of 9/11 48


iiChapter Three The Construction of Intelligence in the 9/11 Public Hearing Transcripts 52 Definitions 55 Intelligence 55 Intelligence Production 56 Intelligence Community 57 Intelligence: Constructions in the 9/11 Public Hearing Transcripts 63 Flow of Information 64 Actionable Intelligence 71 Coordination and Sh aring of Information 77 Access to Information 81 A Law Enforcement and Prosecutor ial Approach to Fighting Terrorism 84 Summary and Discussion 86 Chapter Four Constructions of Accountability in the 9/11 Public Transcripts 93 Who was Accountable for September 11, 2001? 95 Oversight 100 Oversight: The Intelligence Function 104 Accountability: Constructions in th e 9/11 Public Hearing Transcripts 108 Summary and Discussion 118 Chapter Five Alternative Vocabularies for Maki ng Sense of System Resilience: Mechanism and Autopoiesis 123 Mechanistic View of Organizations, Intelligence Failures, and Accountability 126 Mechanistic Thinking: Pre-9/11 Intelligence Failures and Accountability 129 Mechanistic Thinking: Discussion of the 9/11 Public Hearing Transcripts 131 Autopoiesis: An Alternative Descr iptor for System Resilience and Accountability 134 Building System Resilience: An Autopoietic/Sensemaking Model 139 Self-reference 140 Ambivalence 141 Self-reference and Ambivalence: The Case of 9/11 142 An Autopoietic/Sensemaking Model 144 Practical Implications for Futu re Organizational Practice 146 The Role of Leaders: Bu ilding System Consciousness 147 Leadership Practice: Develo ping Key Principles that Express the Overall Identity of the System 148 Leadership Practice: Saturating the Organization with Consistent Messages about the Organization 149


iii The Role of Leaders: Desi gning Heedful Interrelating 152 Leadership Practice: Engaging a Greater Number and Diversity of People 153 Leadership Practice: Loosen ing the Controls on Engagement, Participation, and Action 155 The Role of Leaders: Legitimizi ng Ambivalence in the Workplace 160 Discussion and Conclusions 162 References 167 Appendices 174 Appendix A: Dates of 9/11 Public Hearings and URLS for PDF Records 175 Appendix B: Members of the Nation al Commission on Te rrorist Attacks On the United States 177 Appendix C: List of Witne sses Cited in this Document 178 About the Author End Page


ivList of Tables Table 1 The Influence of Traditional Science on Organizational Research and Practice 6 Table 2 Application of Systems Theory to Social Systems 18 Table 3 Social System Research 31 Table 4 The Relationship be tween ESR and Sensemaking 45 Table 5 Building System Resilience: Co ntributions of Auto poiesis, a Form of Systems Theory 136 Table 6 Autopoiesis: Assumptions an d Implications for Organizational Practices 138 Table 7 Building System Resilience: An Autopoietic/Sensemaking Model 145 Table 8 Building System Resi lience: Leadership 147


vList of Figures Figure 1. ESR Sequence 45 Figure 2. Pre-9/11 U>S. Intelligence Community 58 Figure 3. Information that was genera lly known before September 11, 2001 74 Figure 4. Resilience, in Terms of the ESR Sensemaking Sequence 139 Figure 5. Building System Resilience: An Autopoie tic/Sensemaking Model 145


viMaking Sense of Complex Systems: The Case of 9/11 Sandra M. Cooper ABSTRACT The National Commi ssion on Terrorist At tacks attributed the September 11 attacks on the U.S. homeland to the terrorists’ exploitation of “deep institutional failings.” These findings are si milar to the conclusions of the Presidential Commission investigating the 1986 Challenger accident an d the Columbia Accident Investigative Board (2003). Generally Commissions aim to provide the fullest possible account of events contributing to the catastrophe under investigation and to identify lessons learned, but avoid specifying responsibility and accountability. For this reason, various commission reports have been criticized for being abstract and shallow. These criticisms make a valid point. How commissions make sense of failures has real consequences in terms of preventing reoccu rrences. If these accounts do not satisfactorily address the question, How did this happen?, clear prophylactic measures for the future also remain unspecified. This dissertation calls into question the usefulness of current co nstructions of system failure that focus solely on the abstract role of the institution or system in creating the conditions for failure. For the purpose of acquiring insi ght into our current narratives of system failure and accountability, the 9/11 Public Hearing Transcripts are analyzed. This research is a qualitative textual analysis of excerpts from the Transcripts related to both pre-9/11 intelligence failures and accountability. Using Weick’s view of sensemaking to gain a better understanding of our current constructions of system failure/resilience and accountability, this research identifies the dominant


viiconstructions of pre-9/11 intelligence failures and accountability that are documented in the 9/11 Public Hearing Transcripts and the sensemaking resources that reinforce and solidify these constructions. Verbatim excerpts from the 9/11 documents are included to support claims. The theory of autopoiesis, a form of systems theory, is introduced as an altern ative resource for constructing narratives on system-environment relationships and accountability. Leadership practices that foster system resilience an d individual accoun tability for system-wide performance are presented.


1CHAPTER ONE MAKING SENSE OF COMPLEX SYSTEM FAILURE On September 11, 2001, four U.S. planes hijacked by terrorists crashed into the World Trade Center, the Pentagon, and a field near Shanksville, a small town in Pennsylvania. The loss from these attacks was staggering. Within a few hours nearly 3,000 people died, an estimated 1,60 9 people lost a spouse or a partner, and approximately 3,051 children lost one or both of their parents. Twenty per cent of Americans, it is estimated, knew someone hur t or killed in the at tacks (“9/11 by the Numbers,” 2005, para. 2, 19-21). On 9/11/2 001, in a matter of mi nutes, millions of lives were irreversibly altered. The magnitude of the devastation unl eashed by the September 11 terrorist attacks on the United States remains incomprehensible. What is more, the lack of mobilization to the frequent, although frag mentary, threats to the U.S. homeland before 9/11 continues to be di fficult to understand: In sum, the domestic agencies never mobilized in response to the threat. They did not have the direction, and di d not have a plan to institute. The borders were not hardened. Transporta tion systems were not fortified. Electronic surveillance was not targeted against a domestic threat. State and local law enforcement were not marshale d to augment the FBI’s efforts. The public was not warned. (National Commi ssion on Terrorist Attacks upon the United States, 2004, p. 265) In the end, the Nati onal Commission on Terrorist At tacks upon the United States (2004), i.e., the 9/11 Commission, observ ed, “The terrorists exploited deep institutional failings within our government” (p. 265). A similar explanation had previously been constructed by the Presidential Commission investigating the 1986 Challenger accident—that is, the “human failures behind the [shuttle’s] technical failure” (Tompkins, 1993, p. 127) were “rooted in


2[the] history” of the problems in communica tion at the Marshall Space Flight Center (MSFC) (Presidential Commissi on on the Space Shuttle Challenger Accident, 1986, p. 120). Similarly, the Columbia accident, according to the Columbia Accident Investigation Board (2003), was “likely rooted to some degree in NASA’s history and the human flight program’s culture” (p. 9). The findings of these three investigations are also similar in that: 1. Each of the government-appointed Commi ssions aimed to provide a complete and frank account of the events leading up to a catastrophe. 2. None of the Commissions assi gned individual accountabi lity for those events. 3. All have been criticized for crafting o fficial reports that are abstract, bloodless (Tompkins, 1993, p. 128), and shallow, as if organizations weren’t populated by human beings “with names, power, and ob ligations who are charged with making [them] work” (Meier, Jones, & Moyers, 2004, para.6). Generally, government-appointed Commi ssions aim to provide the fullest possible account of events contributing to the catastrophe under investigation and to identify lessons learned, but in the interest of not engaging in a blame game, they typically avoid specifying responsibility and accountability. For this reason, various commission reports have been criticized for being abstract and shallow. These criticisms make a valid point. How commissions make sense of failures has real consequences in terms of preventing reoccu rrences. If these accounts, in the end, do not satisfactorily address the question How did this happen? clear prophylactic measures for the future also remain unspecified. In light of the national need to ha ve measures for preventing large scale system failures similar to those on Septem ber 11, 2001, this dissertation calls into question the usefulness of retrospective constructions of system failure that do not take into account that small changes in a system can also lead, over a period of time, to large-scale consequences. This is often referred to as the butterfly effect—


3that is, “a butterfly stirring the air today in Beijing can cause a storm in New York next month” (Capra, 1996, p. 134). Si milarly, we have painfully learned how seemingly non-consequential decisions such as flying a shuttle with minutely eroded O-rings can lead to a catastrophic accident When only systems are held responsible for performance, as is the case in the official reports on the Challenger and Columbia accidents and 9/11, the relationship be tween individual ac tions and system performance is evasive, creating the potential for inaction, ineffectiveness, and inadequate performance (Connors, Smith, & Hickman, 1994, p. 23). This dissertation analyzes the retros pective sensemakin g of the events leading to September 11 that is documented in the 9/11 Public Hearing Transcripts, for the purpose of acquiring insight into our current narratives of system failure and identifying alternative narratives for expr essing system account ability and system resilience. Historically, system performanc e and accountability have been dispersed throughout organizations, as a result of division of labor and specialization of knowledge and skills. Under these circumstances, each in dividual is a ccountable for achieving the precise goals of his or her position. This approach to organizations, work, and accountability is mo st effective when the enviro nment is stable enough to ensure that the routine, st andardized performance of in dividuals is summative and cumulatively attains organizational goals. Traditionally, our understanding of system performance and accountability has assumed a stable, predictable environment. Now that we are immersed in an age of technology that enables complex global interdependencies, including events such as international terrorism, one of the major challenges facing orga nizations is to generate fresh ideas and approaches to understanding and fostering accountability in complex, dynamic environments. One possibility is a definition of accountability that aims for a pervasive sense of responsibility and ownership for the surviv al and success of the whole system, as well as for one’s own job. From this holistic perspective, each individual is


4accountable for achieving the goals of the sy stem. This requires all individuals to look beyond their jobs and be keenly aware of how their wo rk interacts with the work of others to achieve the mission of the whole organization, what Weick and Roberts (1993) called heedful interrelating In the absence of this sort of accountability, there is a breakdown of “the nervous sy stem that [connects] the brain to the muscles” (Tompkins, 1993, p. 61), that is, communication that fosters system resilience. The 9/11 Public Hearing Transcripts are artifacts of sensemaking processes that reverberate with themes of accountability, and its breakdowns. After 9/11, the National Commission on Terrori st Attacks upon the United States was established to “investigate ‘facts and circ umstances relating to the te rrorist attacks of September 11, 2001’” (National Commission, 2004, p. x v), twelve public hearings were conducted, testimonies were transcribed and released to the public, and a report was issued. Using Weick’s view of sens emaking (1979, 1995; Weick, Sutcliffe, & Obstfeld, 2005), I aim to reve al the communication beliefs and practices that inform the constructions of institutional failure and accountability in the 9/11 Public Hearing Transcripts. By examining ou r current narratives, I aim to discover opportunities for constructing new distinctions and descripti ons that contribute to our understanding of the kinds of accountability that foster organizational resilien ce. A new, systemic model, autopoietic systems, is presented an d tested as an alternative descriptor for system resilience an d accountability. Systems Thinking The remainder of this chapter offers sy stems thinking as a useful theoretical framework for researching large scale or ganizational failure/resilience and accountability. First, systems thinking is presented as an alternative approach to understanding contemporary organization al issues. Second, the theory and language of systems theory th at are relevant to this project are explored. Third,


5systems thinking is presented as a framew ork for assessing and influencing system resilience. Fourth, an approach to accountability that is grounded in systems thinking is presented. Autopoiesis, a form of systems theory, offers a new image of accountability, one that redefines traditio nal management practices. This section concludes with a look at this new image of organizational relati onships. Excerpts from Tompkins’s (1993, 2005) studies of organizational comm unication at the Marshall Space Flight Center in Huntsville, Alabama during the Apollo era (1960s) and after the 1986 Challenger and 2003 Columbia accidents, as well as related reports ( Columbia Accident Investigation Board, 2003; Presidenti al Commission, 1986), are used to illustrate systems concep ts that offer new ways of considering organizational performance and accountability. Understanding 21st Century Organizational Issues Since the end of the 18th century, the major tenets of the Renaissance, Reformation, Scientific Revolution, and Enlightenment have shaped the lives of the Western world; these include atom ism, reductionism, mechanism, anthropomorphism, instrumentality, standardization, bureaucratization, centralization, universal principles, and a privileging of the masculine. Table 1 is my summary of this mode of th ought and its influence on th e research and practices of modern organizations (Carey, 1989; Cobb, 1926; Eisler, 1987; Morgan, 1997; Spretnak, 1997).


6Table 1 The influence of traditional science on organizational research and practice Research Mechanistic mode of thought Application of mechanistic thinking to organizations What phenomena do we observe and study? A set of mechanistic relations: Relations between clearly defined parts that have some determinate order. Use of labor, machines, materials, instruments, and tools (e.g., tasks, roles, goals) that have been invented to aid in attaining a goal. How is effectiveness defined? Machine-like operations. Rationalized, efficient, reliable, operations. Where do we target interventions? Routinize and mechanize human thought and action. The division, specialization, and standardization of machinery, materials, instruments, tools, and labor. What are indicators of success? Prediction and control; the ordering of our lives; the advancement of civilization. Political, economic, and religious power, control, and influence.


7 Traditional mechanistic approaches to organizations are effective: (a) when there is a straightforwar d task to perform; (b) when the environment is stable enough to ensure that the products produced will be appropriate ones; (c) when one wishes to produce exactly the same product time and again; (d) when precision is at a premium; and (e) when the human “machine” parts are compliant and behave as they have been designed to do. (Morgan, 1997, p. 27) Conversely, mechanistically structured organizations ha ve difficulty responding effectively to the continuous social, economic, and tec hnological changes of the 21st century. Mechanistic structures that are ill-suited for dynamic environments include pre-determined goals, highly-s pecialized functional divisions, hierarchical levels, and task-based position description s (Morgan, 1997, pp. 28-29): 1. Performance that is guided by predeter mined goals is not easily shifted to response patterns that would effectivel y address changes in the environment. 2. Functional silos and hierarchical levels often co ntribute to distortions in information and delays in coordinated responses to unfamiliar situations. 3. Highly specialized functional divisions and hierarchical levels foster myopic views and unilateral actions that often undercut the responsiveness of other functional areas, or the whole orga nization (e.g., when budget decisions heedlessly debilitate an organization’s training, health and safety, and security functions). 4. The precise position descriptions found in most hierarchic al organizations encourage many organizational members to ignore the unfa miliar and “adopt mindless, unquestioning attitudes such as ‘it’s not my job to worry about that,’ ‘that’s his responsibility, not mine,’ or ‘I’m here to do what I’m told’” (Morgan, 1997, pp. 28-30).


8Although many contemporary organizational theorists and others have argued that the concepts of mechanism are in adequate to deal with our globally interconnected world (Eisler, 1987; Mo rgan, 1997; Spretnak, 1997; Wheatley, 1994), most of Western culture continues to su bscribe to them. People rely on these old vocabularies because they are not yet fluent in new ones. This is most evident in our modern organizations. Challenging us to reshape our current worldview, advocates of systems thinking offer new ways of consider ing organizational issues, such as chaos/order, control/autonomy structures, information management, planning, prediction (Wheatley, 1994, p. ix). The traditional way of thinking has been called mechanistic, reductionistic, an d atomistic. Conversely, systems thinking focuses on the whole, connections, relation ships, and context, and has been called holistic, organismic, and ecological. The Theory and Language of Systems Thinking System failures have predominantly been attributed to institutional, historical, and cultural failings. If, however, we acquire a deeper understanding of how both individuals and the system ar e accountable for system failure, strategies for building system resilience can be enhanced. A th eoretical framework th at supports this definition of accountability and system resilience is autopoiesis, a form of systems theory. This theory: 1. Focuses on “ organization and its dynamics, based on the ’triggering’ effects of information transmission” (B uckley, 1967, pp. 1-2). 2. Privileges the relationship of th e system with its environment. 3. Relates communication processe s and system resilience. 4. Provides a theoretical framework for system assessment, prediction, and interventions. This section reviews four systems theory conc epts that are relevant to this project. Conceptualized as a utopoietic systems of communication, social systems can, and


9must, reinvent themselves. Meaning and resonance explain the role of communication in system/environment in terconnections, learning, and system resilience. Requisite variety provides a theoretical framework for strengthening a system’s ability to respond to dynamic en vironments. Collectively, these concepts from systems theory provide a communicati on framework for assessing, predicting, and preventing system failure. Autopoietic Systems Concepts of organismic biological systems have been applied to social systems (Buckley, 1967; Capra, 1996; Luhmann, 1990; McWhin ney, 1993; Varela, Thompson, & Rosch, 1991; Wheatley, 1994). Whereas living system s operate in the medium of life, social systems operate in the medium of conversation (Luhmann, 1990). In addition to being characterized as systems, liv ing and social systems are specifically characterized as open systems; they each enga ge in interchanges with the environment in order to maintain sy stem viability (Buc kley, 1967, p. 50). Whereas living systems characteristically wo rk to maintain their genetically given structures within fairly defined limits (homeostasis), higher level social systems are characterized primarily by th eir morphogenic properties (Buckley, 1967, pp. 4-5). Rather than preserve a given, fixed struct ure, social systems adapt through growth, development, change, and continuous learning; that is, they create, elaborate, or change structures in order to remain vi able as ongoing systems (Buckley, 1967, p. 5). However, while continuously reinvent ing themselves, they also continue to maintain the integrity of their structures and continue to be recognizable as themselves. This developmental process is called autopoiesis from the Greek word for self reproduction. Reproduction, as L uhmann (1990) points out, is a production of products. The medium of social system s is communication; consequent ly, a social system is an autopoietic system of communi cation that, in a recursiv e network of communication,


10not only constructs its own unity, structur e, and communication, but also can learn and change itself through its communication. The self organizing principle of autopoietic social sy stems can be illustrated by the various configurations of family in our society: We have married and unmarried couples; biological adopted, step, and foster children; heterosexual and same sex parents and partners; and sing le-parent, two-parent, and extended families. However it is configured, a fa mily does not exist until people come together, assume roles, and be gin to interact. Over time those interactions lead the participants to seek definition about their relationship, that is, what they are trying to do, and are capable of doing, together. This is the self organizing process of autopoietic systems; in a re cursive network of communicati on, the system constructs its unity, structure, and communication. Autopoietic systems can also learn and change themselves. Continue to consider, as an example, certain family sy stems. The children grow older, go to school, and engage in academic and extracu rricular activities. Some graduate, some marry, some have their own children. Each conscious or unconscious choice made along the way by the members of the family is some kind of real-time reorganization for that family system—for example, who is in the family and what they do together; the formal and informal role s that the members assume; the quantity, forms, and content of family interactions; and the quality of relationships among family members. On a daily basis, these unique interactions constitute, maintain, perpetuate, and characterize the family. This year, for example, my family is in the process of reorganizing (i.e., reinventing) itself as my daughter transitions from middle school to high school and I temporarily shift from full-time employee to fulltime graduate student. Thes e are significant reconstructi ons of major pieces of our lives; as a result, how much time we sp end together, how we use that time, the distribution of household responsibilities, and so forth are being renegotiated in day-


11to-day, formal and informal conversations. Consciously and un consciously, through our interrelated words, meanings, and action s, my daughter and I not only construct a unity and structure for our family system, we also change it. At the same time, we also fundamentally remain recognizable as the Cooper family. In larger systems such as universities, government agencies and Fortune 500 companies, autopoiesis is harder to see, as large organizations often appear as powerful super realities imposed on their members. To the contrary, it is the daily interactions within a system, whatever its size, that creates an d maintains the system’s reality. Carey’s (1989) discussion of the ritual view of communication aligns with the conceptualization of social systems as au topoietic systems of communication that create, maintain, and perpetuate themselves. Viewed as rituals, communication is a pattern of actions or words with some re gularity and precisen ess that reinforces social cohesion and the maintenance of soci ety. A newspaper, for example, portrays and confirms particular views of the world. As such, it offers form, order, and tone to life. It does not describe the world, but rather portrays multiple views of it and “invites our participation on the basis of ou r assuming, often vicari ously, social roles within it” (p. 21). Through portrayal, invitation, and reader participation, the newspaper contributes to the creation, main tenance, and perpetuation of particular views of the world (p. 33). A newspaper is one of countless human-made objects that serve to perpetuate (or contest) our understanding of the world. Other examples include novels, plays, art, events customs, and traditio ns. Organizational examples include vision, va lues, and mission statements ; policies and procedures; bulletin boards; and parking spaces. Exampl es also include the communication that Tompkins (1993) identifies as the essence of organiza tions—upward communication, downward communication, fa ce-to-face communi cation, mediated communication, and decentralized and centraliz ed communication (pp. 17-28) In this way, a social system is an autopoietic, recursive netw ork of communications where communication


12both produces, and is produced by, the system (Berger, 1967; Berger & Luckmann, 1966; Blumer, 1969; Mead, 1967). System/Environment Interconnections: Meaning and Resonance An autopoietic system, by way of re cursive networks of communication, creates its structure, unity, and communi cation (Luhmann, 1990) as its members go about defining their relationship to on e another and identify ing what they are capable of doing together. By means of communication, the system and its components can be defined; and by process of exclusion, what is not of the system— the so-called environment —can be identified. In practice, however, drawing bound aries between the system and the environment is never easy, and system/boundary designations are always fuzzy and shifting. First, the system/environment boundaries are always politically inspired and incessantly negotiated. Similar to a rubber band, the system/environment boundary expands and contracts, taking on various shapes, as the boundaries (inclusion/exclusion) are continuously nego tiated. Second, none of the members is engaged solely with one system as all members participate to some degree in multiple systems, processes, and relationsh ips. With autopoietic systems, there is always the question: Who is in (and to what extent), and who is out? Although the nature of a system ma y change with time, the process of drawing boundaries remains important. Considering a university as an example, the question is Who is part of the university system? The university/environment boundaries can be drawn so that professo rs, staff, and students constitute the system: Classes of professors and students supported by staff, enact a world—that is, a recursive system of communication wi th structure and unity. Arguments can also be made that faculty and staff are in, and students are part of the environment; or faculty and students are the system an d staff, like vendors, are part of the environment. The status of parents is also muddy: Are th ey insiders, or outsiders?


13The answer is that we can draw the sy stem/environment boundaries however we choose. While there is no ri ght answer to the question What is environment to the system? it is important to recognize that each answer has particular practical implications. A striking example, to return to the family context previously discussed, is the instance when a blood relati ve is rejected by a family as one of its members, or vice versa. Once we construct or re-const ruct a boundary, the question How do we relate to what is not in the system? must be addressed. Ot her questions to consider include: Who is included in the recursive network that is engaged in communicatively constituting the system ? Should parts of the designated environment be part of the recursive network of communication that is the designated system? Are ther e times, for example, when the community leadership should be designated as part of a state-wi de university system, possibly to engage in defining the missions of the various universities and their outcomes and services? Once lines have been drawn to distin guish the system from its environment, the system produces very selective system /environment interconnections. Luhmann (1990) uses the concepts of meanin g and resonance to designate the system/environment interplay. Meaning is the representation of world complexity— that is, the system’s representations of the environment at any given moment. Resonance implies that the resilience of a system is connected to the system’s relation to the environment, which in turn is dependent on the system’s representations of its environm ent. Contrary to the adage seeing is believing an autopoietic system sees, and typically responds to, what it already believes/knows. An organization can not be resilient if it does not have the tools to describe the environment, a necessary condition to act in a responsive way. Excerpts from the Report of the Presidential Co mmission on the Space Shuttle Challenger Accident (Presidential Commission, 1986) can be used to illustrate the


14concepts of meaning, resonance, the re lationship between th e two, and system stability. In Chapter V of the Report, “The Contributing Cause of the Accident,” the Commission reported how they “sought to identify the human failures behind the technical failure” (Tompkins, 1993, p. 127): The decision to launch the Challenger [s ic] was flawed. Those who made that decision were unaware of the recent history of problems concerning the Orings and the joint and were unaware of the initial written recommendation of the contractor advising against the launch at temperatures below 53 degrees Fahrenheit. . .They did not ha ve a clear understanding of Rockwell’s concern that it was not safe to launch because of ice on the pad. If the decision makers had known all the facts, it is highly unlikely that they would have decided to launch on Januar y 28, 1986. (Presi dential Commission, 1986, p. 82) More specifically, the Commission conclu ded that the decision to launch the Challenger was based on “incomplete and some times misleading information, a conflict between engineering data an d management judgments, and a NASA management structure that permitted internal flight safety problems to bypass key shuttle managers” (Tompkins, 1993, p. 128). Summarized from a systems perspective, the complexity of the si tuation was inadequately represented throughout the system and triggered a faulty response. When the system’s representations or descriptions of its en vironment are inadequate, system resonance is low and system vulnerability is high Thus, on January 28, 1986, seven crew members on board the Challenger lost their lives. For a system to survive, it must be able “every moment [to] select a new state which is different from the previous one” (Luhmann, 1990). A stable system, contrary to popular belief, is not a syst em that does not change. In short, its structures must be capable of continual adaptation, learning, and innovation, which


15in turn is accomplished by way of “meani ng” (Luhmann, 1990), that is, the capacity to grasp world complexity at any given moment. Meaning is accomplished through difference technique : The system acquires a grasp on the states and the events that appear to it as information by producing its own distinctions and descriptions (Luhman n, 1990, pp. 17-18), that is, its own knowledge and communication. In this way, the system remains dependent on autopoiesis; the continuous generation of knowledge/co mmunication determines the system’s capacity to react to wh atever is environment for it. Requisite Variety Difference technique is synonymous with differentiation within the system. By system differentiation, the system attempts to reduce the environmental complexity by placing a variety of possibilities within the environment and determining what works. This introduces the concept of requisite variety (Conant & Ashby, 1970). The law of requisite variety states: “The variety within a system must be at least as great as the environmental variety against which it is attempting to regulate. Put more succinctly, only va riety can regulate variety” (Buckley, 1967, p. 495). Weick (1979) applies the concept of requisite variety to organizations: Organizational processe s that are applied to equivocal inputs must themselves be equivocal. If a simple process is applied to complicated data, then only a small portion of that data will be registered, attended to, and made unequivocal. Most of the input will remain untouched and will remain a puzzle to people concerning what is up and why they are unable to manage it. . The inability of people in organizations to tolerate equivocal processing may well be one of the most important reasons why they have trouble. It is their unwillingness to meet equivocality in an equivocal manner that produces failure, nonadaptation. (p. 189)


16Traditionally, our organizations have been structured to respond to stable environments and, as a result, may not have the variety of knowledge and experience to survive in dy namic, equivocal environmen ts. Systems with greater complexity (i.e., greater capa city to represent environmental complexity, knowledge) have more and different kinds of relation ships with their environments. Recent practices suggest that our go verning bodies are beginning to recognize the need to diversify the input that informs decision making in today’s dynamic, equivocal environments. These practices include the recent re-emergence of town hall meetings at the federal, state, and local levels of government, as well as recruitment efforts to attract employees from diverse cu ltural backgrounds to public service. These system/environment interconnectio ns have the potential to produce the variety of knowledge, experience, and information in the system that can better serve the public. An extreme example of an organization’s effort to enhance its informationdriven decision-making proc esses is the practice of penetration exercised at the George C. Marshall Space Flight Center (MSF C) in Huntsville, Alabama in the 1960s. Penetration was the practice of assigning MSFC engineers to contractor plants in order to monitor the work of the contractor s. Through penetration, the government hoped to become a “smart buyer” (Thompso n, 2005, p. 127) when the hardware was ready for delivery to NASA. Tompkins (1993) reports, fo r example, an incident in which the Marshall Center engineers knew mo re about the problems with the second stage of Saturn V than the contractor who built it: One such contractor delivered a rocket stage to Huntsville. . The Marshall personnel quizzed the contractor on the po ssibility of cracks in the stage. The contractor’s people finally admitted ther e might be some cracks in it. How many cracks? “Twenty-one,” was the answer.


17 “No, there are 26 cracks,” asserted an MSFC official. The rocket stage was submitted to an examination and X-rayed. The stage contained twenty-six cracks—along with a few workers’ tools and lunch boxes and other debris that shou ldn’t have been there. (p. 69) NASA in this case had penetrated the manufacturing organization, observed the production processes, monito red workbenches, and quizze d contractor employees. Through selective system/environment inte rconnections and requisite variety, NASA (an autopoietic social syst em) had increased its ability to realize (meaning or organizational con sciousness) and respond to una nticipated events in the environment (environmental resona nce or system resilience). Assessing and Influencing System Resilience The purpose of this research project is understanding, and deterring, largescale system failure. To this end, au topoietic systems are presented as an alternative for conceptualizing system accoun tability and system resilience. To guide research and practice, the theory on auto poietic systems must provide direction on the following: What phenomena do we observe, study, and assess? How is effectiveness defined? What are relevant interventions, for the purpose of enhancing effectiveness? What are measures of success? Table 2 is my summary of the framework provided by the theory and language of autopoiesis and its application to research on social systems.


18Table 2 Application of systems theory to social systems Research The language of systems theory Application of autopoiesis, a form of systems theory, to social systems What phenomena do we observe and study? Autopoiesis: A network that creates and recreates its own unity, structures, and elements. A recursive network of communication, where the communication produces, and is produced by, the system. How is effectiveness defined? Requisite Variety: The system’s ability to respond to anomalies in the environment. Viability: The ability to represent and respond to complexity; maintenance of society in time. Where do we target interventions, for the purpose of enhancing effectiveness? Meaning: The representation of world complexity that is realizable by the system at any point in time. Organizational Consciousness: The production of communication/knowledge throughout the system. What are measures of success? Resonance: System complexity reduces environmental complexity. Resilience: Ability to respond to greater environmental complexity.


19Summarized, a social system is an autopoietic system of communication whose effectiveness and viability are related to its ability to represent the complexity of the environment. Efforts to enhance th e system’s effectiveness target its communication/knowledge base—that is, its ability to represent the environment’s complexity, with the goal of enhancing the system’s complexity. With increased ability to respond to dynamic environments, system vulnerability is decreased. In this way, the theory and language of autopoietic systems provide a new language for understanding and influencing system performance. Structures of Accountability At first glance, it would seem that a ccountability for the performance of social systems resides solely with the system’s ability to respond to the complexity of the environment. To stop here, however, is to align systems thinking with previous abstract accounts of institutional failure. To the contrary, viewed from a systems perspective, accountability does not reside wholly with the system as an abstract entity. Systems thinking requires one to see a syst em as a functional whole and to understand the interdependence of its parts. To assign accountability for system performance is to be attentive not only to the whole system, but also to the participation, contributions, and interrelating in the system. This kind of accountability is captured in the “method of operation” (Tompkins, 2005, p. 87) at the Marshall Sp ace Flight Center (MSFC) during the 1960s, the Apollo Era. Known as automatic responsibility it was first and foremost a system of professional and individual accountability: [Automatic responsibility] fixed responsibility in any individual who had the technical competence to recognize and fix a problem; if the individual was unable to fix it, he or she was to communi cate it up the line so that technical strength could be brought to bear on the problem. Every individual was supposed to make NASA a smart buyer and to keep the technical lines of


20communication open. (Tom pkins, 2005, p. 127) In the absence of this sort of accountability, there is an inevitable breakdown of the communication that fosters system resilience. Nearly thirty years after the successful Apollo launch, Tompkins (2005) concluded that it was not coincidental that the descriptions of many of the MSFC communication practices immediately preceding the 1986 Challenger accident did not fit the philosophy of automatic responsibility practiced during the successful Apollo era (pp. 116-117). Autopoiesis provides a theoretical framework for understanding the ways in which system participants, individually and collectively, are accountable for their participation, contributions, and communi cation. The remainder of this section presents this non-traditional image of accountability. Individual and collective performance. The language we consciously or unconsciously use to create social reality is the result of a series of selections. Selection involves the conscious or unconsci ous imposition of me aning on equivocal displays in an attempt to reduce thei r equivocality (Weick, 1979, p. 131). The meaning that is ultimately selected is influenced by numerous factors. Retention, a repository for interpretations, affects subsequent interpretations of equivocality. People try to fit equivocal displays into what they have known all along. When the curre nt situation doesn’t fit with past experience, the situation is often ignored or misread (Weick, 1979, p. 177). Decision make rs also influence selection: Decision makers in orga nizations intervene between the environment and its effects inside the organization, which means that selection criteria become lodged more in the decision makers th an in the environm ent. What the decision makers attend to and enact, th e cues they use, wh y they use those cues, their pattern of inattention and their processes for scanning and monitoring all become more influential as sources of selection criteria. Reality


21as perceived by the members becomes mo re the source of selection within the organization than does reality as perceived by some omniscient, less involved observer. (Weick, 1979, p. 125) This, however, is only half the story. Meaning, action, and accountability do not reside wholly with tradition, the institution, or the system. Members of organizations can avoid bein g “victimized by . thei r parochial volition” by systematically interjecting “chronic do ubt” (Weick, 1979, p. 177). Experience retained as memory can serve as a constrai nt on interpretation and action; whether memory persists and dominates can be a choice, an interpreti ve choice or an enactment choice (p. 217). Interpretive choice. Interpretive choice appears to organizational actors as a question: “’Knowing what I know now, sh ould I change the way I label and connect the flow of experience’” (Wei ck, 1979, p. 217). The choi ce is about the extent to which previous interpretations should inform current explanations of events. The concept interpretation choice (p. 217) creates opportunities to locate instances where the organization’s interpretive structures curb its ability to recognize anomalies. For example, because the Challenger had flown with O-ring erosion and nothing happened, flying with minute amou nts of erosion was not interpreted as a high risk venture (President ial Commission, 1986, p. 148). Enactment choice. Enactment choice prompts a re lated question: “’Knowing what I know now, should I act differen tly’” (Weick, 1979, p. 217). This choice concerns whether a previous action that responded to, or created, change should guide current actions. The concept enactment choice like interpretive choice, provides an opportunity to understand and prevent social system failures. First, enactment choice assists us wi th locating actions and outcomes that were plausible, but not initiated; were initiated but not coordinated; or were coordinated but not coordinated inadequately. For example, fo r one reason or another, the seriousness


22of the problems concerning the Challenger’s O-rings was not adequately reported “up the line” (Tompkins, 1993, p. 130). Individual and collective accountability. The concepts interpretive choice and enactment choice focus our at tention on the constraints and opportunities available at a given moment in time. We can conti nue to reenact what we already know and do, or challenge our knowledge and practi ces and construct new interpretations and responses. However, repeated enactments or persistent discrediting of the past undermines a system’s adaptability. Weic k (1979) suggests a compromise. For an organization to survive, it must develop an ambivalent attitude towards experience, one that simultaneously respects and challenges its history. The chief outcome of organizations, as Weick (1979) points out, is the production of “s table interpretations of equivocal displays” (p. 229). These explanat ions of what an organization is trying to accomplish, and how to go about it, serve as guides for future action and interpretation. Collectively, these explanat ions become an enac ted organization, the familiar world of mission, vision, values, policies, procedures, and practices. Together, they provide some stability as people go about their business. As organizations are reenacted moment by mo ment, individuals in the organization also face the critical question of what to do with what they know, as changes in the environment may challenge the usefulness of the experience or memory of the organization. Weick (1979) describes member s of adaptive organi zations as, “people who oppose, argue, contradict, disbelieve doubt, act hypocritically, improvise, counter, distrust, differ, ch allenge, vacillate, questi on, puncture, disprove, and expose” (p. 229). All of these actions embody ambivalence, some uncertainty toward what is generally accepted as the way to do business. The tension between the organization’s need for both stabilit y (an ongoing reenactment of the system) and flexibility (adaptation to changes in the environment) reveals the ways in which the organization (the familiar world everyone knows) and individuals (choices of


23interpretation and enactment) are accountable for system performance. A New Image of System Resilience and Accountability Accountability and systems thinking are not often joined in public thinking. For example, in his research at the Marshall Space Flight Center (MSFC), Tompkins (1993, 2005) made the following observat ion about automatic responsibility, the system of professional and individual accountability at MSFC that held individuals responsible for resolving problems or communicating them “up the line” (2005, p. 127): [Automatic responsibility] was a radical departure from typical bureaucratic practice. It was such a radical innova tion that since 1967 I have found few executives who felt they co uld administer such a prin ciple. They feared it would foster anarchy. How could you run an assembly line with such a principle? they asked. (1993, p. 67) Fearful of radical departures from typica l bureaucratic practices, leaders often make it difficult, if not impossible, for th e organization to respond to disruptive events. The typical organization’s emphas is on policy, procedures, protocols, and chains of command constrains individuals’ abilities to respond to the unexpected. The tendency is to try to fit a square peg into a round hole by normalizing the unexpected event and implementing preconce ived protocols. When routines are activated, people assume that the world they face is similar to the world that existed at the time the routine was first drafted (Weick & Sutcliffe, 2001, p. 43). This practice ignores unfamiliar information that suggests that trouble is incubating and escalating (so-called red flags ), and generates responses that are consistent with previously crafted plans and operating proc edures for correct performance. Work rules, though usually well intentioned, constrain what workers see and do (Tompkins, 1993, p. 67). Wheatley (1994) captures the essence of a new image of system resilience


24and accountability, one that is informed by autopoiesis and accounts for individualorganization accountability: If we allow autonomy at the local level, letting individuals or units be directed in their decisions by guideposts for organizational self -reference, we can achieve coherence and continuity. [Self organizing autopoietic systems succeed] when the system supports the independent activity of its members by giving them, quite literally, a strong fr ame of reference. When it does this, the global system achieves even grea ter levels of autonomy and integrity [i.e., complexity]. (p. 95) Luhmann (1990) concurs. Systems are domain s of instability and chaos. What is stable are functions and struct ures (e.g., values, strategic direction, practices). They constitute a worldview that is shared, yet co ntinually contested. As such, they serve as provisory but relatively stable points of departure for further operations. During his stint as organizational communication consultant at the Marshall Space Flight Center (MSFC), Tompkins (1993, 2005) identified institutionalized values that contributed to the success of man’s landing on the moon. One was the practice of “Monday Notes” (Tompkins, 1993, pp. 62-66; 2005, pp. 82-85). During the growth stage of MSFC, the director as ked managers, lab directors, and project managers who were removed from him by at least one layer of management to send him, and the other contributors, a weekly one-page note summarizing the week’s progress and problems. The advantages re ported by the contributors included: keeping the director informed of problems and progress; foster ing formal, regular, horizontal communications throughout MS FC; keeping channels of communication open during times of decreased or limited face-to-face communi cations; assuring problems did not fall through the cracks by creating more channels than are logically or ideally necessary; and surfacing co nflicts. “In short,” Tompkins (1993) summarized, “von Braun’s simple reques t for [Monday Notes] had generated a


25rigorous and regularly recurring discipline of communication within the organization” (p. 65). With values-based structures and proce sses such as Monday Notes in place, systems can achieve coherence while si multaneously leaving space for the individuality, diversity, and learning that is fundamental to continuity. Wheatley (1994) states: “What is critical is the relationship created between the person and the setting. That relationship will always be different, will always evoke different potentialities. It all depends on the players and the moment” (p. 34). System resilience, according to advocates of self organizing autopoieti c structures, depends on organizational func tions and structures that not only provide direction, but also allow for local autonomy, the readiness of players, the moment, and a pervasive system of accoun tability. A critical leadership role is designing the underlying structures—that is, the interrelationships that shape the conditio ns where types of behaviors and events become likely (Senge, 1990, p. 341). Thes e underlying interrela tionships (e.g., the purpose, vision, and core values of the system) are impo rtant because these structures foster certain be haviors; by changing the underlying structures, we can promote different patterns of behavior. In this capacity, no one has a more sweeping influence on the system than its leaders (p 341). The work of the leader/designer includes seeing the system as a whole in which the parts are not only internally connected but also linked to the environme nt, clarifying how the whole system can work better, and designing the underlying heedful interrelating of people, processes, and structures that support system resilience (pp. 341-345). Recognizing the role of leaders in designing the underlying structures of organizations, recent high pr ofile lawsuits (e.g., Enron) have aimed to hold individual CEOs culpable for their roles in system fa ilure. Similarly the project manager for Discovery preemptively claimed resp onsibility if any system failures prevented the


26shuttle from returning to Earth. Although these recent instances bring attention to issues of leadership responsibility and culpability, they do not meaningfully assist us with understanding and preven ting system failure. In th e worst case scenario, they obfuscate a more productive understanding of the nexus of individual and system accountability that fosters sy stem resilience. Autopoiesis provides a fresh alte rnative for understanding system performance. To varying degrees, the fami liar world constructe d by organizations provides both opportunities and constraint s for individual cont ributions. For the opportunities and constraints it creates, the organization is accountable. As contributors to the system, members of th e system have opportunities to express themselves, by either reinforcing or cont esting the world enacted by the system. Concomitantly, the system and its individual members are accountable for what they contribute to the on-going production and reproduction of the sy stem, and the reality that is created. This new way of talking about system performance and accountability is supported by theory and research on system failure/resilience in risky (Perrow, 1999) and high reliability (Weick & Sutcliffe, 2001) organizations. Organizational Theory Weick and Roberts (1993) developed the concept of collective mind to describe system performance in situations requiring error-free operational reliability. Collective mind is conceptualized as a patt ern of heedful interrelating in a social system: Actors in the system construct their actions (contributions), understanding that the system consists of connected actions by themselves and others (representation), and interrelate their actions within the system (subordination). (Weick & Roberts, 1993, p. 357) In practice, the concept of collective mind resembles automatic responsibility – the


27system of professional and individual accountability practiced at the Marshall Space Flight Center (MSFC) in the 1960s. Au tomatic responsibili ty, the “method of operation” (Tompkins, 2005, p. 87), prov ided direction on how to participate, contribute, and interrelate in the system. It assumed not only technical expertise, but also system knowledge. Employees we re expected to address problems by resolving them or communicating them to others. The goal at MSFC was successful flight missions; the practice of automati c responsibility ensu red all parts of the organization were interfacing and communica ting with the other parts of the system working toward this end. The concept of collective mind provides a theoretical framework for researching system performance. First, it specifies empirical materials for researching and understanding system resi lience. Weick and Roberts (1993) argue that patterns of interrelating (contributing, representing, and subordinating) within a system are as close to a physical substr ate for holistic consciousness or knowledge as we are likely to find (p. 365). Groups connect their actions with more or less care, and focusing on the way this is done reveals collective me ntal processes that differ in degree of development (p. 360). To understand consci ousness within the system is to be attentive to the participat ion, contributions, and nature and degree of interrelating in the system; to study interrelating is to grasp the intelligence or consciousness of the system, a necessary co ndition to act in a responsive way. Automatic responsibility, Mo nday Notes, and penetration were identified as organizational practi ces at MFSC in the 1960s that cu mulatively contributed to the successful Apollo mission. All are examples of patterns of interrelating that are accessible for researching and analyzing sy stem consciousness, responsiveness, and resilience. Second, the concept of collective mind establishes a performance model; it specifies a goal or vision to be realized. Weick and Roberts (1993) conceptualize


28organizational performance as interrelations of actions of a system that are mediated by the consciousness wi thin the system: There are group actions that are possibl e only when each participant has a representation that includ es the actions of others and their relations. The respective actions converge relevantly, assist and supplement each other only when the joint situation is represented in each and when the representations are structurally similar. Only when th ese conditions are given can individuals subordinate themselves to the requ irements of joint action. These representatives and the actions that they initiate . bring group facts into existence and produce the phenomenal solidity of group process. (Asch, 1952, pp. 251-252) A performance model establishes the goal to be realized. Ideally, the actions of each member of the system are informed by kn owledge of the environment, actions of others, and the interdependencie s in the system. In this way, the joint situation is represented in each response by individuals and the system. Third, the concept of collective mind can account for variations in system performance. Where the concept of consciou sness suggests a kind of capacity in an ongoing stream of interactions, performance is the set of actions that physically exists within a particular system. Examples include the previously mentioned practices of automatic resp onsibility, Monday Notes, and penetration at MSFC. Conversely, breakdowns in performance likel y indicate a correspon ding deficiency in consciousness. When consci ousness is deficient: Individuals represent others in the sy stem in less detail, contributions are shaped less by anticipated responses, and the boundaries of the envisaged system are drawn more narrowly, with the result that subordination becomes meaningless. Attention is focused on the local situation rather than the joint situation. . Key people and activities are overlooked. . There is less


29comprehension of the implications of unfolding events, slower correction of errors, and more opportunities for small errors to combine and amplify. . There is a greater chance that small laps es can enlarge into disasters. (Weick & Roberts, 1993, p. 371) Fourth, the concept of collective mi nd identifies targets for improving performance—that is, forms of social processes, connectio n, and interrelating (Weick & Roberts, 1993, p. 360): People act heedfully when they act more or less carefully, critically, consistently, purposefully, attentively, studiously, vigilantly, conscientiously, pertinaciously. . When heed declines, pe rformance is said to be heedless, careless, unmindful, thoughtless, unconcerned, indifferent. Heedless performance suggests a failure of intelligence. . It is a failure to see, to take note of, to be attentive to. (Weick & Roberts, 1993, pp. 361-362) Activities can be interrelated more or le ss adequately depending on the care with which contributing, representing, and subord inating are done (p. 363). It is these varying forms of interrelating that embo dy the degree of care or consciousness within the system; it is thes e varying forms of consciousness that are actualized in patterns of behavior that rang e from “stupid to intelligent” (p. 361). The more that heedfulness is reflected in a pattern of interrelation s, the more developed the collective mind and the greater the capability to meet emergent situational demands. The concept of collective mind provides a theoretical foundation for analyzing and improving system performance. It sp ecifies empirical mate rials for analyzing organizational performance, performance goals to be real ized, indicators of actual performance, and targets for performance en hancements. Table 3 is my summary of the framework provided by the application of autopoiesis, a form of social systems theory, to research on social systems. This table points out how abstract systems


30theory concepts (e.g., recursive networks of communica tion) are related to empirical materials of our social world (e.g., interrelating) that can be observed, assessed, and changed.


31Table 3 Social system research Research Application of autopoiesis, a form of systems theory, to social systems Applied social system research What phenomena do we observe and study? A recursive network of communication, where the communication produces, and is produced by, the system. The participation, contributions, and interrelating of the members of the system. How is effectiveness defined? Viability: The system’s ability to represent, communicate and respond to the complexity of the environment. Joint action: The organization’s situation is represented in the actions of its members; members subordinate themselves to system requirements. Where do we target interventions, for the purpose of enhancing effectiveness? Organizational consciousness: The production of communication/knowledge that reduces the systems’ vulnerability. Forms of interrelating that develop the collective mind: The care with which contributing, representing, and subordinating are done. What are measures of success? Resilience: Ability to respond to complexity. Capability to meet situational demands.


32Organizational Rese arch and Practice Research on organizations that requir e error-free operations further informs the use of system theory as a framew ork for assessing and improving system resilience (Perrow, 1999; St eier & Eisenberg, 1997; Tompkins, 1993, 2005; Weick & Sutcliffe, 2001). For exampl e, high reliability organi zations provide important lessons about managing in complex, unpredictable environments because of what they do on the “input side” (Weick & Su tcliffe, 2001, p. 9)—that is, what they pay attention to, how they process information, and how they struggle to maintain alertness. They are distinguished by a “continuous updating and deepening of increasingly plausible interpretations of what the context is, what problems define it, and what remedies it contains” (Weick & Su tcliffe, 2001, p. 3). To this end, high reliability organizations complicate rather than simplify their proce sses of attention. They: 1. Maintain complicated mental models of how events unfold. 2. Treat any lapse of performance as a symptom that something is wrong. 3. Encourage the reporting of errors. 4. Position themselves to see as much as possible. 5. Encourage boundary spanners. 6. Develop situational awareness. 7. Hold a deep knowledge of the technolo gy, the system, one’s coworkers, one’s self, and the raw materials. 8. Cultivate diversity. 9. Defer to expertise. 10. Push decision making down and around (Weick and Sutcliffe, 2001) High reliability organizations deal with the unexpected by engaging in activities that develop the system’s consciousness—that is, it s ability to grasp the complexity of its environment.


33The design of an electronic system at NASA that would foster “sharing what we learn or claim to know” (Steier & Eisenberg, 1997, p. 52) exemplifies an organizational intervention to develop the care or consciousness within a system. “In the end,” Steier and Eise nberg (1997) argued, “It is th e case of one organization redefining its idea of learning from one of information records storage and retrieval to one focused on gaining insight in relationship through dialogue” (p. 52). The use of systems concepts such as co llective mind and heedful interrelating theory for explaining, and po tentially preventing, variance in system performance is illustrated by Tompkins ’ s (1993) case study of NASA Marshall Space Flight Center (MSFC). Tompkins reports that description s of certain communica tive events leading up to the launch and the subsequent fatal accident of the Challenger in 1986 did not fit with the practices he had observed at the Marshall Center in the 1960s when Apollo was successfully launched: One example was the highly publicized teleconference held on Monday, January 27, 1986, the night be fore the launch. . One Marshall manager was quoted . as saying he was “appalled” by Thiokol’s recommendation not to launch . .; apparently [the manager] had not been made aware of the critical nature of temperature considerations. (Tompkins, 1993, pp. 9-10) Tompkins (1993) concluded that this was in consistent with the heedful practices of open communication that he had observ ed at MSFC in the 1960s (p. 10). Weick and Roberts (1993) presume th e presence of consciousness and interrelating within all systems. “What may vary across [systems or within a system at various points in time] is the felt need to develop these processes to more advanced levels” (p. 358).


34CHAPTER TWO THE CASE OF 9/11 This research project analyzes retros pective accounts of events leading to September 11, 2001, for the purpose of acqu iring insight into narratives of system failure/resilience and accountability. Specific ally, this project focuses on accounts of pre-9/11 intelligence failures and accountability provided by participants during the 9/11 Public Hearings and documented in the 9/11 Public Hearing Transcripts. Although the 9/11 Commission intended a review of institutional failures, the failure of intelligence functions dominated the focus of the Commission ers and witnesses, not only in their review of the activities of the intelligence community, but also in their reviews of othe r functional areas in the gove rnment, for example, aviation security, border control, immigration. In this light, my resear ch questions are as follows: Research Question 1: What does the language in the Public Hearing Transcripts reveal about the sensemaking processes and structures that shaped these accounts, specifically with regard to pre-9/11 intelligence failures and accountability? Research Question 2: Viewed through the lens of autopoiesis, how does the analysis of the accounts of pre-9/11 intelligence failures and accountability in the 9/11 Public Hearing Transcripts inform a new vocabulary of system accountability and system resilience, one wi th practical implications for future organizational practice? The conceptual and methodological fr amework that informs this research project is Weick’s view of sensemaking (1979, 1995; Weick, Sutcliffe, & Obstfeld, 2005). This chapter presents this framework. In addition, I expl ain the protocols for selecting and analyzing excerpts from the 9/11 Public Hearing Transcripts for this research project. Using the sensemaking perspective as a lens, my goals are to


35(a) reveal the communication beliefs that constitute the constructions of system failure and accountability in the 9/11 Public Hearing Transcripts and (b) identify alternative narratives for expressing syst em resilience and accountability. Sensemaking Sensemaking has been co nceptualized as the proc ess of structuring the unknown (Waterman, 1990, p. 41); active ag ents constructing sensible, sensable (Huber and Daft, 1987, p. 154); a proce ss through which interpretation of discrepancies are retrospectively develope d (Louis, 1980, p. 241); “the reciprocal interaction of information seeking, meaning ascription, and action” (Thomas, Clark, & Gioia (1993, p. 240); and “the activity of ‘making’ that which was sensed” (Weick, 1995, p. 30). The sensemaking perspective is unique in that it goes beyond interpretation (attending to cues, interpreting, externalizing, and linking these cues) to include an explanation for how cues got there in the fi rst place; how particul ar cues are singled out from an ongoing flow of experience; and how the in terpretations and meanings of these cues are altered and made more explic it and sensible as a result of activities (Weick, 1995, p. 8). Weick (1995) argues that autopoieti c systems should be concerned with sensemaking, as their effectiveness and viability depend on this ability to make sense and respond to unanticipated events in the environment: It is the very openness associated wi th [autopoietic systems] that makes distinctions between out there and in he re inventions rather than discoveries, that results in people creating their own constraints, and that triggers the strange sequence in which outputs become the occasi on to define retrospectively what could have been plausible inputs and throughputs. (p. 70) These are the very problems that are th e focus of sensemaking—how we construct


36the explanations we construct, why, and with what effects (Weick, 1995, p. 4). Weick’s (1995) dissection of sensemakin g into seven properties puts “some boundaries around the phenomenon of sensemaking” (p. 18), providing a rough guideline for inquiry into sensemaking processes. Definitions and examples of each of the sensemaking properties follow. Co llectively, these seven properties (pp. 18 – 65) provide a conceptual fr amework for understanding how the events leading to September 11 were constructed by witnesses during the 9/11 Public Hearings. Sensemaking is Grounded in Identity Construction People establish and maintain identity by taking cues from others while simultaneously making an effort to influe nce the definition of self that is being constructed for them. An example of th is sensemaking property comes from my simultaneous appointments as Director of two depart ments during the past two years. In an email confirming an impending Leadership Team Meeting, the Associate Vice President requested an update from the team on its succession planning project. Knowing that heads would turn to me as the Director of Orga nization Development and Training for this update I prepared a report; however, to avoid the abnegation of my identity as the Director of Envi ronmental Health and Safety, I recruited another member of the team for the presen tation, and participated as a member of the facilities team in the subsequent discussion. This complex mixture of proactivity and reaction is commonplace in sensemaking. The sensemaking perspective on the construction of identity captures the e ssence of autopoiesis, a form of systems thinking that takes into account the interd ependence of the parts that constitute the system. Viewed from this sensemaking/systems perspectiv e, identity is constructed from cues from others and one’s own efforts to influence the definition of self. This nexus of self and other constitutes the self that defines and then responds to the world (Weick, 1995, p. 20).


37In a similar way, a system establishes and maintains its identity. That is, the core, distinctive, and enduring character of the system (Albert & Whetten, 1985; Weick, Sutcliffe, & Obstfeld, 2005, p. 416) is a nexus of cues from the environment and the system’s own efforts to influence the definition of its identity: Who we think we are (identity) as organi zational actors shapes what we enact and how we interpret, which affects wh at outsiders think we are (image) and how they treat us . . [Organizational] sensemaking, filtered through issues of identity, is shaped by the recipe “h ow can [we] know who we are becoming until [we] see what they say and do with our actions?.” (p. 416). According to Weick et al. (2005), the estab lishment and maintenance of identity is a core preoccupation in sensemaking, i.e., th e turning of circumstances into a situation that, comprehended explicitly in words, se rves as a springboard for action (p. 409; see also Weick, 1995, pp. 18-24). For an understanding of the it that is defined, it is important to look at that which has inform ed the self conscious sensemakers who are engaged in making sense of the situation. For the purp ose of this project, the definitions of self used during sensemak ing processes provide insight into the why and how of the explanations constructed in the 9/11 Public Hearing Transcripts, as well as other plausible iden tities that may be useful prospectively. Weick (1995) places identity first on the list of sensemaking properties because the establishment and maintenance of identity, a core preoccupation in sensemaking (p. 20), influences how other aspects, or properties, of sensemaking are understood. This positions identity as the primary sensemaking property in constructing the meanings that inform and constrain action. Sensemaking is Retrospective Meaning comes with the kind of retrospect ive attention that is directed to the experience; it “is not ‘attached’ to the expe rience” (Weick, 1995, p. 26). As a result, many possible meanings are plausible, and the problem for the sensemaker is


38equivocality. When people are overwhel med by equivocality, they use values, priorities, and preferences to help them gi ve meaning to that elapsed experience. For example, when in doubt, I prefer to tr ust others. Identifica tion of the values, priorities, and preferences used to crea te meaning from equivocality can provide insight into how, why, and to what effect part icular explanations are constructed. Sensemaking Enacts Sensible Environments Enactment is first and foremost about action. People produce the environment they face, and th is environment constrains th eir actions. For example, organizational charts, position descriptions, and job titles delineate the ways in which employees are to participate in and contri bute to organizational activities. The custodial worker cleans carpets, the offi ce assistant manages calendars, and the senior fiscal assistant reconc iles ledgers. This is where sensemaking most clearly becomes a process that creates objects for sensing. Creating is not the only thing that can be done with action. Lines of action may be stopped, abandoned, postponed, pl anned but not implemented, initiated and then transformed, redirected and inhibited, as well as expressed. Examples include the processes, or lack of processes, that shaped the widespread disappointment with Continuous Quality Improvemen t (CQI) at the end of the 20th century. Frequently organizations embarked on the use of CQI tools to generate e fficiencies, customer satisfaction, and cost sa vings without building the underlying philosophy and structures that are required to realize the tangible results of CQI. According to Weick (1995), “The act that never gets do ne, gets done too late, gets dropped too soon . is seldom a senseless act” (p. 37). Sensemaking is about action but it is also “as much a matter of thinking that is acted out conversationally in the world as it is a matter of knowledge and technique applied to the worl d” (Weick et al., 2005, p. 412). The centrality of the “saying” in sensemaking and or ganizational action includes “the talk that leads to a


39continual, iteratively developed, shared understanding of the [circumstances] and the persuasive talk that lead s to enlistment in action” (Wei ck et al., p. 412). Neither talk nor action, from the perspective of sensemaking, is inherently more important. Until talk brackets action and gives it meaning, it remains an “’indistinguishable part of the swarm of flux’” (Weick et al., p. 412). The centrality of talk and action in organizational sensemaking has been coined as “thinking that is acted out conversationally” or “acting thinki ngly” (Weick et al., p. 412). The presence (or absence) of talk and ac tion that takes on “the form of an externally specified objective reality where transacting parties play out preordained roles and . ‘routines’” (Ring & Van de Ve n, 1989, p. 185; see also Weick, 1995, p. 36) is the substance of sens emaking, and contributes to an understanding of how current and alternative interpretations and enactments are constructed. Sensemaking is Social People actively shape each other’s me aning and conduct, whether they are physically present or absent. According to Weick (1995), “What I say and single out and conclude are determined by who socia lized me and how I was socialized, as well as by the audience I anticipate will audit the conclusions I reach” (p. 62). My Type A personality, like so many of my colleagues, is deeply rooted in family, the Anglo Saxon Protestant culture of our childhood, and the work ethic of our employers. People who study sensemaking pay attentio n to the communication that socializes, defines connections, establishes entities to which people can orient, signifies what is important, and binds people’s time to projects. Examples include “symbols, promises, lies, interest, attention, thre ats, agreements, expectations, memories, rumors” (Weick, 1985, p. 128), upward and downward communicati on, face-to-face and mediated communication, and centra lized and decentrali zed communication (Tompkins, 1993, pp. 17-26). The forms, qu alities, and timing of relationships, interrelating, and coordination provide insight into various ou tcomes of sensemaking.


40Sensemaking is Ongoing People are always in the middle of continuous flows of activity. As we say, a million things are going on (Weick et al., 200 5, p. 411). All of th ese activities furnish “an undifferentiated flux of fleeting sense impressions” (Chia, 2000, p. 517), from which cues may or may not be extracted for closer attention. When a flow of taken-for-granted activity is interrupted, an emotional response is typically induced, which then paves the way for emotion to influence sensemaking. For example, only weeks after Hurricane Katrina ravaged New Orleans and the coast of Mississippi, Hu rricane Rita was moving up the Gulf of Mexico toward the Texas coastline. Before the devastation of Hurricane Katrina, it was often the case that people disregarde d official mandatory evacua tions of coastal areas and flood zones, boarded up thei r homes, stocked supplies, and hunkered down to wait out the storm. Hurricane Rita, however, triggered a different response. On the heels of Hurricane Katrina, Rita struck fe ar and terror throughout the state. The unprecedented mass exodus of over one million evacuees, mostly from nonevacuation zones in Houston, is a powerf ul example of emotion-laden sensemaking induced by the threat of si gnificant disruptions to ongo ing, taken-for-granted day-today activities. It is precisely when ongoing flows of activity are interrupted, or those flows are significantly threat ened, that interpretations of events and actions are infused with and influenced by feelings (Weick, 1995, p. 45). Understanding the influence of emotion on the co nstruction of meaning provides insight into the various outcomes of the sensemaking process. Sensemaking is Focused on and by Extracted Cues Extracted cues are simple familiar structures th at are seeds from which people develop a larger sense of what may be occurring. For ex ample, the blatantly negative and accusatory to ne in an annual report recently published by an institution’s Auditor General’s office generated some internal sensemaking regarding


41the subsequent resignation of its director. People make sense of what they extract; if events are not extracted fr om the ongoing flow of acti vity, they are not available for sensemaking. What is and is not noticed by sensemakers provides insight into meanings that are, and are not, constructed. Sensemaking is Driven by Plau sibility rather than Accuracy People believe what can account for a se nsory experience, as well as what is interesting, attractive, emotionally appealin g, and goal relevant. Stories, myths, metaphors, epics, paradigms, etc. pull to gether disparate elements and provide a socially acceptable, credible, coherent, an d reasonable accountin g of events. For example, the termination of senior-level employees is of ten accompanied by official explanations, for example, “small children at home” and “recent medical conditions.” Often these explanations are accepted as plausible explanations. If we are to understand the products of sensemaking, we must understand them as plausible explanations judged only by their usefulness, not their accuracy: Sensemaking is not about truth and getting it right. Instead, it is about continued redrafting of an emerging story so that it becomes more comprehensive, incorporates more of th e observed data, and is more resilient in the face of criticism. . Stories tend to be seen as plausible when they tap into an ongoing sense of current c limate, are consistent with other data, facilitate ongoing projects, reduce equivocality, provide an aura of accuracy . and offer a potentially exci ting future. (Weick et al., 2005, p. 415) The plausible explanations of pre-9/11 intelligence failures and accountability that dominate the 9/11 Public Hearing Transcripts are the focus of this research project. These accounts will be analyzed for the pu rpose of acquiring insight into current narratives of system failure and accountability and to identify alternative narratives for expressing system resilience.


42Discussion and Summary Weick’s dissection of sensemaking in to seven properties provides a conceptual framework for understanding se nsemaking processes. This framework includes an explanation of how cues appe ar in the environment (e.g, the enactment of sensible environments), how particular cues are singled out from an ongoing flow of experience (e.g., the iden tities that define and respond to extracted cues from the ongoing flux of the environme nt), and how the interpretations and meanings of cues from the environment become more exp licit and sensible (e.g., the plausible retrospective explanations that emerge and become more comprehensive and resilient). Focusing on how we cons truct our explanations, the sensemaking perspective provides a conceptual framewor k for analyzing the constructions of pre9/11 intelligence failures and accountability that are documented in the 9/11 Public Hearing Transcripts. Revealing how these interpretations were constructed, alternative constructions can be considered. Research Protocols This dissertation conceptualizes the events leading to 9/11 as a system of organizational communi cation—that is, as a system of interrelated words, meanings, and actions that lead up to the terro rist attacks on September 11, 2001. Organizational communication is “the stud y of sending and receiving messages that create and maintain a system of consciousl y coordinated activiti es . of two or more persons” (Tompkins, 1984, pp. 662-663; 1993, p. 24; see also Bantz, 1993). Messages can be spoken, wri tten, or conveyed by gestures and include not only content but also metacomm unication—that is, the messa ges associated with the manner or mode in which the content is conveyed and interp reted (Watzlawick, Beavin, & Jackson, 1967). They include simple missi ves (e.g., messages on the bulletin board) as well as more important ex amples such as the exercise of authority (Tompkins, 1993, pp. 24-25). Additionally they include the crucial tensions of


43upward and downward communication, face -to-face and mediat ed communication, centralized and decentralized communi cation, and formal and informal communication (Tompkins, 1993, pp. 17-26). Precedents for understanding comp lex system performance from an organizational communica tion perspective includ e the analysis of the Challenger and Columbia accidents by Tompkins (1993, 2005), Weick’s (1990) discussion of a near collision of two airlines at Tenerife, and Steier and Eisenberg’s design of “conversational domains” at NASA (1997). This research project is a qualitative textual analysis of system failures documented in the 9/11 Public Hearing Tran scripts. Words, sentences, paragraphs, discrete accounts of events, and conversational exchanges will be excerpted from the Transcripts. The selection of excerpts will be guided by the research questions. To this end, excerpts related to both pre-9/11 intelligence failures and accountability will be extracted from the 9/11 Public Hearing Transcripts. Excerpts from the 9/11 Public Hearing Transcripts As researcher in this project, I aim to create insight by researching various views of pre-9/11 intelligence failures and accountability documented in the 9/11 Public Hearing Transcripts. To this end, both dominant and marginalized views of pre-9/11 intelligence failures and accountability are targeted for extraction and further analysis. Examples of the kinds of excerpts to be extracted include: 1. Definitions, descriptors and uses of intelligence, intelligence failure, and accountability. 2. Struggles and negotiations over the definition of a situation. 3. Arguments, claims, and evidence. 4. Analogies to other situations, r oot metaphors, and rigid frames. 5. Coherence and fidelity. 6. Cross talk


447. Traditional and non-traditional vocabulary choices; labels; root metaphors; and rigid frames. 8. Recommendations indicating al ternative plausible accounts. Analysis To guide research and practice, We ick et al. (2005) have compressed the properties and distinguishi ng features of sensemakin g into a conceptual and methodological framework (Figure 1), called the “enactment model” or the ESR sequence (p. 414). Figure 1 ESR Sequence Source: adapted from Weick, Sutc liffe, and Obstfeld, 2005, p. 414. The ESR sequence: proposes that sensemaking can be trea ted as reciprocal exchanges between actors (Enactment) and their enviro nments (Ecological Change) that are made meaningful (Selection) and pres erved (Retention). However, these exchanges will continue only if the preserved content is both believed (positive causal linkage) and doubted (negative causal linkage) in future enacting and selecting. (Weick, Sutcliffe, & Obstfeld, 2005, p. 414) Table 4 is my depiction of th e compression of specific features of sensemaking into the ESR sequence (Weick, Sutcliffe, & Obstfeld, 2005, 414).


45Table 4 The relationship between ESR and sensemaking Sensemaking sequence Features of sensemaking Ecological change Actors (concerned with identity), extracting anomalies from the flux of circumstances, begin to change the flux of circumstances into orderliness. Enactment Noticing, bracketing and enacting order into flux, identities/actors are shaped by externalities. Selection Retrospective attention, mental models, and articulation reduce possible meanings of the bracketed material and tentative and provisional plausible stories are constructed. Retention Plausible stories are made more substantial when linked to past experience, connected to significant identities, and used as a source of guidance for further action and interpretation. With consideration of the ways in which previous knowledge (retention) informs selections and enactments during the sensemaking pr ocess, ESR serves as a


46conceptual framework for analyzing sy stem resilience (reciprocal systemenvironment exchanges) and accountability. When previous knowledge and experience are used ambivalently, we not on ly benefit from lessons learned, but also can update our understanding of, and resp onses to, our environm ent, including how “outsiders” treat us (Wei ck et al., 2005, pp. 414, 416). Through growth, development, change, and continuous lear ning, systems can create, elaborate, or change structures. The ESR sequence capt ures this dynamic, the tension between the ongoing reenactment of the current system and innovation and adaptation. Whether the system’s sensemaking reso urces are enacted or contested, the response is always a contribution to the ongoing production and reproduction of the system. For these responses, both the sy stem and individuals are accountable. How we construct the explanations we co nstruct, why, and with what effects is the focus of sensemaking that is captur ed in the ESR process. Applied to this research project, the ESR sequence is us ed as a guide for revealing the processes and structures that shap ed the accounts of pre-9/ 11 intelligence failures and accountability documented in the 9/11 Public Hearing Transcripts. Additionally, it will inform a subsequent discussion on practical implications for building organizational resilience. Overview of Similar Studies The research design for analyzing sele cted excerpts from the 9/11 Public Hearings has been informed by Dombrowski ’s (1991) qualitative textual analyses of testimonies and reports relating to the space shuttle Challenger accident. Dombrowski’s (1991) study was designed to elucidate whether th e decisions leading to the shuttle accident were primarily determined by pr ocedures or by an individual who then used procedures to effect or ju stify a decision. Descr iptions of decisions and decision-making processes were extracted from the testimonies presented to the Presidential Commission on the Space Shuttle Challenger Accident and the Ninety-


47ninth Congressional Committee on Science and Technology, the two principal government agencies investigating this di saster. The portrayal of decisions and decision-making processes were also extracted from the reports of the Presidential Commission and Congressional Committee. Excerpts were organized into three categories of judgments: PE – people pr imarily; PR – procedures primarily; and AM – ambiguous, indeterminate, or balanced. Fi ndings from each of the categories were summarized and supported by excerpts. Conclusions were drawn, e.g.: The evidence and testimony before both the commission and the committee show that personal judgment in all its contingency and fallibility, rather than procedural shortcomings, accounts fo r the loss of the Challenger. The conclusions and recommendations of both investigations (especially of the [President’s] commission), however, make little mention of personal judgment or responsibility. (Dombrowski, 1991, p. 213) Implications for future prac tice were presented, e.g.: We need to be mindful in profession al communication and public policy discussion of the nature of humans as language-using creatures guided by their own constructions, for these same constructions, while powerfully operative, are often left hidden an d unacknowledged. We should, for example, resist the inclination to accept uncritically the “bottom-line” conclusions and recommendations of report s, instead taking pains to identify the methodological and conceptual assumptions which condition these conclusions and recommendation s. (Dombrowski, 1991, p. 215) Additional qualitative text ual analyses that have informed the methodology for this research include Lighthall’s ( 2002) analysis of published and archival testimony of participants in the decision to launch the Challenger for the purpose of identifying lessons for training engineers and management and Gross and Walzer’s (1997) analysis of the explan ations of the cause of the Challenger disaster by the


48Presidential Commission. The Case of 9/11 The following provides an overview of the methodology I used for identifying and analyzing excerpts from the 9/11 Public Hearing Transcripts. In the first phase of this research, I printed a copy of the HTML transcripts located on the website of the National Commission on Te rrorist Attacks upon the Un ited States. The HTML transcripts (approximately 2000 pages) were chosen over PDF files (approximately 3000 pages) due to the larger number of pages in the PDF file. Each of the transcripts for the nineteen days of hearings was placed in a manila folder; each folder was given a name that reflected the hearing number and day of the hearing (e.g., 2.2 indicated the second hearing, the second day of the hearing). I initially read through each of the nineteen transcripts to accomplish a high level understanding of the co ntent. During this review of the transcripts, all testimony relating to information, intelligence, and intelligence failures was highlighted in yellow. Testimony relating to accountability and performance management (e.g., goals, standards) was high lighted in blue. During this reading of the transcripts, each of the transcripts wa s reviewed several times. Before moving on to the next day of testimony, I revi ewed and re-reviewed the transcripts to ensure that I had a grasp of the content of the testimony, I had color-coded the testimony appropriately, and there was no qu estion in my mind that I could clearly articulate the link between the content of the transcript and the color coding assigned to it. At this stage of the research, I deci ded to err on the side of too many examples. To this end, any doubt I had ab out the applicability of testimony to this research project resulted in a decision to highlight it and include it for future review and decision making. It was also during this phase of the research project that I came to realize that intelligence, intelligence failure, and accountability dominated


49the focus of the 9/11 Public Hearings. At the end of this initial review of the project, the majority of each of the transcripts appe ared as yellow, blue, or green highlights. As a result, throughout the project, I vacillated between having the good luck of having so much data and being overwh elmed by the amount of material. The second reading of the transcripts was divided into two phases. First, I read all the yellow-highlighted testimony for each of the nineteen days of the twelve hearings. During this reading, I labeled each of the yellow-highlighted passages in the transcripts by selecting language from the passage as a label to reflect its content (e.g., sharing of information), wrote this label on a yellow post-it note, and attached the post-it note to the highlighted page. The hearing number, day of the hearing, and page number of the highligh ted testimony were also written on the post-it note in the event it became separate d from the transcript. The sticky side of the post-it note was attached vertically to the document. The label assigned to the transcript was written on the opposite end of the post-it note. This way, I could flip through a file and fairly easily locate the various labels. Second, I read all the bluehighlighted testimony for each of the hearings. Using blue highlighted post-it notes, I repeated the above process. During the third review of the transcripts, I focused on the labels attached to testimonies on information, intelligence, intelligence failures, and accountability. For each of the identified labels (e.g., access to information) I used Microsoft Excel to compile related excerpts, noting the hearin g number, day 1 or day 2 of the hearings, and the verbatim excerpt on a spreadsheet After compiling labels and verbatim excerpts from only a few days of transcripts, I realized the approach was too cumbersome for the amount of testimony relate d to this project. My second effort focused on compiling, for each of the labels, only the hearing numbers, days of the hearing, and page numbers of the testimonie s. This approach was productive, in the sense that it provided the location for each of the related excerpts but not a grasp of


50the content. At this point, I realized I must work directly from the transcripts. With pads of flipchart paper, nineteen manila folders, and a copier, I moved the project from an office desk to the floor, walls, and dining table in the great room of my house. Because of the large number of excerpts attached to various labels, I took each previously identified label (e.g., flow of information) and traced its path through the nineteen days of hearings. La bel by label, I compiled topics discussed during the 9/11 Public Hearing Transcripts and related excerpts. One label at a time, I took notes and taped copies of related excerpts to flip chart pages. Similar statements were grouped together on the flip chart pages (e.g., copi es of excerpts or notes from the transcripts blaming the Presid ent for 9/11 were si tuated on the page, separate from the testimonies blaming the intelligence community for 9/11). As a result, categories began to appear. Afte r reviewing the ninet een transcripts and placing notes and copies of testimonies on the flipchart pages, the flipchart pages began to tell a story about the particular label under review at that time. For example, it was during this process that I came to understand the frequency and consistency of the definition of intelligence that was dominating the 9/11 Public Hearing Transcripts. It was also during this process that I was able to identify less frequently-used definitions of constructs. Th is process also allowed me to review my previous decisions and once again asse ss the fit between th e content of the transcript and the label attached to it. Occasionally, labels were reassigned to other categories. At the end of this step, I was able to identify the constructs and meanings that dominated the transcripts. Repetitive use of words and meanings was the criterion for identifying dominant constructs. Once dominant constructs were identi fied, the selection of excerpts for the dissertation became a challenge due to th e overwhelming number of examples for any of the dominant constructs. I select ed excerpts for this dissertation that, consistent with other possi ble excerpts, more clearly reflected the use of the


51construct by the witnesses and commissioners. Once dominant constructs were identifi ed, they were reviewed to determine if system knowledge and experience (e.g., or ganizational constraints, organizational premises, plans, expectations, acceptable ju stifications, and traditions) (Weick et al., 2005, p. 409) had served as a resource in the dominant constructions of pre-9/11 intelligence failures and accountability in the 9/11 Public Hearing Transcripts. A system sensemaking resource qualified as such if it had informed the construction of multiple constructs. Additionally, I reviewed excerpts that challenged the dominant explanations of events leading to 9/11, fo r prospects of alternative perspectives on system failure and accountability and a new vocabulary for constructing accounts of system resilience and accoun tability. Using the PDF search function, language from the HTML file was used to locate and cite the page numbers of excerpts in the PDF file also on the website of the National Commission on Terrorist Attacks upon the United States. The findings from this analysis of the 9/11 Public Hearing Transcripts are presented in the following format. Chapter Three provides a detailed discussion of the construction of intelligence failure documented in the 9/11 Public Hearing Transcripts. Verbatim excerpts from the 9/11 documents will be used to support claims. Chapter Four provides a detailed analysis of the dominant accounts of accountability in the 9/11 Public Hearing Transcripts, along with supporting verbatim excerpts from the transcripts. Chapte r Five discusses the do minant sensemaking resources that informed the accounts of pre-9/11 intelligence failures and accountability in the 9/11 Public Hearing Tran scripts. Practical implications for future system resilience and acco untability, based on a discussion of the theory of autopoiesis and accounts of intelligence failures and accountability in the 9/11 Public Hearing Transcripts, are presented.


52CHAPTER THREE THE CONSTRUCTION OF INTELLIGENCE IN THE 9/11 PUBLIC HEARING TRANSCRIPTS On March 31, 2003, the National Commi ssion on Terrorist Attacks upon the United States officially open ed the 9/11 Public Hearings. A number of witnesses who had “a particular interest” in the even ts of September 11, 2001 (Kean, 1.1.10)1 appeared before the Commission during the various sessions of the 9/11 Public Hearings. These included survivors of the attack, fa milies of victims, first responders, public officials, scholars and practitioners in the areas of national security and policy, and co ngressional members who sp onsored legislation that brought the Commission into existence (see Appendix A fo r the dates of the twelve Public Hearings, Appendix B for the me mbers of the National Commission on Terrorist Attacks upon the United States, an d Appendix C for a list of witnesses cited in this paper). These Commissioners and witnesses may have had various particular interests for participating in the 9/11 Public Hearings (Kean, 1.1.10), but there was at least one point of consensus from which all subsequent testimonies ensued, not only during the first session of the public hearings but throughout the twelve hearings: On September 11, the United States government failed at its primary duty to provide common defense (Push, 1.1.163) and protect its people (Gorelick, 1 Twelve Public Hearings were conducted by the National Commi ssion on Terrorist Attacks upon the United States. Transcripts from nineteen days of testimony are retained as pdf Federal records managed by the National Archives and Records Administration (see Appendix A for the url to the pdf file for each of the hearings). These web pages were frozen on Septembe r 20, 2004, at 12:00 a.m. Seven of the twelve hearings were two-day sessions. References to testimony presented during two-day hearings are designat ed in this dissertation by For example, 1.1.4 cites Hearing 1, day 1, page 4. Twelve of the Public Hearings were single-day hearings and are cited as hearin For example, 4.19 cites Hearing 4, page 19. In each instance, the name of the Commissioner and witness providing the testimony is provided.


531.1.36). For the friends and families of th e victims, the survivors of the attack, the Commissioners, and members of Congress wh o had sponsored the legislation that brought the Commission into ex istence, “The September 11th attacks represented a massive failure in the most fundamental duty of our government: the security of the American people from foreig n attack” (McCain, 2.1.7). Commissioner Cleland, in his opening statement, ad mitted, “Almost without question, we could and should have been be tter prepared, we know that, to protect our homeland against the terrorist assault” (1.1.57). Clearly, the pre-9/11 attacks by al Qaeda, such as the 1993 bombing of the World Trade Center and the 2000 attack on the USS Cole were not sufficient to: Make intelligence bureaucracy shed th eir turf-consciousne ss and their Cold War mentalities or our border-control agencies to overcome inertia and budget shortfalls or the airlines and airports to tighten security, even if it meant some added inconvenience to the traveling public or the executive or legislative branches to prioritize ho meland security above other spending programs. (Cleland, 1.1.61) According to Commissioner Kean (2.2.7), no body had anticipated this kind of event and the government was unprepared for it. In portrayals such as the one pr esented by Cleland, the dominant construction of intelligence and its role in the events leading to 9/11 emerged. 9/11 was not a failure of a single person or a department of government but rather a systemic breakdown and problems associated with in telligence, as well as other functions, contributed to this failure. A ccording to the final report of the Joint Congressional Intelligence Committee Inquiry, quoted by Commissioner Cleland during the Public Hearings: “Prior to September the 11th, the intelligence community was neither well organized nor equipped and did not adeq uately adapt to meet the challenge


54posed by global terrorist s focused on targets within the domestic United States. These problems greatly exacerba ted the nation’s vulnerability to an increasingly dangerous and immediate international terrorist threat inside the United States.” (Cleland, 1.1.57-58) Cleland assumed the Commissi on would not only affirm those intelligence deficiencies but would find corresponding lapses in bord er control, aviation security, and a host of other fields (Cleland, 1.1.56). Others agreed. Whatever failures occurred in the intelligence agencies may have been matched in seriousness in other agencies (Pelosi, 2.1.5; see also Graham 2.1.27; Kean, 2.2.1; Shelby, 2.1.26). Investigating these myriad failures was th e charge to the National Commission on Terrorist Attacks upon the Unit ed States (McCain, 2.1.8). The 9/11 Public Hearing Transcripts offi cially document the discourse of the 9/11 Public Hearings. This chapter, drawing on the documented testimony in the Public Hearing Transcripts, presents the re curring themes that emerged during the hearings. The following are discussed: 1. Definitions that are relevant to an understanding of the accounts of pre-9/11 intelligence failures provided by wi tnesses and Commissioners in the 9/11 Public Hearings. 2. The themes that dominated the constructions of pre-9/11 intelligence failures by the 9/11 Public Hearing participants. 3. The conditions within which these stories became dominant constructions in the 9/11 Public Hearing Transcripts.


55Definitions “Perhaps intelligence is a term . that exists in an ‘Alice in Wonderland’ world where words can mean whatever one capric iously says that they mean” (Ransom, 1975, p. 154). In an effort to understand better how the 9/11 terrorist attacks on the United States could have happened in a nation th at is as militarily, technologically, and economically strong as the United States (Kean, 1.1.6), the 9/11 Commission investigated, along with myriad other failures, the “failure of intelligence” (Jenkins, 1.1.305) so it could be “improved” (Ben -Veniste, 1.1.93) or “fixed” (Harman, 2.1.38). The following review of definitions from scholars and practitioners in the areas of national security and policy are pr ovided as a backdrop to the accounts of pre-9/11 intelligence failures that are presented in the next section of this chapter. Intelligence Intelligence, in the context of national security, is information (Berkowitz, 1996, p. 35). What distinguishes it from ot her information is its purpose or use. In the context of national securi ty, intelligence is what decision makers need to know before choosing a course of action (Ranso m, 1970; Rosenbaum, 1971; p. 114). It is information that serves the interest of policymakers (Bru emmer, 1992, p. 890), assists those who implement policy (Ber kowitz & Goodman, 1989, p. 109), and improves decision makers’ ability to und erstand issues (Goodman, 1984-85, pp. 161162). It includes gathering, interpreting and selectively comm unicating national security information to decision makers. Ultimately, all intelligence activities should inform the decisions or further the policies of the President (Flanagan, 1985, p. 67). In common usage, the word intelligence has come to have many meanings (e.g., espionage, covert political interventions, paramilitary action, and counterterrorism). In fact, Ransom pointed out in 1975, the word intelligence h as no precise meaning in common usage: “Preside nts (e.g., Ford) and even secretaries of


56defense (e.g., Schlesinger) and directors of central intelligence (e.g., Colby) have exhibited the common, carele ss habit of referring to ‘intelligence’ as meaning both information and secret political action ” (Ransom, 1975, p. 154). According to Ransom (1975), this is not just a matter of semantics. The sloppiness of the definitional problems “has affected th e methods of organizing and managing American intelligence agencies” (Ransom, 1975, p. 154). For example, the usage of the word intelligence to refer to both information and secret political activity can result in information being excessively classi fied, an issue that was of great interest to the Commissioners during the 9/11 Public Hearings. Ransom added, “Conceptual confusion about ‘intelligence’ makes soluti ons difficult” (Ransom, 1975, p. 154). Intelligence Production Intelligence production is a “linear and single-tracked” process (Berkowitz, 1996, p. 47) by which information is acquired and converted into a product for “’consumers’ of this information” (Flanaga n, 1985, p. 62; see also Berkowitz, 1996, p. 47). The process involves several functions: collecting data, analyzing it, and disseminating the final product (Ben-V eniste, 6.50; Berkowitz, 1996, p. 47; Berkowitz & Goodman, 1989; Cline, 198990, p. 695; Deutch, 4.30; Goodman, 1984-85, p. 161; Kean, 6.3; Sc hulhofer, 6.5; Ziglar, 7. 1.134). In theory, the intelligence product begins and ends wi th policymakers who are the consumers of the information. In practice, it is up to the managers to gauge and anticipate policymakers’ needs (S chlesinger, 4.3). Among the most important sources of intelligence are foreign and domestic media, reports on political and social developments from various government agencies around the world, a number of intelligence organizations, and military intelligence services (Flanagan, 1985, p. 64). These collection entities produce a tremendous volume of information that mu st be sorted, processed, and converted into “raw intelligence reports’ that can be disseminated to intelligence analysts and,


57in some instances, policymakers (Flanagan, 1985, p. 64). Analysts use the collected data to develop products that are then coordinated, edited, and delivered to the consumer (Berkowitz, 1996, p. 47). “Finishe d intelligence” repres ents a very careful review of information from all available so urces by analysts or analytic teams who are familiar with the issue or the geographic regions (Flanagan, 1985, p. 65). In this context, available sources refers to agencies and depart ments authorized to collect and transmit national security information. The Central Intelligence Agency, Defense Intelligence Agency, and the State Department’s Bureau of Intelligence and Research are principal producers of national level fi nished intelligence (Flanagan, 1985, pp. 61, 65). Additionally, the line between raw intelligence (i.e., data) and finished intelligence, (i.e., information that has been reviewed for national security decisions and policymaking) is blurry. In terms of the ESR Sensemaking Sequence (Chapter Two), the difference between data and finished intelligence is the difference between numerous possible meanings and fewer plausi ble explanations. Guided by mental models acquired during work, training, and life experience, th e multiple possible meanings of data are reduced to these fewe r plausible explanatio ns (Weick, Sutcliffe, & Obstfeld, 2005, pp. 411, 414). In this way, people make sense of, and respond, to equivocality. Intelligence Community For ease of reference, the agencies and departments that are involved in foreign intelligence or counterintelligence activities (Flanaga n, 1985, p. 60) are collectively known as the Intelligence Community. Authorized by the National Security Act of 1947 and subsequent orders, the Intelligence Community is responsible for “intelligence activities ‘necessary’ for the conduct of foreign relations and the protection of the National Securi ty of the United States” (Goodman, 198485, p. 161).


58The agencies formally designated as members of the pre-9/11 Intelligence Community included the Central Intelligence Agency (CIA), the Defense Intelligence Agency (DIA), the National Security Agency (NSA), the military and special collection offices in the Pentagon, the State Department ’s Bureau of Intelligence and Research (INR), the Treasury Department’s Office of Intelligence Support, the Federal Bureau of Investigation (FBI), and a unit of the Department of Energy (Deutch & Smith, 2002, p. 66; Goodman, 1984-85, p. 161; Nati onal Commission on Terrorist Attacks upon the United States, 2004, pp. 407-408) Figure 2 is a snapshot of the Intelligence Community on the eve of September 11, 2001. Figure 2 Pre-9/11 U.S. intelligence community Source: Deutch & Smith, 2002, p. 66. An overview of the agenci es reviewed by the 9/11 Public Commission follows. It is important to note that many of the structures and functions in place during the


59Cold War continued to be in place on the eve of the 9/11 terrorist attacks. Changes had been made, only slowly, “at an in cremental pace” (National Commission, 2004, p. 407). On the eve of September 11, the Director of Central Intelligence (DCI) had authority for the coordination of the variou s agencies and entities that constituted the Intelligence Community. However, a number of departments with intelligence missions had some independence from th e DCI, “in order to serve their unique departmental needs” (Flanaga n, 1985, p. 58; see also Deutch, 4.18). The only agency directly controlled by the DC I was the CIA (Bruemmer, 1992, p. 868; Goodman, 1984-85, p. 161; National Commission, 2004, pp. 409-410). The CIA had the broadest mandate of a ll Intelligence Community agencies. It supported the DCI in the coordi nation of community functions. It was also involved in the production of a broad array of intelligence reports, counterintelligence activities abroad, clandestine collections of foreign intelligence, and the development of data collection systems supported by technology. Although the CIA had the broadest mandate, the lion’s share of the national intelligence resources and personnel were in the Department of De fense (Flanagan, 1985, p. 61; Kerr, 4.72; National Commission 2004, pp. 409-410). The Defense Intelligence Agency (DIA) provided intelligence and counterintelligence support to the Office of the Secretary of Defense, the Joint Chiefs of Staff, and the Unified and Specific Command operations; additionally, the DIA coordinated intelligence activities of the m ilitary services (the foreign intelligence and counterintelligence activities of the Army, Navy, Air Force, Marine Corps, and various offices that collected specialized intelligence that supported national, department, and military service needs) (Deutch & Smith, 2002, p. 66; Flanagan, 1985, p. 61; National Commission, 2004, p. 408). The National Security Agency (NSA), under the authority of the Secretary of


60Defense, was responsible for the operation of the nation’s technical intelligence systems (e.g., satellites) and the CIA carried out human intelligence (i.e., clandestine collection of foreign intelligen ce). The Secretary of Defense and the DCI shared authority for setting priorities for the collection of foreign intelligence (Deutch & Smith, 2002, p. 65; Flanagan, 1985, p. 61; National Commi ssion, 2004, pp. 407, 409). In the case of foreign threats within the United States, th e Federal Bureau of Investigation (FBI) had the responsibility for setting and implementing intelligence collection priorities. The Department of State’s Bureau of Intelligence and Research (INR) produced “finished” intelligence tailored to the Department’s needs and participated in the development of national intelligence reports, coordinated the State Department’s relationships with other foreign intelligence agencies, and distributed reports from the United States diplomat ic and consular posts abroad to the Community (Flanagan, 1985, p. 61; National Commission, 2004, p. 408). The Department of the Treasury maintained a small intelligence unit which, in cooperation with the State Department, collected openly available financial, monetary, and economic data for use in th e department and in national intelligence products (Flanagan, 1985, p. 62; National Commissi on, 2004, p. 408). Oversight of intelligence activities was performed by several executive and legislative branch entities including the President, National Security Council (NSC), and Senate and House permanent intelligen ce oversight committees. The Office of Management Budget (OMB) influenced in telligence policy by reviewing budget proposals and justifications, monitoring the Community’s budge ting process, and providing the DCI with a budget (Flanagan, 1985, p. 72; Nati onal Commission, 2004, p. 410). The intelligence activities of the Treasury and State Departments, the military services, and the counterterrorism role of the FBI predated the CIA. In 1947, when


61nation states such as the Soviet Union were the envisioned enemy of the United States, the National Security Act established the CIA as the preeminent U. S. government agency to receive and interpret foreign intelligence from all sources. The act also codified the posi tion of the Director of Cent ral Intelligence (Bruemmer, 1992, p. 867). The effect was the organizati on of U.S. intelligence functions based on the distinction between domestic law en forcement and foreign national security concerns. The FBI was responsible for the former and the remainder of the Intelligence Community (e.g., the CIA, NSA, and DIA) was responsi ble for the latter. This fragmented approach to intelligence, conceived in 1947, cont inued to affect the way the U.S. organized and managed inte lligence for the remainder of the 20th century and the beginning of the 21st century: Law enforcement’s focus is to collect evidence after a crime is committed in order to support prosecution in a court of law. The FBI is reluctant to share with other government agencies the information obtained from its informants for fear of compromising future court action. On the other hand, the CIA collects and analyzes information in order to forewarn the government before an act occurs. The CIA is reluctant to give the FBI information obtained from CIA agents for fear that its sources and methods for gaining that information will be revealed in court. (Deutch & Smith, 2002, p. 64) The FBI had primary responsibility for counterintelligence and counterterrorism within the United States. The CIA and DIA had primary re sponsibility for threats to the U. S. that “would come from overseas or be overseas” (Roemer, 1.1.302). This fragmented approach to intelligence, in place on the eve of the 2001 terrorist attacks, was a concern that dominated te stimonies of witnesses during the 9/11 Public Hearings. Definitions of intelligence and Intelligence Community used by scholars and practitioners in the fields of national security and polic y have been provided as a


62backdrop to the accounts of pre-9/11 intelligence failures that are presented in the next section of this chapter. Keeping in mind that our definitions “create what we disingenuously pretend they merely describe” (Carey, 1989, p. 32), it is not unreasonable to assume that the following may be encountered in the reading of the 9/11 Public Hearings: 1. One of the uses of the word intelligence in the Public Hearing transcripts will denote information that has been ga thered, analyzed, and disseminated for national security purposes (e.g., decisions and policymaking). 2. Intelligence will be treated as a product or a thing that can be gathered, analyzed, and disseminated. 3. The words intelligence and information may be used imprecisely and interchangeably as a matter of course. 4. The descriptions of the methods and pr actices for organizi ng and managing pre-9/11 intelligence will resonate with the above definitions of intelligence and Intelligence Community as these definitions do minated scholarship and practice during this period of time. For example, intelligence is defined in the literature and practice as information provided by the Intelligence Community for the purpose of national security decision or policymaking. In the 9/11 Public Hearing Transcripts, we can expect the line between what is and is not intelligence and who does/does not inform policymaking to be similarly demarcated. In this vein, we may encounter testimony regarding valuable information that was generally available (i.e., information relevant to 9/11), but was not treated as intelligence because it was not coming from an intelligence official. 5. The Commission’s charge to identify intelligence failures and provide recommendations will lead to an in-depth exploration of the intelligence process (i.e., the collection of information, its analysis, and the dissemination


63of intelligence), as it was generally believed that “failure in any of these steps” would lead to intelligence fa ilure (Cline, 1989-90, p. 695; see also Berkowitz & Goodman, 1989). 6. The intelligence failures identified by the 9/11 Commission (e.g., the fragmented nature of U.S. intelligence) and the solutions to these problems (e.g., reorganization of the Intelligence Community) will be logical extensions of the dominant definitions of intelligence and Intelligence Community that were in place during the investigation of 9/11. Intelligence: Constructions in the 9/11 Public Hearing Transcripts The ESR sensemaking model (Weick et al., 2005) provides a conceptual and methodological framework fo r understanding system failure/resilience. Using this model, system resilience can be viewed in terms of the reciprocal exchanges between ecological changes (environment ) and actors (system). These exchanges become meaningful as labels are attach ed to “the undifferentiated flux of experience,” and used as “common currenc y for communication exchanges” (Chia, 2000, p. 517). The labels that are attached to the flux of ex perience narratively reduce the stream of experien ce to plausible stories. Thes e stories are important, as they have the potential for guiding future interpretations of events and actions (Weick et al. p. 415). Keeping in mind the prospective as well as retrospective nature of sensemaking, this chapter identifies the five dominant themes in the accounts of pre-9/11 intelligence failures provided by witnesses and Commi ssioners during the 9/11 Public Hearings: 1. Flow of information 2. Actionable intelligence 3. Coordination and sharing of information 4. Access to information


645. Dominance of law enforcement and pr osecutorial approach to terrorism These stories appeared during the official opening of the 9/11 Public Hearings on March 31, 2003, and reverberated througho ut the twelve public hearings. With hindsight, we see these topi cs recurring and coalescing in to dominant descriptors for the pre-9/11 failures in intelligence. This section focuses on each of the five themes that dominated the Commissioners and witnesses’ co nstructions of events leading to the failure of 9/11. In the concluding section of this chapter, I review these five constructs to determine the conditions within which these stories became dominant constructions in the 9/11 Public Hearing Transcripts. Similarly, th e next chapter identi fies and discusses the dominant constructions of accountability in the Public Hearing Transcripts. Flow of Information It is not surprising that flow of information was a recurring theme in witness accounts of pre-9/11 intelligence failures, as this concept, flow of information was at the heart of the intelligence production model that dominated scholarship and organizational practices at the turn of the 21st century. In theory and practice, national security intelligence was define d as the product of a process in which information is collected from various sources in the field, disseminated to analysts for further analysis, and then further disseminated for the purpose of decision and policymaking. Critical to this process was the movement or flow of information through this intelligence production process (Axley, 1984; Feldman and March, 1981). In the 9/11 Public Hearing Transcripts, the movement or flow of information was generally depicted as a “one-way street” of information. For example, Commissioner Ben-Veniste (1.1.93) used this metaphor to characterize the flow of information from the FBI to state and local authorities. The flow of information to the Federal Aviation Administ ration (FAA) was similarly ch aracterized. According to


65Schiavo, former Inspector General for the Department of Transportation, it was the responsibility of the FBI to assess a threat and then provide intelligence, which the FAA “fans out” to the airlines (2.2.95); that is, they transmitted advisories, warnings, etc. (Canavan, 2.2.60). The process from December 2000 until October 2001 was outlined by Canavan, FAA Associate Administ rator for Civil Aviati on Security: Throughout 2001, as the intelligence reporting volume increased . my office issued at least 15 information circulars . focusing on domestic and international terrorism threats di rectly against aviation. . For example, one information circular . issued in the summer of 2001 updated airline security personnel of developments that terrorists and criminals had in disgui sing firearms. . We pushed real hard to get ever ything we got from the intelligence community into the field. (Canavan, 2.2.60-61, 68) The FAA, cast as consumers of intelligen ce, relied completely on the Intelligence Community (U.S. agencies and departme nts authorized to conduct foreign intelligence and counterintelligence agen cies) to provide the best quality of intelligence so they could issue information circulars to the aviati on industry, security professionals, corporate secu rity directors, senior mana gement personnel, ground security coordinators, superv isory personnel at overseas locations, local airline managers, and law enforcement when appr opriate and on a n eed-to-know basis (Canavan, 2.2.61). During the summer of 2001, the entire United States Counterterrorism Group, incl uding law enforcement and intelligence agencies, sent out notifications that heightened security measures should be put into place immediately. During this period, the FAA advanced this information in the form of Security Directives and Inform ation Circulars to airlines, ai rports, and all officials for whom it had regulatory oversight (Canavan & Fielding, 2.2 84). These initiatives were intended to “get them to conform to the existing regulations” (Canavan &


66Lehman, 2.2.69). During the 9/11 Public Hearings, witnesses from North America Air Defense (NORAD) also relayed an a ccount similar to the one presented by the FAA. Restricted by law from collecting domestic intelligence (Verga, 7.1.144), the department of Defense relied on the other in the Intelligence Community for domestic intelligence. This intelligence served as the basis for developing training, tactics, and procedures fo r combatant missions. In accordance with pre-9/11 report s from the Intelligence Community, NORAD defense forces were positioned for ex ternal threats to the United States and were therefore looking “outward” (Verga 7.1.159). The FAA was also looking outward, as they too had received intelligence indicating the impetus of the heightened state of alert during the summer of 2001 was “overseas” (Garvey, 2.1.91, 93, 98). In both instances, acco rding to witnesses from FAA and NORAD, neither had any specific intelligence indicating a domestic terrorist threat to commercial aviation prio r to the attacks. Commissioner Lehman found the un-prep aredness of FAA and NORAD for the terrorist attacks on September 11 to be an incredulous sequence of events: Despite [a] long litany of events and intelligence reports of the growing probability that aircraft would be us ed as weapons, nothing ever got to [Mineta, Secretary of Transportation], nothing apparently got to [General McKinley], and I assume, General Arnold, nothing got to you. (2.2.31). This according, to Lehman, was a significant failure. The United States intelligence community existed “to provide product precis ely” to the users who were tasked with defending the country (Lehman, 2.2.31). There are numerous examples in the 9/11 Public Hearing Transcripts where consumers of information testified that the flow of critical information had been impeded such that they had not received th e intelligence that might have assisted


67them with foreseeing or deterring the attacks of 9/11. For example: 1. The FAA had no knowledge of the “Phoenix” memo, an internal FBI memo written in July 2001, sugge sting the Bureau invest igate the use of civil aviation schools by indi viduals who may be aff iliated with terrorist organizations (Rai dt, 7.2.7). 2. The Commissioner of Immigration and Naturalization Services (INS) was not briefed or advised of a heightened period of alert with respect to the possibility of Islamic fundamentalist terrorist activity when he assumed the position of Commissioner in August of 2001, nor was he aware of anyone in the National Security Section of INS who was aware of it (Ben-Veniste & Ziglar, 7.1.146-147). 3. The Bureau of Consular Affairs had not received intelligence on any of the 9/11 hijackers from “the ag encies designed to colle ct it,” although some agencies possessed information on two of the terrorists who would participate in the September 11 attacks (Ryan, 7.1.19). 4. Two of the hijackers were on a wa tch list established by the State Department, but these names were not provided to the FAA; consequently, two hijackers, Khalid al Mihdhar and Nawaf al Hazmi, were not subject to additional scrutiny or barred from flying (Zelikow, 7.2.14-15). 5. Information from the FBI and CIA on al Qaeda’s use of fraudulent passports was not available to Customs, Immigr ation, or Consular officials who examined hijacker passports befo re 9/11 (Ginsburg, 7.1.4-5). 6. Immigration inspectors were not given any information that would have prevented them from admitting any of the hijackers into the country (Fine, 1.2.39). These examples are presented to illustrate how flow of information, a dominant construct in the 9/11 Public Transcripts, was used by the Commissioners


68and witnesses in their accoun ts of the pre-9/11 intelligence failures. Commissioner Ben-Veniste stated it as a “fact” (7.1.146): We had a considerable amount of collected intelligence prior to 9/11, but it seems as though our failures involved the inability to disseminate . the information in a way in which we could interdict those individuals who participated in the 9/11 plot. (Ben-Veniste, 7.1.146) The question was, how could it be that “t he product did not get” to those who could use it? (Lehman, 2.2.32) Discussions on flows of information in the Public Hearing Transcripts centered on the distribution of intelligence; that is, the Intelligence Community had not distributed the information to its consumers. Primarily, consumers of intelligence testified that they were not provided or did not receive the intelligence that would have assisted them with fo reseeing or deterring the at tacks. This emphasis on information transfer (Eisenberg & Good all, 1993, pp. 22-24) reflected several fundamental assumptions that dominate or ganizational practice and scholarship, including the fields of comm unication and national security and policy, during the 20th century and the early years of the 21st century: 1. Intelligence is a concrete substance that travels, or is moved, from one place to another. In the Intelligence Community, as well as in the 9/11 Public Hearing Transcripts, intelligence was referred to as a product that was disseminated by the Intelligence Comm unity to its various consumers. 2. The receipt, or non-receipt, of intelligence causally or quasi causally affected the performance of the consumers of inte lligence. In the 9/11 Public Hearing Transcripts, consumers of information testified they were not prepared for, or in a position to deter, the 9/11 at tacks because they had not received relevant information from the Intelligence Community.


693. Success of intelligence transmission is mitigated by breakdowns, defined as problems during the transmission of information that cause the flow to stop completely; barriers, defined as obstacle s that simply impe de or slow down the message during the transmission proc ess; or gatekeeping, defined as the control of the information flow by the source, with the source of the information determining who receives what information (Krone, Jablin, & Putnam, 1987, p. 23). In the 9/11 Pub lic Hearing Transcripts, it appears that information was purposefully withheld from those who could have deterred or mitigated the 9/11 terrorist attacks. For example, the FBI provided information to consumers when the bure au had something that was credible, specific, and relevant to their operations (Flynn, 7.2.41; Manno, 7.2.42). However, where there was no direct connec tion to a specific operation, it was very difficult to obtain clearances from the FBI to obtain access to information (Manno, 7.2.37). It appears, from th is example and others, that the FBI played a gatekeeping role and controll ed the distribution of intelligence information. Although there are accounts of gatekeep ing in the 9/11 Public Hearings, they are few and far between and do not in th emselves account for the numerous stories of consumers not receiving information. In the end, only on a few occasions, during moments of expressed frustration, did I find the Commission struggling to gain additional insight into the recurring storie s on the difficulties with the flow of pre9/11 intelligence. On those rare occasions, and only momentarily, mainstream thinking appeared to be challenged by the Commissioners. The following two examples exemplify the occasional on-the -margin observations, if not judgments, from Commissioners. In one example, it appears that th e taken-for-granted passive role for consumers of intelligence (e.g., policymakers) was challenged:


70[Should consumers of intelligence] take a much more active role in directing the priorities of what is to be collected, and direct the intelligence community to find a way to collect it rather than taking a more passive role and saying okay, tell me what’s happening. (Lehman, 4.38) In the next two examples, th e standard definition of intelligence as that which is provided by the Intelligence Community was under fire. This exchange occurred between Commissioner Lehman and Mary Ryan, Assistant Secretary of State for Consular Affairs from May 1993-July 2002: Ryan: So I think that knowing certainly what we did in Consular Affairs, prior to September 11th, 2001, we did—we took every step that we could . . Lehman: I would respectfully disagree with you. I don’t think the record shows that at all. In some of the interviewing of some of your officials that were doing the actual consular functions in Saudi Arabia at the time, they said in so many words, gosh, if we only knew . . Well, hello. I mean, did anybody read the newspapers? I mean, there were books. . Did you only have robots . ? Don’t they read the papers? (Lehman & Ryan, 7.1.29-30) Similarly, Commissioner Ben-Veniste found it “a bit fatuous” that the United States was so unprepared on September 11, give n the generally available information on terrorists using airplanes (2.2.27), as well as “reports floating around in the intelligence community” and “about a ha lf a dozen novels and movies” about airplanes being used as weapons (Lehman, 2.2.14). In the 9/11 Public Hearing Transcripts, a dominant construct shaping the accounts of pre-9/11 intelligence failure was flow of information ; information that was not sent, provided, or disseminated was the meaning most often attached to this construct by the Commissioners and witnesses. On the surface, it may appear as a gatekeeping issue; that is, the Intelligence Community was withholding critical


71information. And in some instances, that was the case. A closer look, however, may challenge the conceptual an d definitional models of intelligence and Intelligence Community that had informed both pre-9/11 intelligence work and its analysis by the 9/11 Commission. Actionable Intelligence “We really had no credible or actionable intelligence that told us this was really going to happen. In other words, this is a real threat, we are hearing, this, this, this, this and this” (Canavan, 2.2.85). One of the questions pose d by the Commissioners on the first day of the 9/11 Public Hearings was, “How do we improve . actionable intelligence . the specifics of that information . the right information” (Roemer, 1.1.103)? References to receiving, or not receiving, specific, right, hard, credible information, that is, actionable intelligence, from th e Intelligence Community was a dominant theme in the 9/11 Public Hearings. The following composite of testimonies by various FAA witnesses exemplifies the use of this term in the public hearings. During the spring and summer of 2001, the Intelligence Community issued reports, at a “near frantic level,” suggestin g some type of terrorist attack somewhere in the world (Gorelick, 2.2. 15). The entire counterterrori sm community including law enforcement and intelligence agencies we re placed on the highest alert, and information circulars were dist ributed to all interested parties (e.g., airlines, airports, and all security officials) (C anavan, 2.2.61; Fielding, 2.2. 84). For example on July 18, the FAA issued information circulars stating, “We have no specific information on [a] threat to civil aviation. The FAA urge s all civil aviation security personnel to continue to demonstrate a high degree of alertness” (Canavan, 2.2.83). In the United States, this heightened level of intelligence warning about impending attacks (Gorelick, 2.2.15) was in terpreted as “chatter” (Mineta, 2.2.15). The term chatter appears in the 9/11 Public Hearin g Transcripts as an antonym to


72actionable intelligence. Actionable intellig ence refers to specific information about a terrorist attack (who, what, when, where, how) dissemina ted by a credible source (e.g., an agency in the Intelligence Community). Chatter, on the other hand, refers to increases in communicati on activity without corresp onding increases in its specificity or credibility. Reports issu ed by the Intelligence Community, before September 11, 2001, weren’t specific about what the threat might be and, as a result, did not result in any action by the government. For example, there was no indication that attacks woul d be focused specifically against airlines (Canavan, 2.2.61; Mineta, 2.2.14); theref ore additional prevention or mitigation measures were not initiated. The FAA responded only to specific, ac tionable information that was provided to them by the CIA and FBI. For exampl e, the twenty names on the pre-9/11 FAA no-fly list had been specifically identified to the FAA by the Intelligence Community as the terrorists that the FAA ought to be concerned about (Minet a, 7.2.28). In the months preceding September 11, in response to the specific credible threat information provided to them, the FAA “worke d hard to make changes in the aviation security baseline” (Garvey, 7.2.17). Where there was specific and credible information that people were actually targeting civil aviation, the names of those individuals were put on secu rity memos directing the ai r carriers not to transport these people (Manno, 7.2.27). Additiona lly, the 61,000 names in TIPOFF (the U.S. government’s only pre-9/11 terrorist watchlist) could be searched “if you had a name” to search against th e database (Manno, 7.2.27). Pre-9/11, the kinds of protocols that the FAA had in place anticipa ted specific kinds of crises (e.g., bad weather and Y2K) (Garvey, 2.1.108). Although information regarding the use of airplanes as weapons of mass destruction and the possibility that this could happen in the United States was accumulating, the FAA (working with the intelligence community) in each case


73deemed the source of information as not ac tionable; that is, it was not specific and did not come from credible sources (Garvey, 2.1.106,114). There was “hard intelligence” indicating seriou s threats against aviation in the Pacific (Flynn, 7.2.18), but the FAA had no specific knowledge of a terrorist plot to hijack aircraft and use them as weapons against targets in the Un ited States, or any plot that resembled such an operation prior to 9/11 (Garve y, Gorton, & Manno, 7.2.23). In her testimony, Garvey reported, “The FAA did not have any credible or any specific information which indicated the type of [d omestic] attack we saw on September 11” (Garvey, 7.2.17). There was no actionable intelligence that even hinted that there was a real threat to aviation (Canavan 2.2.85). According to the testimonies provided by a number of FAA consumers of intelligence, the Intelligence Community was providing national security information, but it was not actionable; therefore, it did not result in the kind of preparation that one, with hindsight, might expect. The meaning ascribed to actionable intelligence in the 9/11 Public Hearing Transcripts resonates with the use of finished intelligence in the intelligence production model that dominates current public administration scholarship and practice (Flanagan, 1985, p. 65). That is, finished intelligence is actionable intelligence; it can guide policy and decision making. In contrast, information does not qualify as finished intelligence to be ac ted upon if it does not specify threats or does not originate from formal intelligence production processes. An overview of the kinds of information that was generall y known prior to the September 11, 2001, terrorist attacks, based on the testimony of various witnesses during the 9/11 Public Hearings, is provided in Figure 3.


74Figure 3 Information that was generally known before September 11, 2001


75Figure 3 (continued) Information that was generally known before September 11, 2001


76 According to multiple witnesses in the 9/11 Public Hearings, intelligence had to be actionable (i.e., provide information about a sp ecific attack) if it were to be useful (Manno, 7.2.21). “Solid inte lligence” (Fielding, 2.1.105)—that is, intelligence that had been run to the ground by the intell igence community (Garvey, 2.1.106 )—was was not provided; therefore appropriat e actions were not taken. During the 9/11 Public Hearings, a number of accounts of pre-9/11 intelligence failures focused on information that was not disseminated. That is, information did not flow from the Intell igence Community to national security decision makers. When witness testimonies specifically attributed the failures of 9/11 to a lack of actionable intelligence, the failures were constructed as a specificity of intelligence that was not disseminated by the Intelligence Community. This emphasis on accuracy, specificity, and completeness of information reflected two fundamental assumptions dominating organiza tional scholarship an d practices during this period of time: (a) communication is about perfection, fidelity, and accuracy (Berlo, 1960, p. 40) and (b) “accuracy begets effectiveness” (Wei ck et al., 2005, p. 415). This focus on accuracy, grounded in models of rational decision making, dominated organizati onal thinking and practice during the end of the 20th century and the early years of the 21st century. Using rational decision making models: A given problem is evaluated in relation to stable goals and a course of action chosen from a set of alternatives. In this model, accurate information is important in evaluating the feasibility and utility of alternative actions, and accurate perceptions increase decisi on quality. (Weick et al., p. 415) Rational models of decision making are best suited for organizations that reside in stable, predictable environments. However, they are ill suited for organizations that reside in complex, dynamic, unpredictable environments, where organizations must be capable of continuously no ticing, bracketing, and responding to novel events in a


77continuously changing envi ronment (Weick, Sutcliffe, & Obstfeld, 2005, p. 415). Generally, according to Wheatley (1994), our efforts to keep “the lid on” until we “get to the bottom of” unexpected, differe nt, disconfirming, conflicting information have left our organizations “dying, literally ” (p. 108). That is, in complex, dynamic environments, organizations do not have the luxury of compiling information to the point that a threat is fully understandable and, as a consequence, actionable. Additionally, the discussion on actionable intelligence during the 9/11 Public Hearings suggests how a focus on findin g and connecting dots (e.g,. Ashcroft, 10.1.142, 166; Baird, 1.2.146; Ben-Veniste, 11.1.146; Fetchet, 1.1.218; Fielding, 10.1.19; Jenkins, 1.1.305; Lehman, 12.2. 48; McLaughlin, 10.2.46; Reno, 10.1.62; Roemer, 10.1.110) or pieces of informatio n draws attention away from patterns and relationships of events that unfold over time. Testimonies throughout the hearings suggest that there was “a considerable amount” of information prior to 9/11 (BenVeniste, 7.1.146) reflecting trends or patterns of activi ty leading to 9/11. This generally available information, however, was distinguished from the pieces or “dots” of information (the who, what, when, where, and how of terrorist attacks) compiled by the Intelligence Community for national security purposes. Coordination and Sharing of Information In their first report, the Commission staff (National Co mmission, 2004, xiii) set out to reconstruct a “story” that, in the end, is “more telling about the system than the people” (Zelikow, 7.1.78, 88). In this system, the staff concluded, no one was ensuring seamless handoffs of informatio n or coordinating an overall interagency strategy for operations (Zelikow, 7.1.88). The staff account of pre-9/11 events, presented by Dr. Philip Zelikow (7.1.78-86), the Commission’s Executive Director, provides an opportunity to gain a gr eater understanding of the meaning of lack of coordination and sharing in the 9/11 Public Hearings. These events reconstructed by the staff, began with the analysis of si gnals intelligence (SIGNET) by the National


78Security Agency (NSA). Near the end of 1999, NSA analyzed communications associ ated with three men, Khalid, Nawaf, and Salem. Combining the NSA analysis of these communications with information from other sources, officials in the intelligence community concluded that “Naw af and Khalid [were] part of . ‘an operational cadre’ [and] something nefarious migh t be afoot” (Zelikow, 7.1.78). If NSA had been asked to further in vestigate these indi viduals, NSA would have started by checking its own database and the following events would have been set in motion (7.1.81): Analysts would quickly have identified Nawaf as Nawaf al Hazmi. Someone then could have asked the State Depart ment to check that name too. State would promptly have found its own reco rd on Nawaf al Hazmi. That record would have shown that he . had been issued a visa to visit the United States. They would have learned that the visa had been issued at the same place, Jeddah, and on almost the same day as the one given to Khalid al Mihdhar. (7.1.81) But information was not requested from NSA, and NSA did not think it was its job to initiate research on its own (7.1.78). Information reported from, and to, the intelligence community continued to reinforce “the picture of an emerging op eration” (7.1.79). On January 4, 2000, officials tried to track Nawa f’s scheduled departure from Pakistan for Malaysia, only to discover that he had alread y left the country. Other of ficials had more success in tracking Khalid. They learned that his name was Khalid al Mihdhar and that he held a Saudi passport with a visa to visit the United States. When the CIA headquarters searched its database for the names that were being identified, no hits were produced. They did not check the databases at NSA and did not specifically ask NSA to do so.


79 January 12, 2000, official s reported that the “Arabs had dispersed and the tracking was falling apart” (7.1.82). It was not unt il March 2000, when another matter generated renewed interest in “those missing Arabs” (7.1.83), that Nawaf al Hazmi’s departure on a United Airlines flight to Los Angeles on January 15 was reported. Sharing this information with the FBI, from an intelligence perspective, was vital; however, the Commi ssion staff found “no evidence that this information was sent to the FBI” (7.1.83). Mihdhar, o fficials later learned, had also been on the United flight to the United States of America. By March 2000, Mihdhar and Hazmi had already established a residence in San Diego. At the time, no one in the intelligence community knew this (7.1.83). In August 2001, an FBI agent, working with a CIA agent deta iled to the FBI, put together the pieces of the puzzle and “nominated” Hazmi and Mihdhar for TIPOFF, the U.S. government’s only pre9/11 watchlist dedicated to catching terrorists (7.1.84-85). Zelikow added: It is worth noting that the Federal Aviation Administration’s own no-fly list was totally independent from TIPOFF. . So, to be specific, adding Hazmi and Mihdhar to TIPOFF did not put them on a no-fly list and did not keep them from flying on September 11. (7.1.85) Mihdhar and Hazmi were involved in the hijacking of American Airlines 77, which hit the Pentagon. The Commission staff also reported that these hijackers had previously met up with a hijacker who late r joined the hijacking team assigned to United 175. Zelikow concluded with the observation that the staff’s acco unt of the events leading to 9/11 revealed how difficult it was for the intelligence community (e.g., the CIA, FBI, and NSA) to assemble “puzzle pi eces” gathered by the different agencies, make sense of them, and then coordinate needed action with relevant agencies (e.g., FAA) (7.1.86).


80 It is impossible to overlook issues relating to initiative, ownership, accountability, and oversight—all of which wi ll be discussed in the following chapters. Setting those aside for now, we can focus on the construction of coordination and sharing of information in the staff report. According to the staff report, no on e was ensuring a seamless handoff of information or coordinating an overall interagency strategy. That is: 1. NSA was not asked to further investigate, and the State Department was not asked to check their records, for Nawaf and Khalid once they had been identified as part of a disreputable operational cadre. 2. Although the State Department had its own records on Nawaf al Hazmi, the CIA did not have the names in its database. 3. Information on Hazmi’s flight to Los Angeles was not reported to, or shared with, the FBI. 4. Putting the names of Hazmi and Mihd har on the government’s terrorist watchlist, TIPOFF, did not prevent them from flying because the FAA’s own fly list was totally independent of TIPOFF and the FAA did not have these names on its list. In the staff’s reconstruction of this event, there are numer ous instances where pieces of the puzzle were gathered by diffe rent agencies but not shared; that is, it was not handed off or transmitted to other agencies. Informatio n and requests were not given to the NSA, the State Department, or the FBI. Similarly, information was not in (i.e., it had not been moved into) th e CIA database or the FAA no-fly list. Despite the pieces of information resi ding in various loca tions from late 1999 to September 11, 2001, the Intelligence Comm unity lost track of Nawaf al Hazmi and Khalid al Mihdhar. In January 2000, they entered the United States. On September 11, 2001, they hijacked an airp lane, and with two other hija ckers, crashed it into the Pentagon.


81 This construction of events by the Commission staff was dominated by instances where interagency sharing and coordi nation did not occur; that is, agencies did not move information from one place to the next as one would expect, according to the defined process of intelligence production. In the end, the Commission staff had reconstructed an account of devastation caused by intelligence failures, a failure to move national security information along. In this account, as well as other similar accounts in the 9/11 Public Hearing Transcripts, the constructs, coordination and sharing of information referred to the transmission of information from one plac e to another. This construction of information as a “’thing,’ as an inert entity to disseminate,” is reminiscent of “several decades of information theory that treated information as ‘bits’” that could be moved, tracked, passed back and fort h and managed (Wheatley, 1994, pp. 101102), and, in this sense, coordinated and shared Access to Information Who did, and did not, have access to intelligence was a topic of various testimonies during the 9/11 Public Hearings. The following is presented to reflect the ways in which access to information surfaced as a dominant theme in the construction of pre-9/11 intelligence failures in the 9/11 Public Hearing Transcripts. In the summer of 2001, an FBI agent an d a CIA agent detailed to the FBI put the pieces together and figure d out that two terrorists, Nawaf al Hazmi and Khalid al Mihdhar, were in the country and were “clearly here to kill Americans” (Baker, 6.38). Baker, former General Counsel, National Security Agency, re counted the events: The FBI agent who discovered this . made it his mission to try to find these guys. . When he asked for the authority to get the assistance of [the] law enforcement [side of the FBI] . he was told by FBI headquarters, not on your life. You cannot capitalize, cannot do that because there’s a wall


82between law enforcement and intelligence. And [the response of the FBI agent] was to say, the American people will not understand this, someone is going to die because of this. (Baker, 6.39-40) In addition to not having access to FBI law enforcement information and personnel, the FBI agent did not have access to “a lo t of government information” that would have assisted him with tracking down the two terrorists. After seeking and acquiring authorizations to access the government in formation that could be available to him, he encountered a second access issue. He did not have computer access to the records that he had acquired the legal authority to review (Baker, 6.39). Two terrorists were living openly in San Diego. They had California IDs, were making purchases and engaging in financia l transactions, signing rental agreements in their own names, and booking flights on the plane that they would ultimately crash into the Pentagon. The FBI agent who was pursuing the trail of the terrorists had access to databases for arrests and ce rtain automobile registrations, but these two terrorists had not registered any vehicl es and had not been arrested. Because the agent did not have access to other information, it took him nearly two weeks to figure out and check out the hotel address on e of them had put on his visa when he entered the country. Baker concludes: If [the FBI agent] could have been able to find those two guys and then check the links that they had to many of the other terrorists—t here were direct shared addresses as I remember, links to the people who flew into the south tower and the north tower. We had a chance to stop this. The one chance that I can see in all of the errors that were made where we really could have prevented this. (Baker, 6.39) While Baker called it a failure of tools and rules, in the end he had constructed a tale of the agent’s inability to access the people who may have had knowledge and the


83databases that may have had relevant information. Had this information been pursued, according to Commissioner Ben-Ven iste, we certainly could have found the two individuals (6.47). A number of post-9/11 accounts of pr e-9/11 intelligence failures centered on whether the law enforcement community had too much or too little access to information. For the most part, these discussions occurred in the context of the debate between privacy and civil liberties interests. On December 8, 2003, the 9/11 Commission and witnesses examined legal authorities relating to intelligence collection and the law enforcement communi ty’s access to information. Areas scrutinized by the Commission included: the ease with which a field agent, with approval from her supervisor could issue a National Se curity Letter authorizing access to any individual’s tele phone records; whether the FB I should be able to issue administrative subpoenas wh en conducting investigations relating to terrorism; support for a national iden tification card; the Wireta p Act of 1968; the Foreign Intelligence Surveillance Act (FISA); th e Federal Privacy Act of 1974; the Posse Comitatus Act; and the Patriot Act (Baker, Ben-Veniste, Gorelick, Gorton, Kean, Miller, Roemer, Rotenberg, Schulhofer, Thompson, 6.2-61). This section is not intended to revi ew the security/privacy debate but to underscore the saliency of the construct access to information for the Commissioners and witnesses examining the why and how of September 11. The question for the participants in the 9/11 Public Hearings is, Who should and should not have access to national security information? This issue is control, and its saliency reflects common practices for most 21st century organizations where, as a matter of course, “rigid chains of command keep people from talking to anyone outside their departments . and protocol s define who can be consulte d, advised, or criticized” (Wheatley, 1994, p. 17).


84 A Law Enforcement and Prosecutoria l Approach to Fighting Terrorism A dominant theme in the 9/11 Public Hearing Transcripts was the failure of the Intelligence Community to prevent the terrorist attacks on September 11. Descriptions of these failures included the one-way street of information from Federal agencies, particularly the FBI, to state and local authorities as well as a lack of coordination, cooperation, relating, an d sharing of information among the various agencies of the Federal government. A central explanation for these failures was “the dominance of the law-enforcement an d prosecutorial approach to terrorist issues” (Lehman, 1.1.106). The following is a synthesis of these testimonies. The collection, analysis, and dissemination of intelligence in the United States has been divided between the FBI and th e CIA since 1947. The FBI produced domestic intelligence for prosecution and law enforcement, and the CIA produced strategic foreign intelligence for national security policy and de cision making. The consequence was a lack of sharing of in formation between the FBI’s tracking of radicals at home and the CIA’s tracking of radicals abroad (Byman, 1.1.241). This “gap,” exacerbated by minimal effort “to ma rry up . privileged information,” left only the FBI holding some information and the CIA with other information (Byman, 1.1.241). As a result, for example, indicati ons that airlines might be used by al Qaeda hijackers for “destructive purposes or cataclysmic terrorism,” known to the portion of the intelligence community that dealt with foreign intelligence, was not available for domestic use (Cleland, 2.1.109). Within the FBI, the sharing of information was impeded by the central mission of the FBI. The FBI was primarily a la w enforcement agency that focused on prosecuting cases (Byman, 1.1.243). Agents were responsible fo r investigating and bringing terrorists to justice. This involv ed capturing them and making a case that would result in conviction and imprisonme nt (Lehman, 1.1.268). When the FBI is trying to make a case, prosecution superseded the sharing of information:


85[When] the FBI . provides informatio n for the Department of Justice, their indictments come down. Suddenly all the information goes into the black hole. It can’t be shared, it is grand jury information, it is law enforcement information, but it is being used in the adjudication of criminal cases and, therefore, doesn’t get dissemina ted. (Fielding, 1.2.189) As a result, evidence that a national securi ty strategist within the FBI or CIA might think is very important was treated as uni mportant by someone who wanted to build a case. An FBI agent was always, and on ly, thinking about wh ether the evidence was admissible in a courtroom (Sofaer, 1.1.27 0). This is a second way in which the prosecutorial approach to terrorism fostered a lack of intelligence sharing in the intelligence community. September 11, 2001, according to al l accounts in the Public Hearing Transcripts, was a massive failure of the U.S. government, and problems associated with the intelligence function contributed to this failure. A dominant accounting for this failure by the 9/11 Commissioners and witnesses was the law enforcement approach to terrorism that created two barriers, the divide between foreign and domestic intelligence and the divide between law enforcement and strategic intelligence (Steinberg, 4.42). This approa ch, according to the witnesses, fostered a lack of sharing of information between intelligence agencies and within the FBI. The ineffectiveness of a la w enforcement/prosecutorial approach to fighting terrorism, described by witnesse s in the 9/11 Public Hearings, is characteristic of the structures of many modern organization s, and the problems created by these structures: The difficulty of achieving effective responses to ch anging circumstances is often further aggravated by the high degree of spec ialization in different functional areas within the organization. . Interdepartmental communications and coordination are of ten poor, and people often have a


86myopic view of what is occurring, there being no overall grasp of the situation facing the enterprise as a whole. As a result the actions encouraged by one element of the organization often entail negative consequences for others, so that one element ends up working agains t the interests of another. (Morgan, 1997, p. 29) The challenge, according to Commissioner Le hman, is to “come up with procedures to insure that there is full sharing among a ll the offices, . even at the expense of perhaps weakening the evidentiary sanctity of a prosecution” (1.1.107). This challenge is not to be minimized. The highly functional silos that dominate our current way of thinking about organization al structure, communi cation, efficiency and effectiveness have their origins in 17th century Newtonian sci ence and have been deeply rooted in management practices si nce the Industrial Revolution (Wheatley, 1994, p. 27). Summary and Discussion The constructs information flow, actionable intelligence, sharing and coordination of informatio n, access to information, and a law enforcement/prosecutorial appr oach to fighting terrorism dominated the accounts of pre-9/11 intelligence failures in the 9/11 Public Hearings. In the terms of the ESR model (Weick et al., 2005) presented in Chapter Two, constructs dominate as a result of redundant and consistent use of institutional frameworks to make sense of an experience: Explicit efforts at sensemaking tend to occur when the current state of the world is perceived to be di fferent from the expected state of the world . . To make sense of the disruption, . “reasons” are pulled from frameworks such as institutional constraints, organi zational premises, plans, expectations, acceptable justifications, and traditions inherited from predecessors. (Weick, et al., 2005, p. 409)


87 In this section, the transmission vie w of communication is presented as an institutional premise that significantly in-formed the dominant accounts of pre-9/11 intelligence in the Public Hearing Transcripts. In testimonies on flow of information, the focus was on the dissemination of information by the Intelligence Community to its consumers. In debates over access to various kinds of information, the focu s was on information that was or was not made available to various individuals and agencies within the Intelligence Community. In accounts of sharing and coordi nation of information, the focus was on the hand off of information within the Intelligence Community. In discussions on actionable intelligence, the focus was on the specificity of information that was handed off. The dominant explanation fo r these failures was the dominance of the law enforcement and prosecutorial approach to terrorist issues. With each of the constructs, the focus was on whether information was moved through the intelligence production process. Fundamentally, each of the constructs assumed a flow of information from a source in the Intelligence Community to a re ceiver or consumer of national security intelligence. Peeling away the language that marks the differences among these constructs, we find resounding similarities regarding the sense that was made of the pre-9/11 intelligence failures. Each can be understood in terms of what was or was not sent, imparted, transmitted, transferre d, passed, conveyed, given to others, or “otherwise gotten from person to person” (Axley, 1984, p. 430). In the language of the transcripts, the underpinning construct is what was and was not disseminated. Portrayed as a linear single-tracked process by which information is acquired and converted into a product for consumers of information, the intelligence production process constructed by the 9/11 Public Hearing participants reflects the transmission model of comm unication, the dominant view of communication in modern Western society. During the 19th century, the movement of information and


88the movement of goods and people were seen as essentially identical processes, and both were described as communication (Carey 1989, p. 15). This transmission view of communication has domina ted American thought and culture, and possibly all industrial cultures, since the 1920s (Carey, 1989, p. 23) and continues to be the most common view of co mmunication (Eisenberg & Goodall, 1993, p. 22): Our basic orientation to communication remains grounded, at the deepest roots of our thinking, in the idea of tr ansmission: communica tion is a process whereby messages are transmitted and dist ributed in space for the control of distance and people. (Carey, 1989, p. 15) When communication is viewed as transmi ssion, it is constructed as a process and technology that spread, transmit, an d disseminate knowledge, ideas, and information, with the goal of controlling space and people (Carey, 1989, p. 17). In this view, communication has been called “a metaphorical pipe line through which information is transferred from one person to another,” to accomplish their goals and objectives (Axley, 1984, p. 429). This meta phor figuratively de picts communication as a conduit where: (1) language transfers thoughts and feelings from person to person; (2) speakers and writers insert thoughts an d feelings in words; (3) words contain the thoughts and feelings; and (4) liste ners or readers extract the thoughts and feelings from the words. (Axley, p. 429) A communication model that represents the transmission view of communication is David Berlo’s SMCR model (1960): Commu nication occurs when a sender (S) transmits a message (M) through a channel (C) to a receiver (R) (Berlo, 1960, pp. 30-38). This transmission view of communication continues to dominate dictionary definitions of communi cation. Accounts of individual and coll ective behaviors, informed by this model of communication, organi ze experience so that:


891. Meaning is treated as a commodity to transfer from one place to another. Communication involves the physical transfer of thought, emotion, meaning, etc. from one place to another (A xley, 1984, p. 429; Wheatley, 1994, p. 101). 2. Communication is presented as a uni-d irectional process, or at best, a sequential process (Eisenberg & Goodall, 1993, p. 23). 3. The message receiver is portrayed as passive and uninvolved (Eisenberg & Goodall, 1993, pp. 23-24). Once the sender finds the right words to accomplish the transfer of thought, emotion, meaning, etc., then the fidelity becomes guaranteed, even routine. All th e receiver or listener needs to do is “unpack the thoughts from the words” (Axley, 1984, p. 433). 4. Accurate, perfect information is regard ed as a tool for accomplishing goals (Berlo, 1960, p. 40; Eisenberg & Goodall, 1993, p. 22). 5. Success is defined in terms of message di stribution, or the passing of ideas or meaning from one person to another. To achieve desired results, the message must be distribu ted (Axley, 1984, p. 433; Carey, 1989, p. 15). The intelligence production process cons tructed by the 9/11 Public Hearing participants resonates with the conduit or transmission model of communication: 1. Intelligence is constructed as a produc t and the U.S. Intelligence Community exists “to provide product” (Lehman, 2.2.31) precisely to the users who are tasked with defending the country. Anal ysts collect data to develop products for consumers. 2. Finished intelligence (i.e., actionable inte lligence) is solid, specific, credible. The finished product, in contrast to raw intelligence data, is used as the basis for action, national policy deve lopment, and decision making. 3. Action is causally or quasi-casually affected by the receipt or non-receipt of intelligence. For example, witnesses testified they could not prepare for


90terrorist attacks to the American ho meland in 2001 because they had not received actionable intelligence. 4. Intelligence production is a one-way street of information and consumers of information are passive recipients of in formation. For example, despite the “great deal of information that [cam e] in, sometimes 300 messages in a day,” the FAA administrator never asked “for additional assessments” from the Intelligence Community (Field ing & Garvey, 2.1.106-107). Authorized by the National Security Act of 1947, dozens of agencies and departments were formally ch arged with performing the intelligence activities (i.e., collecting, analyzing, and disseminating information) that are necessary for the conduct of foreign relations and national security of the Unit ed States. These agencies, linked by their charge, have come to be known as the Intelligence Community. With this legislation, an official definition of what is and is not intelligence was clearly demarcated: In telligence is the product of a process (collection, analysis, and dissemination of information) performed by the Intelligence Community for officials who are resp onsible for nation al security. With these definitions, we see an invisible hand sh aping the accounts of the Commissioners and witnesses in the 9/11 Pub lic Hearings. In their review of pre9/11 intelligence failures, the participants turn their attention to the intelligence production processes that have been charged to the Intelligence Community. Using “unquestioned constructs” (Schutz, 1967; see also Ax ley, 1984, p. 428) to apprehend and organize witness testimon y, the Commissioners embarked on an analysis within a taken-for-granted fr amework, the transmission model of communication, that narratively reduced the kinds of explan ations and solutions that were available to them. According to the transmission view of communication, succe ss is defined in terms of message distribution. Similarly, the success of the Intelligence Community


91is defined in terms of its dissemination of na tional security inform ation to consumers. Axley (1984) reminded us that little doubt remains about the capacity of metaphors (such as the conduit metaphor) to create pers pective, to structure points of view (p. 129). In this vein, it not unexpected that the dominant accounts of pre-9/11 intelligence failures are accounts of failures in dissemination. To wit, the consumers of intelligence testified that they depended on information from the Intelligence Community and that they did not receive it. As we discuss the dominant constructs of pre-9/11 intelligence failures in the Public Hearing Transcripts, it is important to remember that experience is the “moment of vision” before intellectualization takes place (Weick, 1995, p. 24). For example, many of us can vividly recall an instance in a young child’s life (e.g., falling off a bed, or stumbling and falling onto a rough sidewalk) when the wide-eyed child looked for others’ reactions before laughi ng, wailing, or merely getting up and brushing himself off before resuming play. This example exemplifies not only the moment of experience before intellectualization or meaning takes place, but also the multiplicity of meanings that could be atta ched to a single instance of experience. Additionally, it exemplifies the partial nature of the experience that is captured in the interpretation that is constructed (e.g., th e fallen child may respond differently to the parents’ but not a stranger’s reaction). The intellectualization of the events leading to September 11, 2001, includes the labels, e.g., lack of coordination, that were being employed in the 9/11 Public Hearings and discussed in this paper. In an effort to understand the complexity of 9/11, we (i.e., first the Commission and witnes ses, then the writer of this paper) are singling out events and orga nizing them into plausibl e explanations (Weick, 1995, pp. 55-61); that is, we are intellectua lizing the experience of 9/11 with the constructs we use to make sense of our experiences. The products of our efforts will not mirror the events leading to 9/11, bu t are a retrospective accounting of the

PAGE 100

92experience based on the constr ucts we have available to make sense of the world. As a result of the complexity of experience and the partial, partisan interpretations we apply to experience, the constructs that we create to explain the world are not as clear cut and mutually exclusive as we migh t like them to be. The event is always more complex than the stories we can te ll about it, and the stories reflect the individual and collective theories we use to make sense of the world. This chapter presented a detailed discussion of the construction of intelligence failures documented in the 9/11 Public Hearing Transc ripts. Chapter Four provides a detailed analysis of the dominant accounts of accountability in these transcripts. By looking at the constructions that were front stage in the 9/11 Public Hearings, we are creating opportunities to also consider alternative constructions that are not currently informing how America prepares its defense against this newlyacknowledged enemy. One of these, an autopoietic open systems perspective, will be explored in Chapter Five.

PAGE 101

93CHAPTER FOUR CONSTRUCTIONS OF ACCOUNTABILITY IN THE 9/11 PUBLIC HEARINGS Throughout the 9/11 Public Hearings Commissioners and wi tnesses declared their goal to assign accountability for the terrorist attacks on September 11, 2001. For the most part, the particip ants in the 9/11 Public Hearings stated their intentions not to “point fingers” (Kean, 1.1.11), participate in a “blame game” (Ben-Veniste, 1.1.30), or conduct a “witch hunt” directed at one particular agency or particular individuals (McCain, 2.1.7), as there was nothing to be gained by finding, placing, asserting, or assigning blame or pointing fingers (Corzine, 2.1.63; Waizer, 1.1.116). According to the testimonies of many of th e participants in the 9/11 Public Hearings, it wasn’t about blame but about accountability (Corzine, 2.1.63): Americans needed and wanted to know how their government may have failed them (Lautenberg, 2.1.80). Commissioner Roemer stated: “As our Declaration of Independence proclaims, those holding power, ‘deriving their powers from the consent of the governed,’ should be accountable to their citizens” (1.1.49-50). Holding government officials and employees acco untable was what the Americ an people needed, wanted, and demanded (Fetchet, 1.1.187; Lautenberg 2.1.80; Lieberman, 2.1.16), and that was what the Commission set out to do (Roemer, 1.1.49). As the hearings proceeded and th e events leading to 9/11 were reconstructed, the search for accountability prompted accounts of blame and exoneration. In the end, we find that the initial search for accountability for 9/11 was subsequently enacted as stories of: 1. Exoneration (i.e., I fulfilled my responsibilities). 2. Finger pointing and blame (i.e., the deflection of accountability for 9/11 to others who did not fulfill their responsibilities). This chapter has four sections. The firs t section presents the constructions of blame and exoneration the dominant responses by participants in the 9/11 Public

PAGE 102

94Hearings to the often unstated, but ever present question: Who was accountable for September 11, 2001? As the construct oversight is a core element of accountability in government institutions (Kearns, 1998), the constructions of “oversight” by the 9/11 Public Hearing Commission and witnesse s are also presented. Excerpts from witnesses’ accounts of pre-9/11 intelligence failures, the dominant focus of the 9/11 Hearings, are used to illustrate the constructions of blame, self exoneration, and oversight in the Public Hearing Transcripts. Whereas the first section of this chapter focuses on the construction of pre9/11 events, the second section focuses on various post-9/11 recommendations for securing America against future terrorist attacks. These recommendations provide the greatest opportunity to review the meaning of accountability in the Public Hearing Transcripts. Because self exoner ation and blaming do minated witnesses’ accounts of the events leading to Septem ber 11, these accounts provide very little insight into witnesses’ views of accountability. Additionally, there was no agenda item on the Public He aring dockets entitled accountability. Therefore, various recommendations for post-9/11 accountability are compiled to provide insight into what the Commission had in mind when th ey demanded accountab ility for officials with critical national security responsibiliti es (Lieberman, 2.1.16). This compilation is presented in the second section of this chapter. In the 9/11 proceedings, witnesse s and Commissioners used the words responsibility and accountability interchangeably. I attempt to distinguish between these in the third section of this chapter. The dominant themes of accountability presented in the first three sections of this chapter are summarized and discussed in the final section of the chapter. Who Was Accountable for September 11, 2001? Whatever the witnesses and Commission ers said about not attributing blame, the pointing of fingers was pervasive and far reaching during the Public Hearings,

PAGE 103

95ranging from the “dysfunctional society in the Arab Muslim countries” (Push, 1.1.222) and “the murderous group of thugs that hijacked and cr ashed those planes” (Shelby, 2.1.22) to our Western sense of invulnerability and the complacency in which we all shared (Cleland, 1.1.61). Th ose who were targeted in the 9/11 Public Hearing transcripts as being culpable for September 11 included: 1. The United States government (T hompson, 1.1.55)—that is, U.S. policymakers (Gasiorowski, 3.59), senior political decision makers (Lehman, 4.38), Congress (Cleland, 1.1.58; McCa in, 2.1.10; Roemer, 6.21; Shays, 2.1.78; Wermuth, 1.2.142), the Cl inton and Bush administrations (Schulhofer, 6.20; Shays, 2.1.78), Presidents Clinton and Bush (Kerrey, 7.1.55), elected offici als with oversight responsib ilities (Fetchet, 1.1.183) 2. The intelligence community (Cleland, 1. 1.57; Fetchet, 1.1.182; Gunaratna, 3.28; Kleinberg, 1.1.192; Lieberman, 2.1.15) 3. Other government agencies (Barr, 6.104) —that is, border control, aviation security (Pelosi, 2.1.5; Shelby, 2.1.26), the State De partment (Fine, 1.2.40) 4. The airline industry (Push, 1.1.216), airport securi ty (Bloomberg, 1.1.72), aviation security program (Mead, 2.1.104), the FAA (Mead, 2.1.110) 5. The CIA (Lehman, 4.73-74) and FBI (MacGaffin, 6.93) 6. The many bureaucratic steps between th e development of a counterterrorism budget within the FBI, or a CTC bu dget within the CIA, and a final administration budget request from th e Office of Management and Budget (Shelby, 2.1.23) 7. The whole political campaign contri bution system (Push, 1.1.216) 8. T-U-R-F (Harman, 2.1.46) 9. Policy failures (Ranstorp, 1.1.257) 10. Our systems, our bureaucracies, and our inflexibility toward change (Sincock,

PAGE 104

96 1.1.157) 11. The media (Shays, 2.1.78) 12. Human, budgetary, and organization al deficits (Schulhofer, 6.6) In testimonies provided by those who were targets of blame for the events leading to 9/11, there were recurring accoun ts of self exoneratio n and deflection of accountability, based on the fact that they were fulfilling what they understood to be their responsibilities. The following testimony provided by Ms. Garvey, Administrator of the FAA from August of 1997 until August 2002, exemp lifies this stance: On September 10th, 2001, by statute, civil aviation security in the United States was a shared responsibility. Air carriers had the primary responsibility for screening passengers and baggage, and for applying security measures to everything that went on th eir planes. Airports were responsible for keeping a secure ground environment and for pr oviding law enforcement support. Government’s role, that is the FAA’s role, was regulatory. By rulemaking, the FAA set the security standards for U.S. airports, for U.S airlines worldwide, and for foreign air carriers flying in the United States. The FAA also ensured compliance wi th those standards. . On September 10th, aviation security was re sponsive to the assessed threat based on information from intelligence and law enforcement agencies. Within the FAA, the Office of Civil Aviation Security was the primary office responsible for security. The FAA, as others may point out, was not an intelligence gathering organization. Threat analysis was . based on raw and finished intelligence products supplied to the FAA from these communities. . And on September 10th, based on intelligence reporting, we saw explosive devices on aircraft as the most dangerous threat. .

PAGE 105

97We were also concerned about what we now think of as traditional hijacking. (Garvey, 2.1.87-90) According to Garvey, the FAA was exoner ated for the events leading to 9/11. By being responsive to threats identified by the Intelligence Community, they were fulfilling their understanding of their responsibilities. By not providing solid intelligence on the attacks that would occur on 9/11, the Intelligence Community was culpable (Garvey, 2.1.88). Similar a ccounts of exoneration and deflection of accountability were relayed throughout the 9/11 Public Hearings. The FAA, NORAD, and Consular Affair s were among those who pointed to the Intelligence Community as the cause of th eir failure to prepare for, mitigate, or prevent the attacks on 9/11. As a result, there was considerable discussion during the Public Hearings about the Community’s failures, responsibilities, and accountability. These discussions, howeve r, also quickly prompted stories with themes of self exoneration and blame. For example, a ccording to Senator Shelby, the Intelligence Community’s performance pr ior to September 11 was not the fault of Congress: Defenders of the intelligence communi ty’s performance during the Clinton administration and prior to September 11th, insinuated that it was really the fault of Congress that the intelligence community failed to detect and deter the attacks on the World Trade Ce nter and the Pentagon . . Unfortunately, a surprising number of my colleagues in Congress seemed to give credence to the suggestion that September 11th was in some way our fault. I have at times been a harsh and I believe a constructive critic of the intelligence community. I have never asserted, however, that the attacks of September the 11th were anyone’s fault other than the murderous group of thugs that hijacked and cras hed those planes in to the symbols of American military and economic power. We should all keep that in mind as

PAGE 106

98we search for the truth here. (Shelby, 2.1.22) Congress, according to Shelby, had emph asized counter-terrorism and counterintelligence as fundamental policy priorities for years. As part of his testimony, Shelby provided an overview of the “a ggressive steps” taken by the Senate Intelligence Committee to address “what we re becoming very cl ear indications of fundamental weaknesses in our ability to attack the terrorist target” (Shelby, 2.1.23). These steps included: 1. Identifying counter-terrorism and counter-intelligence as two of the five highest priorities of the intelligence community. 2. Listing terrorism, and the ability of the U.S. to combat it, as one of the Senate Intelligence Committee’s highest priorities in every one of its bills, since 1996. 3. Revealing, in 1998, that the FBI was failing to address the significant technological challenges that were degrading its ability to track terrorists. 4. Highlighting serious FBIwide deficiencies in information technology and modernization, and the absence of a plan to address them. 5. Recognizing a critical shortage of language skills, including Arabic and Farsi, in the FBI and directing the FBI to revi ew its language recruiting efforts. 6. Working to remove restrictions that un necessarily hindered the collection of terrorism information. 7. Detailing serious problems in information sharing between intelligence agencies and law enforcement organizations. 8. Registering its concern that there was no comprehensive intelligence community estimate on present and emer ging terrorist threats and directing the Director of Central Intelligen ce to produce such an estimate. 9. Working to effect structural and organizational changes within the community.

PAGE 107

9910. Creating three Senate-c onfirmable management positions to address community-wide problems with coordination on collection, analysis, and production issues. 11. Engaging in efforts to increase funding for counterterrorism. These, according to Shelby, were just “a few examples of co ngressional actions” (2.1.24). Congress, according to Shelby’s understanding of its responsibilities, was doing what it could and sh ould be doing—that is, encouraging “these things” (2.1.25). “Ultimately,” according to Shelby, “the intelligence community is led and run by the Director of Central Intelligence (DCI), who deserves mo st of the credit or blame for the decisions he ma kes and the results that he produces” (Shelby, 2.1.25). In his testimony, Shelby iden tified pre-9/11 decisions made by the DCI that reflected a “complete disregard for cong ressional direction” (2.1.25) Examples included: 1. In 2001, prior to September 11, the CI A reported that it would not spend millions of its counterterrorism funds, despite the fact that Congress had fully funded the administration’s request for a Counter-terrorism Center (CTC). 2. The DCI resisted efforts by the Senate Select Committee on Intelligence to remove restrictions that unnecessarily hindered the collection of terrorism information. 3. The DCI, at the time of the Public Hearings in 2001, had not yet submitted names to the Senate Select Committee on Intelligence for Senate-confirmable high visibility management staff positions, wh ich were created in 1996 to “help . manage the intelligence community” (Shelby, 2.1.24). 4. The DCI, at the time of the Public Hearings, had not complied with the 1997 request of the Senate Select Inte lligence Committee to produce a comprehensive intelligence community estimate on present and emerging terrorist threats. Shelby’s testimony serves as an additional exampl e of the recurring themes of self

PAGE 108

100exoneration and the deflection of accountability that dominated the responses to the question most frequently po sed during the 9/11 Public Hearing, Who was responsible for the pre-9/11 intelligence failures? Oversight It is not surprising that oversight surfaced as a dominant construct during the 9/11 Public Hearings. Kearns (1998), revi ewing basic definition s of accountability, identified oversight as a core element of accountability in government institutions; that is, at “the heart of an y accountability system” is “ a higher authority vested with the power of oversight” (p. 144). Recogn izing the prominent ro le that oversight plays in public sector accountability, this section explores definitions and issues that inform our understanding of the constructions of accountability in the 9/11 Public Hearing Transcripts. Definitions of oversight include watchful care, management, supervision, and management by overseeing the performance or operation of a person or group. Similarly, synonyms for oversight include supervision, supervising, and superintendence. Definitions of legislativ e oversight include legislative control of bureaucracy, legislative supervision and monitoring of the executive, the intention to influence the performance of administration, formal and informal efforts to bring agencies into compliance with congressional demands, monitoring the fidelity of elected leaders and their subordinates, and maintaining accountability (Johnson, 1985, p. 550; Ogul, 1976; Ogul & Ro ckman, 1990, pp. 5, 6, 21). The purpose of oversight is to ensu re that those to whom authority is delegated remain responsive (Ogul & Rockma n, 1990, p. 7). The question is, to whom or what should govern mental agencies be responsi ve – Professional norms? Democratic values? Internal constituenci es? Public interest constituencies? Committee preferences? Congress as a whole? Multiple principals, for example, the President, the courts, interest groups and the agencies themselves (Ogul &

PAGE 109

101Rockman, 1990, pp. 7–12)? Acknowledging that they are reiteratin g what many have pointed out, Ogul and Rockman (1990) described oversight of U. S. executive agencies as “an unusually political” undertaking, one that is especially responsive to well-articulated political demands and “possibly subject to excesses of parochial influences” (p. 21). Ogul and Rockman concluded, “The United States political system fails to resolve the issues of who is boss” (p. 12). In fact it was impossible for Ogul and Rockman to imagine the U.S. system of government being otherwise. In a system that emphasizes the independence of institutions and the relative autonomy of congressional committees an d sub-committees, it was unimaginable that the multiplicity of efforts to influence and control agencies could be easily reconciled into a feasible coordinating mechanism for oversight (pp. 21-22). The U.S. intelligence community—that is executive agencies such as the CIA and FBI—is not exempt from the diffuse nature of is oversight. Inspectors general, oversight boards within the executive branch, Congress, courts, and even the media have assumed responsibility for monitoring intelligence activities (Johnson, 1985). Pre-eminent among the over seers of the intelligence communities have been Congress with its congre ssional committees (p. 550) and the President: In theory, the director of central intelligence, as the president’s principal intelligence adviser, presides over the allocation of resources of this entire system. Just so, Congress with the constitutional power of the purse, theoretically oversees the policies, organization, and efficiency of this vast army of intelligence workers. (Ransom, 1975, p. 158) However, the reality has been, and may co ntinue to be, considerably different. According to Ransom (1975), neither Congress nor the director of central intelligence appeared to provide meaningful oversight of the intelligence community. A tug of war for oversight of, and account ability for, the inte lligence agencies

PAGE 110

102was at the heart of an unprecedented comprehensive investigation of the intelligence community in the 70s (Ransom, 1975, p. 160). In 1976 Congress established the Senate Select Committee on Intelligenc e (SSCI); in 1977 the House Permanent Select Committee on Intelligence (HPSCI) was established. With the Intelligence and Intelligence-Related Activi ties Authorization Act for Fiscal Year 1979, the Central Intelligence Agency and other intellig ence agencies became subject to the congressional authorization and approp riation procedures imposed on other executive agencies (Bruemmer, 1992, p. 874 ). Authorities delegated to the SSCI and HPSCI included supervising intelligence policy, monitoring the analysis and production of intelligence, developing legislation that guides and regulates the operations of the intelligence community authorizing the annu al budgets of the Community, and overseeing and advising the President and the Director Central Intelligence on the conduct of clandestine activities (Bruemmer, 1992, pp. 874, 878; Flanagan, 1985, p. 62; Jo hnson, 1985, p. 550). However, in 1985, Johnson noted that full accountability for enforcing the accountability of intelligence agencies rema ined elusive, as Congress could not agree on how much accountability it should require and how they could ensure it (p. 549). In 1985, Flanagan also reported that all was not well with the Intelligence Community. However, he acknowledged that intelligence analysis and collection activities were generally improving, and belie ved this course could be maintained if effective management by the DCI were coupled with “judicious guidance and oversight by the President and Congress” (p. 95). In 1987, Goodman noted the long st anding problems stemming from alarming defects in the executive and cong ressional oversight of the Intelligence Community (p. 123). For exam ple, existing laws, execut ive orders, and regulations did not outline the process by which Congre ss should conduct oversight of the U.S. intelligence operations (p. 124). Goodman ca lled for a “new charter” to address the

PAGE 111

103problems that had repeatedly led to intelligence failures or abuses (Goodman, 198485, pp. 162–169; Goodman, 1987, p. 135). This charter was to delineate the process for Congress to exercise operationa l and fiscal oversight of U.S. intelligence (p.135). Berkowitz (1996), referencing at least six official and unofficial studies within the previous three years on how to improve U.S. intelligence, asserted, “If we have not figured out how to fix the intelligence co mmunity, it is not for lack of trying” (p. 35). The approaches to reform summarize d by Berkowitz in 1996 included various efforts to ensure the intelligence comm unity obeyed the law and complied with oversight requirements. The focus was on efficiency and responsiveness (p. 37) by clarifying roles, missions, an d priorities in the intelligence community. Efficiency studies, streamlining and better prioritiza tion and planning implied that there was nothing wrong with the intelligence system that better management (i.e., oversight) could not fix (p. 40). In 1975, Ransom had written: Neither Congress nor the director of central intelligence appears to have meaningful control. Therefore, it is difficult to focus policy or managerial responsibility or accountabili ty on any particular plac e within the executive or legislative branches. In the absence of such accountability, “oversight” as currently exercised is best defined in the dictionary ’s other meaning of the word—“overlooking” or the absence of careful attention. (pp. 158-159) Ransom’s observation in 1975 reverberat es in numerous comments, questions, and recommendations provided by witne sses and Commissioners during the 9/11 Public Hearings held in 2002 and 2003: 1. Comments, e.g., “I never really understood who was in charge” (Cleland, 2.1.18); “I very much endorse Senato r Graham’s views that congressional oversight is an area that screams for your attention” (Goss, 2.1.35); and “We

PAGE 112

104must . provide meaningf ul oversight of the intell igence community” (Goss, 2.1.32). 2. Questions, for example, “M ore than 12 different agen cies under six different cabinet officers were part of the intelligence community. But who was accountable (Cleland, 4.30)? “Short of the President, who do we look to for these things” (Steinberg, 4.48)? 3. Recommendations, for example, “There’s go t to be somebody in charge of the intelligence community” (Cleland, 2.1.55), and “There has to be continual oversight” (McCarthy, 4.74). Oversight: The Intelligence Function This section focuses on the dominant co nstructions of oversight that appear in the 9/11 Public Hearing Transcripts. It begins with a description of the duties of the General Accounting Office (GAO), an independent, non-partisan investigative arm of Congress: We make recommendations to the agencies who in turn are responsive to the U.S. Congress. . It is for us to present the information to the policymakers, and from there, they can make them do it . . We are a congressional agency and ou r mission is oversight, generally, of the executive agencies. I don’t want to sound like [the agencies] don’t do anything, but generally the implementation [of GAO recommendations] is not the way we want it to be. Some of the recommenda tions that have been made . have been around for years. They were ei ther not implemented or partially implemented. (Dillin gham, 1.2, 104-105) As early as 1975, it was noted that the pote ntial of the GAO as an overseer of the intelligence community had not been realiz ed (Ransom, 1975, p. 163). During cross

PAGE 113

105examination by the Commissioners, Dillingham added, “It has to be a congressional priority that is followed through” if a GAO recommendation is to be implemented (Dillingham, 1.2.104). According to the GAO officer, the GAO can not get things done, but Congress can. Senator Shelby, however, pr esented a different perspective: Congress can encourage these things, and they should. And we have certainly tried, but the legislature merely conducts oversight. We do not and should not, I believe, direct the operational activities of our intelligence agencies . . We can legislate, but there is little we can do to compel compliance. . While Congress oversees the intelligence activities of the U. S., ultimately, the intelligence community is led and run by the Director of Central Intelligence. (Shelby, 2.1.25) The threats to the United States on th e eve of 9/11 includ ed growing threats from Iran, Iraq, and Korea; the military modernization in China; threats in regional areas (e.g., Bosnia and Africa); terrorism in all its dimensions (e.g., chemical, biological, and nuclear; refugee flow; and hum anitarian disasters). When asked by a Senator at an Armed Services Committee meeting which were the most important threats, CIA Director George Tenet re plied, “If any of those happens, I’m accountable, I can’t prioritize” (Gannon, 4.67). On the one hand, the acco untability for various threats was assumed by the Director of Central Intelligence (DCI); on the other hand, others testified at the 9/11 hearings that the DCI, or the people work ing for him, could not be held accountable for the events leading to 9/11 for a variet y of reasons: The pre-9/11 Intelligence Community consisted of more than twelve di fferent agencies reporting to six or more Cabinet officers; most of the resources belonged to the Department of Defense (DOD) and the committees that provided oversight of the DOD; and the DCI

PAGE 114

106controlled no more than 20 percent of the budget (Shelby, 2.1.46). Insufficient authority, lack of resources, and no “report card” (Kleinberg, 1.1.224) have mitigated the accountability of the DCI for the events leading to 9/11. This according to Kleinberg is why “congr essional oversight becomes extremely important” (1.1.224). With this last stat ement, however, we have come full circle. According to various participants in the 9/11 hearings, the responsibility of Congress is to encourage but not direct and compel compliance; the intelligence community is run and lead by the DCI, but he can not be held accountable for the events leading to 9/11 because he did not have the author ity and resources; because the DCI does not have the resources and authority, co ngressional oversight, post-9/11, is extremely important. A higher authority vested with the po wer of oversight is a core element of basic definitions of accountability in the literature on public administration (Kearns, 1998, p. 144). On October 14, 2003, the dominant construction of institutional oversight in the Public Hearing Transcripts was captured in Gorelick’s assessment of the situation: “Right now we have a situation where there is no accountability for the entire government’s intelligen ce capacity, except in the person of the President of the United States and the Nati onal Security Advisor as hi s aide” (Gorelick, 4.14). The testimony of the National Security Advisor at the 9/11 Public Hearing provides an additional oppo rtunity to view various bu t similar constructions of oversight by the witnesses and Commissioner s. On April 8, 2004, Dr. Condoleezza Rice, the assistant to the President for nation al security affairs, appeared before the Commission and outlined the strategy to e liminate al Qaeda that had been developed by the Bush administration over the sp ring and summer of 2001 and approved by the President’s senior national security offi cials on September 4. Excerpts from Dr. Rice’s statement to the Commission provide insight into her interpretation and enactment of oversight:

PAGE 115

107The strategy set as a goal the eliminat ion of the al Qaeda network and threat and ordered the leadership of relevant U.S. departments and agencies to make the elimination of al Qaeda a high priority and to use all aspects of our national power – intelligence, financial, diplomatic and military—to meet that goal. And it gave Cabinet secretar ies and department heads specific responsibilities. (Rice, 9.5) The goal was set, and leadership was orde red to meet the goal. For example, the FBI was charged with determining the domest ic threat. However, during his cross examination of Dr. Rice, Co mmissioner Roemer reported that the Commission had not found anybody who knew anything about a “tasking” of field offices, despite thousands of interviews and a review of millions of pieces of paper: Mr. Roemer: The director . of the FBI during this threat period, Mr. Pickard . says he did not tell the field offices to do this. And we have talked to the special agents in charge. They don’t have any recollection of receiving a notice of threat. Nothing went down the chain to the FBI field offices on spiking of information, on knowledge of al Qaeda in the country, and still the FBI doesn’t do anything. Isn’t that some of the responsibility of the national security advisor? Ms. (sic) Rice: The responsibility for the FBI to do what it was asked was the FBI’s responsibility. Now, I – Mr. Roemer: You don’t think there’ s any responsibility back to the advisor of the President? Ms. (sic) Rice: I believe that the responsibility—again, the crisis management here was done by the CSG [Counterterrorism Security Group]. They tasked these things. If there was any reason to believe that I needed to do something . I would have been expected to be asked to do it. We were not asked to do it. In fa ct, as I’ve mentioned to you –

PAGE 116

108Mr. Roemer: But don’t you ask somebo dy to do it? You’re not asking somebody to . do it. Why wouldn ’t you initiate that (Rice & Roemer, 9.68)? A question posed by Commissioner Go relick during the 9/11 Public Hearing was, “How do you get the organization, th e imagination, the leadership, the urgency if you haven’t invested someone with auth ority to move resources to where they need to go” (4.14)? This overview of th e testimony on oversi ght of the executive offices, however, reveals that Congress, the GAO, the DCI, and the National Security Advisor are among those who have been vested with oversi ght authority for executive offices. In this light, the question that mi ght be posed, based on the above discussion of oversight, is: How do es someone (e.g., Congress, the GAO, the DCI, and the National Securi ty Advisor) exercise oversight once they have been granted official authority to provide leadership? This section has focused on looking ba ckwards, i.e., the construction of oversight of, and accountability for, the events leading to September 11. The next section looks forward. In recommendations for building accountability into government, there are indirect, if not direct, references to what the 9/11 Commissioners and witnesses had in mind when they demanded accountability for those who have national secu rity responsibilities. Accountability: Constructions in the 9/11 Public Hearing Transcripts The greatest opportunity to review the construction of accountability in the 9/11 Public Hearing Transcripts comes in th e form of witnesses’ recommendations for building greater accountability into the government. These recommendations included: 1. Making individuals respon sible for doing their jobs properly (Kleinberg, 1.1.224; Zelikow, 7.1.77), including following rules, procedures, and guidelines (Kean, 6.119; Steinberg, 4.36) and fu lfilling their assigned

PAGE 117

109responsibilities (Yim, 5.84) with hone sty and professionalism (Schulhofer, 6.26) 2. Providing resources, for example, suffici ent time and agents (Schulhofer, 6.7, 11) 3. Giving them both responsibility an d authority for budgets, resources and programmatic efforts (Gore lick, 4.26; Deutch, 4.19) 4. Establishing and reporting according to specific, tangible standards, goals and specific timetables (Brill, 1.2.123; Kleinberg, 1.1.224; Yim, 5.81) 5. Integrating annual performance and acco untability plans of individuals and agencies and linking them to key national indicators (Yim, 5.84) 6. Holding individuals personally accountable for failing or faltering in their duties (Lieberman, 2.1.16) 7. Having consequences for abuses (Kean, 6.119) and poor performance, for example, firing and voti ng them out of office (Kleinberg, 1.1.224), fines (Schiavo, 2.2.87), court mart ials (Gorelick, 2.7.51) Cumulatively, the composite of performa nce management techniques recommended by the participants in the 9/11 Public Hear ings echoed definitions of accountability found in the current literature on political science and public admi nistration (Kearns, 1998, p.143). For example, Kearns ( 1998), reviewing basi c definitions of accountability relating to government institutions, identified: three core elements that are at the heart of any accountability system: (a ) a higher authority vested with the power of ov ersight, (b) an explicit reporting mechanism for conveying information to the higher authority, and (c) a measure or criterion used by the higher authorit y to assess compliance by subordinate institutions. (Kearns, 1998, p. 144) Each of these core elements is identified or implied in the composite of performance management techniques recommended by partic ipants in the 9/11 Public Hearings.

PAGE 118

110Similarly, the features of accountability advocated by the 9/11 Public Hearing participants resonate with Fenstermacher’s (1979, pp. 330-331) four conceptual features of accountability: 1. Accountability is a relational term. To say that A is accountable to B implies that A is held accountable and that B is holding A accountable. 2. The accountability holds between persons.2 3. The relation holds with regard to some standard of performance. 4. The parties to an accountability relation are obligated to provide or receive information. Summarized, “Accountability is a relation between persons, wherein person A is engaged in the performance of specific tasks, and is obligated to inform person B of the standard of performance that is attain ed in these tasks” (Fenstermacher, 1979, p. 331). The literature on accountability distinguishes between basic or narrow definitions of accountability and broader de finitions. The narrower definition of accountability refers to “ans werability” for one’s actions or behaviors to a higher authority (Kearns, 1998, p. 144) and focuse s on tangible outcomes (Kearns, 1998, p. 140). A broader definiti on in the literature on a ccountability in government institutions includes assessment of intang ibles (Kearns, 1998, p. 144), for example, Schulhofer’s recommendation during the public hearings that honesty and professionalism be considered as criteria for assessing performance (6.26). This broader definition re aches beyond obeying instructions (Kearns, 1998, p. 144), to include managing diverse expectations with in and outside the organization (Romzck & Dubnick, 1987, p. 228) and responsivene ss to societal demands (Kearns, 1998, p. 144). According to Kearns (1998), broader definitions that incl ude responsiveness to 2 I n the deflection of accountability for events leading to 9/11, participants in the 9/11 Public Hearings also blamed organiza tions, processes, and structures for the attacks on 9/11.

PAGE 119

111stakeholders and persons in positions of authority are better suited to public organizations whose mandates are to preser ve the public trust and serve the public interest (p. 141). When the testimonies on preand post-9/11 accountability are extracted from the 9/11 Public Hearing Transcripts and compiled, a construction of accountability emerges that is consistent with the scholar ship and practices th at dominate Western society. Used by practitioners and schol ars, the construct often refers to the delegation of authority by “B” to “A,” followed by the answerability of “B” to “A” for the performance of assigned tasks. Sa tisfactory performance is achieved by measuring up to the expectations of “A.” At the core of this generally-used construction of accountability is the notion of fulfilling one’s responsibilities; this is the sense of accountability that also dominated the construction of accountability in the 9/11 Transcripts (e.g., representative s of FAA, Consular Services, and NORAD each testified they were fulfilling their respon sibilities). Therefor e, to understand this construction of accountability more fully, we must dig deeper into the construction of responsibility. In “On Being Responsible,” Haydon (1978) focused on the “notion of responsibility” and the various senses we make of it. Two “senses” that are particularly relevant to the constructions of accountability by the 9/11 Public Hearing participants are liability-responsibility (p. 47) and virtue-responsibility (p. 46). Someone may be held responsible in the liability sense (and hence blamed or praised), when: (a) an agent has particul ar responsibilities as a function of occupying a distinct role and we refer to those as his responsibilities and (b) the agent’s action or omission of action is involved in the causation of an event for which he has the responsibility (p. 46). Haydon’s definition of liability-responsibility resonates with the construction of a ccountability by the 9/11 Public Hearing participants (e.g., assignment of responsibi lities, evaluation of performance based on

PAGE 120

112a person’s performance of those resp onsibilities, and consequences). This sense of responsibility/accountability, according to Haydon, is unitary and monolithic (p. 47); that is, judgments of responsibility and irresponsibility are made in relation to a person’s performance of particular roles (p. 51). It does not account for virtue-responsibility, a judgment that a person is responsible in the “undertaking” of a responsibility (p. 50). This sense of responsibility, virtue responsibility, according to Haydon (1987) can best be elucidated in terms of accountability: We need not assume that the notion of responsibilities has sense only in a context of roles. I suggest that while the responsibilities constitutive of a role are paradigmatic for “roleresponsibility”, ‘responsible’ can be used in the same sense without reference to a role (this use can, I think, be elucidated independently in terms of accounta bility). (Haydon, 1978, p. 50) The narrower sense of responsibility/accountabi lity restricts accountability to fulfilling delegated responsibilities; the broader definition of accountability is not tied to particular responsibilities, but extends to any agent in regards to any decision to act in one way or another (p. 52). The co mponents of this broader notion of accountability, or virtue-responsibility, are: Having regard for consequences . . it is required not merely that one act with regard for the consequences, but that one evaluate and weigh the consequences properly. The agent is expected to pay special attention to consequences involving harm and benefit to others, and to modify his conduct so as to promote benefit or at least minimize harm. . (Haydon, p. 52) Being autonomous. . .The autonomous [human], it would generally be agreed, forms his own judgments; it is often added that he forms them rationally; and sometimes that he also acts on them. The requirement in any case goes beyond . the notion of a conscientious . follower of some

PAGE 121

113conventional code (pp. 52-53) Using this broader sense of accountability, the accountable person has an appreciation of her situation as an agent in a social wo rld and understands that she is expected to perform in a way that prom otes benefit or at least minimizes harm. An example of this sense of accountability is taken from the 9/11 Public Hearing Transcripts. On January 26, 2004, Jose Melendez-P erez, a twelfth-year immigration inspector, appeared before the 9/11 Commi ssion to testify on his encounter with a Saudi male at the Orlando International Ai rport on August 4, 2001. Mr. MelendezPerez opened his testimony (7 .1.57) with two “things” he wanted the Commission to know. On the day in question, he was “j ust doing [his job],” and he was honored and proud that the work he did apparent ly prevented an alleged terrorist from participating in the September 11 attacks on America’s homeland (7.1.57). Second, he relied on his experience, training (7.1 .57), and “a gut feeling that something wasn’t right” (7.1.69). At approximately 17:35 hours on Augu st 4, 2001, Mr. Melendez-Perez was assigned the case of a Saudi national who had arrived on Virgin Atlantic flight 15 from London Gatwick Airport. The excerpts from the cross-examination of Mr. Melendez-Perez by Commissioner Ben-V eniste provide insight into the Commissioners’ construction of a first class profession al doing his job (Lehman, 7.1.72; see also Roemer, 7.1.69): Mr. Ben-Veniste: Let me ask whether it is correct that at no point during the summer of 2001 did you receive any noti fication that there was a higher danger or threat level of potential terrorists coming into the United States? . . Mr. Melendez-Perez: Not that I recall, sir. Mr. Ben-Veniste: And in looking back at this matter, had you received any

PAGE 122

114information as to whether this individual, Mr. Kahtani, had actually been interviewed in connection with his visa application to obtain a visa to visit the United States? Mr. Melendez-Perez: That is negative, sir. Mr. Ben-Veniste: And that is the case that our staff has found no indication that Mr. Kahtani was physically interviewed by a consular officer. Let me ask you this, as of August ‘01, when this incident occurred, among the INS inspectors in Orlando, was there a recognition, sir, that greater deference would be given to a Saudi national than say, a Mexican or a Jamaican? . Mr. Melendez-Perez: Well, since I came in the service in 1992, the [consensus] of the Saudi people was they have to be treated with more tact for their nation. . basically, that feeling is communicated by the more experienced inspectors. Mr. Ben-Veniste: That if you hassled a Saudi citizen or that you took more time up to the point even of permitting entry but you gave them more attention that you might catch some kind of negative criticism? Mr. Melendez-Perez: That is correct. Normally, as a matter of fact, the day that I was working on this particular incident, one of my co-workers stated to me, “You’re going to get into trou ble because you’re trying to refuse a Saudi.” My answer was, “You know, I have to do my job and I cannot use nationality as guidance how to do or conduct my interview or take care of business” (7.1.61-62). Mr. Ben-Veniste continued the cross ex amination by verifying the factors that Mr. Melendez-Perez considered during his secondary inspection of Mr. Kahtani. Mr. Kahtani, a twenty-six-year-old man, was tr aveling alone, wherea s the vast majority of Saudi visitors to Orlando travelled in family groups. He appeared fit and had possibly undergone mili tary training. He appeared unable to speak or understand

PAGE 123

115English. Despite his lack of facility in the English language, he had no hotel reservation and gave contradictory answer s as to who was supposed to assist him either immediately or within three days of his arrival. He was traveling on a one-way ticket and had no good explanation abou t where he was going once he departed from the United States; he did not have sufficient funds to spend six days as a tourist in the United States and then to purchase a return ticket to Dubai. He refused to provide information about the identity of the individual who would be meeting him or the friend who was going to lend him money. And he refused to answer questions once Mele ndez-Perez administered an oath. His attitude was unusual. He was arrogant and combative from the start, fixing Me lendez-Perez with a piercing glare, and became even more confrontational as the interview process proceeded. The language interpreter co nfirmed there was something seriously suspicious about Kahtani (7.1.63-64). Commissioner Ben-Ven iste continued his cross examination: Mr. Ben-Veniste: And on the basis of all these factors, you concluded that Kahtani might well be a hit man here in the United States to do harm. Mr. Melendez-Perez: That is correct. Mr. Ben-Veniste: And yet, despite al l these factors, because you were dealing with a Saudi national, you were not certain by any means that your superiors would agree with your de termination to deny entry. Mr. Melendez-Perez: That is correct. (7.1.64) After Melendez-Perez presented his ca se to the supervisor, the assistant airport director (AAPD) wa s contacted for further inst ructions. The supervisor provided the synopsis and Melendez-Perez answered a number of questions from the AAPD. Additionally, Melendez-P erez explained to the AAPD, “When the subject looked at me, I felt bone chilling cold . bottom line, he gave me the chills. You would have to be present to understand what I’m trying to explain” (7.1.60).

PAGE 124

116 Based on the fact that Melendez-Perez was able to convince the AAPD that Kahtani was “malafide,” i.e., “his true intent in coming to the United States was not clear and he appeared to be very evasiv e” (Melendez-Perez, 7.1.60), Kahtani was asked to withdraw his application or be se t up for expedited re moval. Along with another immigration inspector, Melendez-P erez escorted Kahtani to his departure gate. Perez-Melendez testified, “’Before bo arding the aircraft, the subject turned to the other inspector and myself and said in English something to the effect, ‘I’ll be back’” (7.1.61). On August 4, 2001, Kahtan i departed via Virgin Atlantic flight 16 to London for a connection flight to Dubai (7.1.61). Ben-Veniste closed his cross examin ation of Melendez-Perez with the following: Mr. Ben-Veniste: Now, as we now know, with the benefit of investigations subsequent to 9/11, Mohamed Atta, perhaps the ringleader of all the terrorists here in the 9/11 plot, was at Orlando International Airport on August 4, 2001, the very day that Moha med Kahtani claimed at least in part of his interview with you that someone was upstairs to meet him. And we know that Mohamed Atta made a teleph one call from that location to a telephone number associated with the 9/11 plot. On the basis of that information, as well as significant additional information which we are now not at liberty to discuss in public session it is extremely possible and perhaps probable that Mohamed al Kahtani was to be the 20th hijacker. Based on that premise, and taking into account that the only plane commandeered by four hijackers, rather than five, crashed before reaching its target, it is entirely plau sible to suggest that your actions in doing your job efficiently and competently may well have contributed to saving the Capitol or the White House, and all the people who were in those buildings, those monuments to our democracy, from be ing included in the catastrophe of

PAGE 125

1179/11. (7.1.65) Melendez-Perez stated that he “was just doing [his] job” (7.1.57). According to Lehman, Melendez-Perez was “a first class professional” who did his job (7.1.72). In this context, it appears that doing your job included using skills and common sense (Ben-Veniste, 7.1.68); “feeling in your he art and your gut” (Roemer, 7.1.69); and “doing [a] job in a way that makes us all proud” (Ben-Veniste, 7.1.68). Melendez-Perez chose cert ain sets of consequences, recognizing that he could be called to account for the consequences of his action. Guided by a sense of responsibility that was not defined by a narrow interpretation of his position description, he did what he felt he had to do. He stated, “You’re better off to make a wrong decision and send some body home . than to admit somebody because we will be afraid of . cong ressional letters or letters from somebody” (7.1.66). Mr. Melendez-Perez’s expectation, or his understanding of his job, was to promote benefit or minimize harm. His advice to the higher ups from th e people on the line everyday doing their job was “understand that we have to do what we have to do” (Melendez-Perez, 7.1.66). Melendez-Perez pushed the edge of the envelope with his persistent secondary inspection of the Saudi travel er, and risked getting into trouble and negative criticism (Ben-Veniste, 7.1.62). With this broader construction of accountability, the primacy of promoting bene fit or minimizing harm to others takes precedent over fulfilling role responsibilities. As a result, a person’s decision or action might be judged respon sible even though it runs counter to those particular responsibilities for which he has been charged (Haydon, 1978, p. 51). The testimony of Major General Retired Arnold, Commander of North American Aerospace Defense Command (NORAD ) on the day of the terrorist attacks, provides a clear example of an instance wh ere the right thing to do ran counter to protocol. Hijacking is a law enforcement issue and any assistance the military might

PAGE 126

118provide in these circumstances must first be requested from law enforcement. On September 11, the Boston Center of NORA D had called a possible hijacking within the system and had put the airc raft at Otis on battle statio ns. According to protocol, Boston had to call NORAD for permission to scramble the planes; NORAD in turn was to obtain permission from someone represen ting the Secretary of Defense. Once the request is approved, aircraft are scrambl ed. On September 11, General Arnold testified, “We didn’t wait for that. We scrambled the aircraft, told them [to] get airborne, and we would seek clearances later” (2.2.33). With the benefit of hindsight, Commissioner Goreli ck applauded Arnold’s blatan t disregard of protocol: In fact, General Arnold, I am glad to see and hear that when faced with the judgment of whether you should do your job in defending the United States or wait for someone from the FBI to call you, you decided to get the authority later. (2.2.50-51) At the same time she praise d his decision to act, she paradoxically acknowledged the risk, “It probably could have gotten you co urt-martialed. But one appreciates that sort of leadership” (2.1.51). Summary and Discussion The initial plans of the 9/11 Public Hearing Commissioners and witnesses to assign accountability for the September 11, 2001, terrorist attack s was subsequently enacted as 1. Accounts of exoneration (i .e., I was doing my job). 2. Blame (i.e., deflecting accountability to others who weren’t doing their job). 3. Recommendations for securing America’s future. Similar to the previous discussion on the accounts of intelligence failures, this section begins with the ways in which the transmission model informed the Commission’s construction of preand post9/11 accountability.

PAGE 127

119 According to the transmission view of communication, sp eakers and writers insert thoughts and feelings into words and transfer them to listeners and readers. The process is a uni-directional or sequen tial process. List eners are passive and uninvolved. All they have to do is unpack the information and use it to accomplish goals. Success is defined in te rms of message distribution. Accountability for the events leading to September 11, 2001, was similarly constructed in the Public Hearing Transcripts. There are disseminators of information who claimed exoner ation. They were doing their jobs. Senator Shelby testified, for example, that it was not th e fault of Congress that the Intelligence Community failed to detect and deter the attacks on the World Trade Center (2.1.22). The “aggressive” steps taken by the Senate Intelligence Committee to deter terrorism included emphasizing, identifying, listing, revealing, highlighting, detailing, and registering problems and co ncerns (2.1.23). Congress, according to Shelby, was doing what it could and should have been doing. In the same way, Dr. Rice, Assistant to the President for national security affairs, set goals and charged leadership and field offices to eliminate th e al Qaeda network. Th e responsibility of the FBI was “to do what it was asked” (Rice, 9.68). Similarly, we find examples of receivers of information who claimed exoneration. The FAA, NORAD, Consular Affairs and other consumers of information passively received and used the information that was provided. Despite evidence of the growing threat to American assets an d a history of airlines being used as weapons of mass destruction, additional info rmation was not sought out. They were also doing their jobs. Likewise, blame was defl ected to senders and receivers of information who were not disseminating and using informat ion in ways that are assumed by the transmission model of comm unication. The Intelligence Community, disseminators of national security information, did not provide actionable intelligence to its

PAGE 128

120consumers. As receivers of communication, the CIA and FBI, respectively, did not respond to the direction that was provided to them by Congress and the Assistant to the President for national security affairs. Viewed from a transmission mode l of communication, the dominant constructions of accountability for pre-9/11 intelligence failures in the 9/11 Public Hearings can be summarized in terms of senders and receivers of information who were/were not doing their jobs. Similarly the strategies for accomplishing post-9/11 accountability that were recommended during the Hearings is consis tent with the conduit or transmission view of communication, i.e., accountability is the delegation of authority by “A” to “B,” followed by the answerability of “B” to “A” for her performance of assigned tasks. Satisfactory performance is achieved by measuring up to the expectations of A. Depicted in this construction of accountability is a tr ansmission view of communication. More specifically, account ability is a process whereby: 1. The content of each job, what the worker is supposed to do while at work, is formalized, explicated, and made rout ine in approved descriptions and training. 2. Workers are expected to extract the rule s, responsibilities, and expectations and perform in a corresponding manner. 3. Evaluation of employee performance is subsequently based on how well the employee has extracted and enacted the responsibilities and expectations that had been set forth in detail. The features of the conduit view of communication inhered in this definition include: use of language to transfer thoughts and feelings from one person to another; a sequential flow of information between th ose who delegate and those who are held accountable; the extraction of meaning from words by a passive listener or speaker, for the purpose of accomplishing a goal; an d success defined in terms of the fidelity

PAGE 129

121between intended and enacted meanin g (Axley, 1984, p. 433). For the most part, approved position descriptions are the tools that define performance requirements for a position; however, it is also acknowledged that position descriptions cannot “f ully mirror” or define all requirements (Edwards, 1984, p. 110). Two instances from the 9/11 Pub lic Hearings where this was the case included: 1. General Arnold’s appropriate disregard for protocol when the unprecedented situation on 9/11called for a scrambling of aircraft before acquiring standard clearances. 2. Mr. Melendez-Perez’s relentless interrogations of an arrogant, combatant, Saudi national at the Orlando International Airport despite concerns from his colleague that he may lose his job for hassling a Saudi national. Setting aside irregular duties or even the occasional except ional circumstances requiring efforts outside the job description desired performance in organizations is generally formalized, explicated, and made routine by way of precisely specified tasks in position descriptions. In Chapter Three, the tran smission view of communica tion is presented as the institutional premise that significantly in-formed the dominant constructions of pre9/11 intelligence in the Public Hearing Transcripts. Similarly, this view of communication appears as the dominant institutional premise that significantly informed the constructions of accountability by the participants in the Public Hearings. Dominant frameworks used retros pectively to make sens e of our past also prospectively inform our future. By l ooking at current frameworks, we create opportunities to consider alternative ways of constructing both the past and future. In this vein, Chapter Five searches for a deeper understanding of the institutional and cultural frameworks that informed the construction of intelligence and accountability in the 9/11 Public Transcripts. Also presented is an alternative

PAGE 130

122framework for retrospectively constructing past accounts of 9/11 and prospectively organizing ourselves in ways that more effectively address th e complexities of today’s world.

PAGE 131

123CHAPTER FIVE ALTERNATIVE VOCABULARIES FOR MAKING SENSE OF SYSTEM RESILIENCE: MECHANISM AND AUTOPOIESIS Mental models acquired and retained through life inform the interpretations of, and responses to, events (Weick et al ., 2005, p. 411). In the case of the 9/11 Public Hearings, the transmission view of communication has emerged as the dominant mental model informing the retrospective construction of the events leading to September 11, 2001. This obse rvation is aligned with Carey’s (1989) perspective of nearly two decades ago that the transmission vi ew of communication is a dominant “motif” of Western Society; therefore, “Our thou ght and work have been glued to a transmission vi ew of communication” (p. 19). Additionally, the transmi ssion view of comm unication is “congenial with the underlying wellsprings of Am erican culture” (Carey, 1989, p. 19)—that is, the instrumental, machine-like vi ew of society, organiza tions, and communication (Eisenberg & Goodall, 1993, p. 22). We live in a mechanistically-oriented nation. Not only have machines raised levels of productivity, they also have influenced virtually every aspect of our lives; for example, scientists, philosophers, and psychologists have produced mechanistic interpretations of the natural world and human mind and behaviors (Morgan, 1997, p. 12). This is readily seen in the mechanistic construction of communication and intelligence production processes as the transmission of information from a send er (e.g., the Intelligence Community) to a receiver (e.g., consumers of intelligence). This “dialectic” between our material and subjective worlds was explored in The Social Construction of Reality (Berger & Luckmann, 1966, p. 61): Our understanding of the world is not only shaped by our taken-for-granted reality, but also re -enacted and perpetuated, as this understanding informs our participation in, and contributions to, this world (Berger & Luckmann, 1966, pp. 21-22).

PAGE 132

124 With the goal of further revealing th e sensemaking structures and processes that shaped the 9/11 Commission’s acco unts of accountability and pre-9/11 intelligence failures, this chapter explores the mechanistic view of Western society (i.e., the underlying wellspring of American culture) that has significantly informed these accounts. Additionally, this chapter presents autopoiesis as an alternative descriptor for system resilience and accoun tability for organizations residing in complex, dynamic, unpred ictable environments. The 9/11 Public Hearing Transcripts and the following discussion on mechanism and autopoiesis also re flect the changing meaning of systems in organizational research and practice. Essentially, general systems theory is a science of organizing and or ganization (Ruben, 1979, p. 95). Aspects of it can be traced to Aristotle who conceived of a state as “villages, which are in turn made up of households, which contain families” (Ruben 1979, p. 96). Conceiving of things in terms of wholes and interrelated parts cont inues to be a basic concept in general systems thinking. The more recent history of general systems theory dates back to the 1950s when the theoretical biologist Lu dwig van Bertalanffy, credited by his peers as the father of modern genera l systems theory (Ruben, 1979, p. 97), developed the open systems approach to organizations. This approach builds on the principle that organizations, like organisms, must achiev e appropriate relations with their environment if they are to survive (Morgan, 1997, p. 39). The basic framework emerging from an open systems approach incl udes the following: 1. A system implies wholeness and su ggests the presence of parts in relationship with one another (Ruben, 1979, p. 99). 2. Systems are embedded within physical spatial, temporal, and sometimes symbolic sets of environments and conditions (Ruben, 1979, p. 101).

PAGE 133

1253. Boundaries not only hold together the components which make up a system, but also selectively admit an d block out various inputs (Ruben, 1979, pp. 100-101). Open systems survive only through conti nual exchanges with their environments, maintaining themselves through “metabolism of matter-energy and the metabolism of information” (Ruben, 1979, p. 104). In contrast, “closed” sy stems, a mechanistic model which comes from conventional physics, function in isolation from their environment (Ruben, 1979, p. 104). Much of organizational theory since the 1920s has sought to overcome the limitations of the mechanistic perspect ive, and many of the most important developments in organizational theory over the past sixty year s reflect a definite move away from a closed to an open syst em perspective as a source of ideas for thinking about organizations. However, mechanistic models of organizing haven’t simply disappeared with the advent of op en systems theory. In fact, there are schools of system theory (e.g., structur al functionalism) that are extremely mechanistic. Part of the appeal of system s theory is its ability to account for the environment, while allowing for description, prediction, and control (e.g., norms, rules, procedures, hierarchy), concepts th at have roots in mechanistic thinking (Eisenberg & Goodall, 1997, p. 98). Additi onally, people often rely on the language of the old vocabularies of mechanism because they have not yet acquired fluency in new ones. The following discussion on me chanism and autopoiesis, with examples from the 9/11 Public Hearing Transcripts, provides a contemporary snapshot of how these dynamics appear in practice and the ways in which scholars and practitioners are turning to systems concepts to keep abreast with increasingly complex environments.

PAGE 134

126 Mechanistic View of Organizations, Intelligence Failures, and Accountability Generally, when speaking of organizations, we have in mind a state of orderly relations between clearly defined parts th at have some pre-determined order for delivering a pre-determined product. For example, the Intelligence Community was created to collect, analyze, and disseminate information for the purpose of national security. Although the image that undergirds this process is implicit, we are referencing a set of mechanic al relations created to ac hieve other ends (Morgan, 1997, p. 13). Guided by the implicit image of a machine, we expect organizations to perform similarly: to produce predictable ou tcomes by performing “in a routinized, efficient, reliable, and predictable way” (Morgan, 1997, p. 12). Organizations have be come progressively more mechanized since the widespread use of machinery during the Industrial Revolu tion, as evidenced by the increased presence of the following in the wo rkplace: division an d specialization of labor; control of outcomes by supervision and use of machinery; reduced discretion of workers; the separation of planning and labor; and the introduction of new procedures and techniques for the purp ose of routinizing employee performance (Morgan, 1997, p. 15). This is the cu ltural milieu in wh ich Senator McCain, congressional sponsor of the 9/11 Pub lic Commission, charge d the Commission to “recommend . reforms [that] . .rationalize the way intelligence information is collected, analyzed, dissemin ated and acted upon to impr ove the effectiveness of our efforts to deter, preempt and count er extremist terrorism” (2.1.8). Table 1 (Chapter 1, p. 6) provides an overview of the infl uence of mechanistic thinking on organizational research and practices. For example, an analysis of organizations based on a mech anistic mode of thought fo cuses on a review of the labor, materials, instruments, and tools (including goals, tasks, roles, processes, and techniques) that are being used to attain a goal or outcome. Similarly, the

PAGE 135

127interventions designed to improve mechanisti cally-driven performance focus on efforts to further routinize and automate human thought and action. Potential targets for these efforts include revisions to the machinery, mate rials, instruments, tools, and labor involved in the production of specified outcomes. The ways in which mechanistic thinking influences organization al analysis and interventions can be seen in the 9/11 Commission’s investigation of th e events leading to September 11, 2001. For example, On April 1, 2003, Gl enn Fine testified before the 9/11 Commission “about the work of the Department of Justice Office of the Inspector General [OIG] on border security issues” (1.2 .5). Based on the “significant body of work by the OIG” during the previous seve ral years, Fine presented “several broad themes that the Commission may want to examine relating to border security” (1.2.13). The following excerpt from Fine’s testimony is presen ted to exemplify the ways in which mechanistic thinking framed issues during the 9/11 Public Hearings: First, information and intelligence sharing among all levels of government remains a critical component of the effort to prevent terrorist attacks in the United States. Without adequate intelligence, the ability of front line employees to screen effectively those who seek to enter the country is limited. Second, our reviews have foun d that the current systems for identifying when aliens enter and leav e the country are clearly inadequate. Implementing an effective entry-exit tr acking system is a daunting challenge that will require substantial efforts and a large investment of resources. Third, we encourage the Commission to focus on the often-overlooked issue of human capital. To fulfill its mission, the Department of Homeland Security must have sufficiently trained immigration staff and supervisors. Historically, this has been a challenge for the INS. Fourth, I think it is also important to note that timely and consistent

PAGE 136

128 processing of the millions of benefit applications has been a longstanding problem for the INS, and now for the DHS. An enhanced focus on border security should not override this im portant service-related responsibility. And finally, the transfer of the INS to the Department of Homeland Security presents enormous management challenges. The transfer will not, in itself, resolve the issues I have identifi ed today. Solutions to border security issues will require innovation and aggressive management oversight. (Fine, 1.2.13-14) When issues are framed in terms of th e efficiency and effectiveness of flows of information, tools, people, processes, and structures, proposed solutions also focus on these phenomena. For example, better traini ng for baggage screeners and more effective use of the explosive de tection machines were included in recommendations for making us safer when we fly (Dzakovic, 2.1.138; Mead, 2.1.92). These examples have been presented to illustrate the dominant thinking that was in play during the Commission’s investigation of the events leading to the acts of terrorism on September 11, 2001. The ne xt section specifically considers the influence of mechanism on th e discussion of intelligence and accountabi lity during the 9/11 Public Hearings. Mechanistic Thinking: Pre-9/11 In telligence Failures and Accountability In Chapter 3, we saw how the pre-9/11 intelligence production process constructed during the 9/11 Public Hearings resonates with the transmission model of communication, a mechanistic view of the world. As witnesses and Commissioners constructed their accounts of pre-9/11 intelligence failures, they also recommended “fixes” (e.g., Fielding, 7.2. 119; Tenet, 8.2.39), for example, the solutions for addressing problems of info rmation flow and sharing and coordination information. The two strategies that dominated these discussions during the 9/11

PAGE 137

129Public Hearings were: (a) revamping or restructuring the Intelligence Community and (b) deploying a performance management system or system of accountability. The following quotes from Commissi oners illustrate the marriage of mechanistically-framed problems (i.e., failures to disseminate information) and mechanistically-driven solutions in the 9/11 Public Hearing Transcripts (e.g., restructuring the Intelligence Community for the purpose of holding a single person accountable, increasing centralization, and ensuring division an d specialization of labor). 1. Commissioner Roemer: A little bit ab out the problems of communication and information sharing between the CI A and the FBI. . How do you fix it? Do you create more agencies to do it and more stove piping . or do you create this new Terrorist Threat Integration Center? Where do you put it, Homeland Security or outside of it? (1.1.307) 2. Commissioner Kean: We’ll be asking ourselves whether the FBI . or whether a new entity should be established to perform collection, analysis and dissemination of intelligence within the United States, primarily to prevent, curtail and combat terrorism. (6.2-3) Similarly, the following recommendations for organizing the post-9/11 Intelligence Community are presented to further illustrate the influence of mechanism on the construction of solutions, as well as problems, during the 9/11 Hearings. Proposed strategies for prev enting, preempting, or deterring future terrorist attacks on the U.S. homeland included: 1. Making do with the current organization of domestic intelligence with significant improvements; fulfilling existing authorities, as there is no real need for additional organizational structure (Fielding & Wermuth, 1.2.180-183).

PAGE 138

1302. Developing a virtual organization that pulls information into central places within the Federal Government and then moves it to people who need it, when they need and want it (Baird, 1.2.147-149). 3. Starting with “a blank piece of paper” (Larsen, 1.2.188-189) and ending up with an information integration center that will help with prevention, mitigation and response (Larsen, 1.2.162), but cannot report to the Director of Central Intelligence (B aird, 1.2.185; Larsen, 1.2.187) or the Department of Justice (Larsen, 1.2.187). 4. Using a different model (for example, a separate British MI5 type of organization) (Deutch, 4.20; Gorton & Push, 1.1.212); the Canadian bifurcation or French models; or so mething else (Ben-Veniste & Larsen, 1.2.199-200). 5. Instituting a system or a new organization of domestic intelligence, for example, creating a separate domestic security organization in Justice, but separate from the FBI, in the Departme nt of Homeland Security, or as an independent agency reporting to a Di rector of National Intelligence (Gorton & Steinberg, 4.46). 6. Placing a new domestic intelligence service under the direction of Central Intelligence, reporting along with the CIA and Department of Defense Intelligence Service to a Director of Central Intelligence (Deutch, 4.1819). 7. Creating a Director of Central Intelligence (Deutch & Fielding, 4.22) or a Director of National Intelligence, somebody with centralized power and authority within the intelligence community, including budget, implementation of policy and marshaling resources in the war against terrorism (Ranstorp, 1.1.315). Many of these discussions were driven by the need to “break down the walls”

PAGE 139

131(Deutch, 4.20; Mueller, 10. 118, 122; Rice 9.10; Thompson, 6.15) within the FBI, and between the FBI and the CIA, for the pu rpose of ensuring appropriate access, information flow, and the sharing and coordination of information. In this light, the discussion on the restructuring of the intelligence community appeared as a dominant solution to many of the issues that were most frequently discussed in the 9/11 Public Hearings. Similarly, the second dominant fix recommended by the participants in the 9/11 Public Hearings was a performance management system or system of accountability th at clearly delineated performance expectations, processes for reporting outcomes, and rewards or cons equences for performance (Chapter 4). In each of the two recommendations that dominated discussions on deterring future terrorist attacks, the focus was on improving the current practices of the Intelligence Community by further routinizing processes and delineating structures and lines of authority. Mechanistic Thinking: Di scussion of the 9/11 Pu blic Hearing Transcripts In the 9/11 Public Hearing Transcrip ts, we find proposed system responses (two dominant recommendations informed by mechanistic thinking) to perceived environmental exigencies. This section focuses on the enviro nmental exigencies constructed by the 9/11 Public Hearing participants (i.e., international terrorism in the 21st century) for which these recommendations are a proposed response. September 11, 2001, according to a number of the 9/11 Public Hearing participants, was a “wake up call” (e.g., Fetchet, 1.1.180-181; Ranstorp, 1.1.312; Shays, 2.1.77). That is, as a nation, we came to realize that we live in a world in which threats developed an ocean away can st rike with “horrifyin g impact within our own borders” (Hamilton, 1.1. 17). Our “downfall” before 9/11, according to Commissioner Cleland, was a blind eye to the changing dynamics in the environment (Cleland, 1.1.62). Since 9/11, according to a number of the 9/11 Public Hearing witnesses, we

PAGE 140

132are a changed nation. However, we are re minded by various witnesses during the 9/11 Public Hearings that our mindsets, not the attacks on 9/11, have changed everything. Nearly 60 years ago with the 1947 National Security Act, intelligence structures, responsibilities, and authorities were established to address the challenges of the Cold War. On the eve of the terrorist attacks, however, communism had not posed a nati onal security threat to United States for nearly ten years; yet, our national intelligence ag encies had not re-oriented its mindset, culture, and institutions to the new form of terrorism that was emerging (Graham, 2.1.42): Our [pre-9/11] collective policy intelligence minds weren’t tuned to look the right way, even though th ere probably should have been plenty of evidence that we should have. . Our mental filters aren’t designed right . You want to proactively . try to properly tune your filters before something happens. (Hamre, 6.99) In the 1980s and 90s, a new terrorism ba sed on ethnic hatreds or religious fanaticism had emerged. Enabled by tec hnology, the modus operandi of terrorists had changed from “we ‘will cause damage and terror but not kill lots of people’ to ‘we will terrify people and kill thousands of them to get their attention’” (Roemer, 1.1.52). With technology to aid in the pl anning, coordination, an d implementation of terrorist acts, “small groups” had become serious threats to national security (Cleland, 1.1.63). This new form of terrorism was charac terized by various wi tnesses during the 9/11 Public Hearings as: 1. An “open-ended thing” (Jenkins, 1. 1.318), with no end state (Mead, 2.1.91) 2. A global network with a pr esence in over ninety-eight different countries (Ranstorp, 1.1.313)

PAGE 141

1333. A changing adversary, continually chan ging the way it operates (Mineta, 2.2.3) Additionally, it was characterized as being: 1. Unencumbered by physical and international borders (Cleland, 1.1.63); for example, a terrorist with a wireless Internet co nnection in a cave in Afghanistan can learn as much ab out America as any of us know (Schumer, 2.1.66) 2. Empowered by technology “to the po int that small groups, or even individuals, can now inflict a degree of death and destru ction heretofore reserved to great armies” (Cleland, 1.1.63) 3. Emboldened by the fact that terrori sm has breached the borders of the United States (Wermuth, 1.2.212) This synthesis of post-9/ 11 testimonies suggests a view of the environment that did not exist prior to September 11. In response to these environmental exigencies (an open systems concept), the witnesses recommended further routinizing and controlling the activities of the Intelligence Community by restructuring the Community and implementing a performance management system. On the one hand, the 9/11 Public Hearing witnesses acknowledged dynamic environments; on the other hand, their recommendations were largely informed by mechanism, an approach that is most effective when the environment is stable, there is a straightforward task to perform, one wishes to produce exactly the same product time and again, and the human “machines” will perform as they have been instructed to perform (Morgan, 1997, p. 27). This nexus of the mechanistic and open system views of organizations is clearly articulated in Katz and Kahn’s The Social Psychology of Organizations (1966), “a landmark application of system theory to organizations” (Eisenberg & Goodall, 1993, p. 97):

PAGE 142

134Katz and Kahn argued that organizations were fundamentally open systems in which people were bonded together by psychological constructions and symbolic as well as behavioral respon ses to their environm ents. Furthermore, organizations contain a great deal of variety and therefore uncertainty, owing to the individualistic nature of the humans within them. Variety and uncertainty are in turn balanced by certain control mechanisms (e.g., norms, rules, procedures, hierarchy) that subordinate individu al needs to the requirements of effective organizati on. Creating ways to integrate and coordinate these comp onents—while remaining open to the changing demands of the environment—re quired systems management by persons sensitive to the dynamic inte rdependencies at work. The goal was to ensure that the system remained op en and hence capable of adaptation. (Eisenberg & Goodall, 1993, p. 97) This discussion on “remaining open to changing demands of the environment” (open systems concepts) and “certain cont rol mechanisms” (mechanistic concepts) reveals the mix of sensemakin g structures that shaped both the investigation and recommendations during the 9/11 Public Hearings. It is useful to note the disconnect. On the one hand, the 9/11 Pu blic Hearing witnesses and commissioners took into account the variety and unce rtainty in the environment of the 21st ventury; on the other hand, they adopted a largely mechanistic framing of problems that, in the end, narratively reduced solutions to mechanistic responses that are better suited for stable, predicta ble, certain environments. Autopoiesis: An Alternative Descriptor for System Resilience and Accountability An alternative to mechanistic descriptions of organizations, autopoiesis offers powerful “images, metaphors, and ways of thinking” and talking about system failure/resilience and account ability (Wheatley, 1994, p. ix ). Resilience, defined in terms of autopoietic structures, is the syst em’s ability to create new knowledge and

PAGE 143

135reconfigure itself so it can deal with flux in the environment. Rather than preserve a given, fixed structure, the system adapts through continuous learning and change; that is, it creates, elaborat es, and changes structures in order to remain viable as ongoing systems (Buckley, 1967, p. 5). For this reason, autopoietic systems are frequently referred to as self-organizing or self-renewal system s (Wheatley, 1994, p. 88). Table 5, a summary of the theory on autopoiesis presented in Chapter One of this paper, presents ways in which the language of autopoiesis or self-organizing systems offers an alternative vocabulary to that of mechanism for theorizing organizational resilience.

PAGE 144

136Table 5 Building system resilience: Contributions of autopoiesis, a form of systems theory Autopoiesis, a new vocabulary for building system resilience: Unpredictability and flux in the environ ment are viewed as opportunities for system growth and resilience. Knowledge of the environm ent and the tools and processes for generating knowledge are resources fo r system development. Because people have some control over the tools and processes for generating knowledge, there are opportuni ties for shaping the future of the system. Key to shaping the future is ensuring the tools and processes for creating knowledge are as complex, dynamic, and diverse as its environment. Tools and processes for leveraging system resilience in dy namic environments include (a) a clear sense of identity that guides the action of its members and (b) processes that facilitate participation, the continuous generation of knowledge, and action. Resilient self-organizing sy stems survive by responding to new environmental conditions. This sort of agility, however, exists only when systems continuously generate information/communication/k nowledge about the ever-changing environment. It is important to note, howe ver, that the creation of new information, in the case of self-organ izing systems, is not a me chanistic transmission of information from one place to the next. Focusing on insights from connections of various people/expertise, advocates for self-organizing systems encourage a new form of exchange, interactions, relationships, engagement, and participation for organizations. The differen ce, according to Ruben (1979), is between an information system and a communication system:

PAGE 145

137 As long as the primary focus is upon the mechanical flow of messages . it would still be termed an information system. Only when the significances, competencies, purposes, fu nctions, and desires which [participants bring] to the situation were defined as a crucial as pect of the system, would it meet the definition of communi cation system. (p. 110) Table 6 presents assumptions from the theory of self-organizing autopoietic systems that are relevant to this project on system resilience, and their implications for organizational action.

PAGE 146

138Table 6 Autopoiesis: Assumption s and implications for organizational practices Autopoietic assumptions Implications for orga nizational practices There are “no pre-fixed, definitely describable” environments (Wheatley, 1994, p. x ). A system places a variety of responses into the environment and determines what work s. Through action, we develop not only our systems, but also our environment. Action over planning is preferred by advocates of autopoiesis (Weick, 1979; Wheatley, 1994, p. 37). An autopoietic perspe ctive assumes flux in the environment. Knowledge is provisional, as “everything is always new and different and unique” to us (Wheatley, 1994, p. 7). An autopoietic system fosters resilience and sustainability through its continuous generation of knowledge (Wheatley, 1994, p. 104). Autopoiesis assumes every act of observations loses more information than it obtains, as each interpretation of an event is specific to the observer (Wheatley, 1994, p. 65). The broader the distribution of information, viewpoints, and interpretations, the greater the opportunity to create new knowledge, resilience and sustai nability (Wheatley, 1994, p. 64). “Innovation is fostered by information gathered from new connections” (Wheatley, 1994, p. 113). Autopoiesis encourages relationships that foster the connection of people/information, e.g. inter-disciplinary relationships (Wheatley, 1994, p. 113).

PAGE 147

139Building System Resilience: An Autopoietic/Sensemaking Model Autopoiesis conceptualizes system resilience in terms of the system’s ability to represent its environment—that is, it s descriptions/communica tion/knowledge of the environment (Luhmann, 1990 ). As a consequence, this theory brings to our attention the importance of continuous learni ng for organizations that reside in everchanging environments. Similarly, the construct sensemaking furthers our understanding of system-environment relations by focusing on organizational activities at “moments of fl ux” (Weick et al., 2005, p. 413), i.e., “when the current state of the world is perceive d to be different from the expected state of the world” (Weick et al., p. 409). How organizations construct relations with the environment, why, and to what effect is the focus of sensemaking (Weick, 1995, p. 4). In an effort to guide research and practice in this area, Weick, Sutcliffe, and Obstfeld (2005) compressed the seven sensemaking activities (discussed in the second ch apter of this paper) into a sequence of activities that “mobilize around moments of flux” (p. 413). Depicted graphically in Figure 4, the ESR model conceptualizes system-environment relations, or resilience, as “reciprocal exchange s between actors (Enactment) and their environments (Ecological Change) that are made meaningful [by being Selected] and preserved (Retention)” (Weick, Su tcliffe, & Obstfeld, p. 414). Figure 4 Resilience, in terms of the ESR sensemaking sequence (Weick, Sutcliffe, & Obstfeld, 2005, p. 414) Source: adapted from Weick, Sutc liffe, and Obstfeld, 2005, p. 414.

PAGE 148

140Using the ESR sensemaking model, we can conceptualize the kinds of organizational practices that foster continuous learning, self organizing, and system resilience. Keeping in mind the difficulty of translating open systems concepts to organizational practices, this section focuses on brin ging together two theoretical concepts, self reference and ambivalence from autopoiesis and sensemaking for the purpose of identifying the kinds of orga nizational practices that fo ster system resiliency. Self Reference A fundamental principle of all self organizing principles is self-reference (Wheatley, 1994, p. 94): In response to environmental disturba nces that signal the need for change, the system changes in a way that remains consistent with itself in that environment. The system is autopoietic, focusing its activities on what is required to maintain its own integrity and self-renewal. As it changes, it does so by referring to itself; whatever future form it takes will be consistent with its already established identity. Changes do not occur randomly, in any direction. They always are consistent with what has gone on before, with the history and identity of the system. . When the environment demands a ne w response, the [system] is a reference point for change. (Wheatley, 1994, p. 94) Examples of points of reference available to systems when the environment signals a need for change include the system’s values, vision, culture, competencies, procedures, processes, and practices. Each, according to Weick (1979) is a “historical document,” stored as part of the system’s memory or retention processes that serve as plausible guides for subs equent activities (p. 229). Figure 4 graphically portrays the acti vities of selection, enactment, and retention and the role these activities play in establishing points of self-reference as guides for subsequent activities. During the process of selection, a number of

PAGE 149

141possible meanings for equivocality in th e environment are reduced to plausible accounts of the events. It is important to note that these accounts are simultaneously the adoption of a descripti on of the environment and a loss of other possible descriptions. Stored as “his torical” documents (Weick, 1979, p. 229), selected descriptions are available to guide future descriptions of unexpected flux in the environment. Similarly, previous actions by the sy stem are available to guide current and future responses to eq uivocality in the environment. Ambivalence Self-reference, however, does not mean that members of an organization must be “victimized by . their paro chial volition” (Weick, 1979, p. 177). Experience retained as memory can serve as a constraint on the ways in which new experiences are interpreted and responded to or experience can serve as a point for interjecting “chronic doubt” (Weick, 1979, p. 177) or ambivalence (Weick, Sutcliffe, & Obstfeld, 2005, p. 414) about previous interpretations and responses. For example, in contrast to mechanistic views of organizing, agents in self-organizing systems are continuous learners whose le ssons from past experience are always open to change. In fact, this ambivalent use of previous knowledge is the crux of autopoietic systems; that is, ambivale nce distinguishes open, deve loping autopoietic systems from mechanistic systems that are closed and static. To avoid stasis, entropy and death, systems must be able to view the system’s history/memory ambivalently, both believing and doubting it in future enac ting and selecting (Wei ck et al., 2005, p. 414): Only with ambivalent use of previous knowledge are systems able both to benefit from lessons learne d and to update either thei r actions or meanings in ways that adapt to changes in the sy stem and its context. (Weick, Sutcliffe, & Obstfeld, 2005, p. 414)

PAGE 150

142Self-reference and Ambivalence: The Case of 9/11 Organizations that assume a stable en vironment tend to be preoccupied with the past, following prescribed procedures and improving standardized processes and procedures when errors occur. In short, history is believed to be the best lesson. This tendency to prepare for the future by reactively responding to past failings was an expressed concern of witnesses in the 9/11 Public Hearings. According to the witnesses, this tendency is a “recurring theme in the governme nt,” both preand post-9/11 (Push, 1.1.171). For example, when al Qa eda mounted land suicide attacks against the U.S. embassies in Kenya and Tanzania in August 1998, Americans increased the perimeter security of their land targets. When al Qaeda attacked a maritime target, the U.S.S. Cole in 2000, Americans increased the perimeter security of their maritime targets. But then al Qaeda again fooled the United States and conducted an airborne suicide attack (Gunaratna, 3.3). Additionally, it was noted during th e Public Hearings that the pre-9/11 aviation security model had been based on lessons learned from previous disasters; for example, screening check point security wa s the direct result of aircraft hijacking worldwide during the 60s and 70s, airport access controls were strengthened after the crash of Pacific Southw est Airlines flight 1777 in 1987, and checked baggage security was strengthened during the 90s after the bombing of Pan Am 103 (Mead, 2.1.92). Similarly, post-9/11 counterterrorism in itiatives continued to focus on lessons learned. Examples provided by witnesses included the ai rport checks on everybody’s shoes after Richard Reid boarded an air line with a shoe bomb (Push, 1.1.171; Thompson, 1.2.82), and the State Depart ment’s temporary de nial of visas, immediately after 9/11, to students wantin g to study flying in the United States (Dillingham, 1.2.37). Consistent with the creativity of sensemaking, witnesses in the post-9/11

PAGE 151

143public hearings looked for alternatives to “closing the barn door after the horse is out” (Cleland, 2.1.104). These included looking to the future as well as understanding our historic shortcomings “to better order our future steps in security” (Ranstorp, 1.1.256; s ee also Corzine, 2.1.62). Post-9/11 we find a greater openness for seeking an understanding of, and responding to, the complex dy namic environment of the 21st century. From the 9/11 Public Hearing Transcripts, for example, there is an expressed interest in achieving a better understanding of: 1. The psychological makeup of the te rrorists, their decision making procedures, and how they identify ta rgets and attack modes (Ranstorp, 1.1.264-265). 2. How terrorism is structuring, changi ng, and adapting (Ranstorp, 1.1.263); what drives and motivates it; the source of its power; the resources at its command; its internal strengths and we aknesses; and the identity, roles, and motives of its allies, enablers, and supporters (Kean, 3.1). 3. The web of relationships in the Muslim world (Fandy, 3.5). 4. The context and forces that make al Qa eda appealing to its recruits in the Muslim world (Fandy, 3.5). The 9/11 Public Hearing Transcripts suggest that there is a growing recognition of the need for a new perspective and language for building system resilience, one that links knowledge, environment, innovation, and accountability. While autopoiesis and sensemaking offer theoretical concepts (e.g., self reference and ambivalence) for understanding the dy namics of stability and innovation in dynamic complex environments the question remains, How do we move from theory to practice?

PAGE 152

144An Autopoietic/Sensemaking Model Figure 5 graphically portrays the ways in which both autopoiesis and sensemaking can inform organizational pr actice. This model presents the ESR Sensemaking Sequence of en actment-selection-retentio n (Weick et al., 2005, p. 414) as activities that consti tute the system’s ability to represent the environment. Additionally, it expands the ESR model to in clude the concept of ambivalence. Along with their graphic depictio n of the ESR sequence, Weick, Sutcliffe and Obstfeld (2005) also identified the critical role that ambivalence plays if systems are to have meaningful exchanges with their environments. That is “These exchanges will continue only if the preserved content is both believed . and doubted” (2005, p. 414). However, this critical activity was no t visually depicted in the Weick et al. (2005) model. Additionally, Figure 5 specif ies the kinds of retain ed plausible stories that inform, according to an autopoietic vi ew of organizations, system-environment activities—that is, the system’s plausible stories of previous system-environment experiences and stories on how members of the system engage wi th one another as well as with the environment (Wheatley, 1994, p. 91). A summary of the concepts from autopoiesis and sensemaking that have informed this model are presented in Table 7.

PAGE 153

145Figure 5 Building system resilience: An autopoietic/sensemaking model Source: adapted from Weick, Sutc liffe, and Obstfeld, 2005, p. 414. Table 7 Building system resilience: An autopoietic/sensemaking model Figure 5 also suggests ways in which le aders can foster a coordinated systeminformed response to the environment. The system’s knowledge (i.e., its previous experiences) is the history that informs the system’s future. However, a selforganizing system assumes that every experience is unique (Wheatley, 1994, p. 38). Autopoietic/sensemaking assumptions The future form of an autopoietic system will be consistent with its previous history. How selections and enactments are ch osen can be both an acceptance and rejection of the system’s past or conventional way of seeing things. The system’s ability to be innovative can exist only if the organization has access to new information about external conditions and internal resources (Wheatley, 1994, p. 91). The knowledge that is generated and th e tools and processes that generate new knowledge are viewed as resource s for building system resilience.

PAGE 154

146Therefore, self reference as a guide for future action is most effective when it is coupled with an ambivalence about the usefulness of previous experiences and the ways in which new knowledge is continuously generated. Given the critical roles that ambivalence and new knowledge play in th e responsiveness and sustainability of organizations, the development of these sensemaking processes and structures are the kinds of contributions for which le aders should be held accountable. Practical Implications for Futu re Organizational Practice An organizational sensemaking model (e.g. ESR) is a composite of multiple sensemaking processes that occur in organiza tions. For example, the repertoire that informs how we participate in organizational life includes the knowledge, skills, abilities, values, identity, etc. that I bring to the position, as well as the organizational premises, pl ans, expectations that I find in the workplace. Additionally, my ongoing performance in th e organization, also retained in memory, may influence how I interpret and respond in the future. When we recognize that organizational sensemaking is the compos ite of an array of diverse sensemaking processes that individuals engage as they go about their work, we realize the multiplicity of individual frameworks, as well as organizational frameworks, that inform individual and organi zational performance. With this in mind, we cannot escape the critical role of leaders as sens emakers within the syst em whose task is to provide their members with a self-referenti al system of organi zational enactments for making sense of, and responding to, changes in the environment. In this capacity, leaders are uniquely positioned to significantly influence the contributions of individual members to organizational life. An overview of the following discussion on the ways in which leaders can uniquely influence system resilience and accountability is presented in Table 8.

PAGE 155

147Table 8 Building system resilience: Leadership Leadership Sensemaking Roles Leadership Practices Building system consciousness Developing key princi ples that express the overall identity of the system Saturating the organization with consistent messages about the organization, what it is, what it stands for Designing heedful interrelating Engaging a greater number and diversity of people Loosening controls on engagement, participation, and action Legitimizing ambi valence in the workplace Admitting, “We don’t know.” Asking, “What is new?” The Role of Leaders: Building System Consciousness Building system consciousness Developing key principles that express the overall identity of the system Saturating the organization with consistent messages about the organization, what it is, what it stands for In mechanistic organizations, leader s attempt to control performance by standardizing organizational inputs and outputs, such as communication. For

PAGE 156

148example, at a recent planni ng meeting attended by mi ddle management, the desired outcome for the meeting was a process for improving communication from the executive vice president (EVP) to the staff wh o report directly to her. Agenda items for accomplishing this task included: What information do we want the executive vice president to share with us, how often should she communicate this information, and to whom should this information be disseminated? The problem was identified in previous meetings as a need for the various administrative units to work as a team. After multiple meetings, the issue was fr amed as a need for better communication from the EVP, and the solution was to be found in established processes for channeling particular kinds of information from the EVP to subordinates. This example serves as a prototype of how proc esses, procedures, and practices are often accomplished in mechanistically-driven organizations. A problem is identified, solutions are generated, and rules are esta blished for the purpose of guiding future actions. Organizations that are viewed as auto poietic processes, in contrast to the above approach, foster organizational co herence by using a few key patterns or principles to guide organization al activity. To this end, leaders of these organizations are accountable for creating an organiza tional environment where ideas connect, cohere, and inform members’ actions. The go al is system consciousness; that is, throughout the system, ideas guide individuals’ interpreta tions of, and responses to, the environment. Leadership practices that foster sy stem consciousness include developing, and saturating the organization with, pr inciples that can guide organizational activity. Leadership Practice: Developing Key Princi ples that Express the Overall Identity of the System These principles can be articulated in the form of inspirin g visions, strong values, organizational beli efs, and concepts. For ex ample, the mission of an

PAGE 157

149academic institution stated in terms of teaching, research, and service qualifies as an inspiring vision. On both the academic and administrative sides of educational institutions, we often find faculty and staff who have joined the institution for this very reason. They want to be part of something worthwhile, something they believe in. In practice, however, this mission seldom guides the whole institution and is often adopted and perpetuated only by the academic leadership. On the nonacademic side, each vice president typica lly has a mission statement for her area, and within each vice pres idential area, divisions of ten have their own mission statements. For example, th e mission of a Division of Physical Plant is to: Maintain and improve a physical environment conducive to teaching, learning, and research. To this end, the Physical Plant Division is responsible for the operation, maintenance and repair of the educational buildings, utilities, and rounds of the University campus. And the mission of a Division of Environmental Health and Safety is to promote and assist in the achievement of a safe and healthful University environment. To whatever extent an employee of either of these divisions migh t find the missions exciting and inspiring, the fact remains that the university is fragmented into competing units to which individuals identify and commit. In organizations guided by the spirit of autopoiesis, leaders are accoun table for constructing ideas that connect, rather than perpetuate, discrete an d distant actions and events. Leadership Practice: Saturating the Orga nization with Consistent Messages about the Organization In organizations that are guided by the spirit of autopoiesis, information is treated as “the primal energy that struct ures matter into form” (Wheatley, 1994, p. xi). In these organizations, leaders are responsible not only for creating but also for saturating the system with clear messages about the organization—i.e., what it is and what it stands for—to serve as self-r eferential resources th at guide current and

PAGE 158

150future actions. Recognizing the array of diverse sensemaking frameworks that individuals bring to the framework, not to mention competing sensemaking frameworks that can be found in organizati ons, it is important for these messages to be ever present as a resource in the work place. This requires a presence greater than the kind that can be found by hanging vision, mission, and value statements on the wall, i.e., it requires an engaged, participating manager such as the one described in the following excerpt from the 9/11 Public Hearing Transcripts: I’ve been working for George Tenet [Director, Central Intelligence Agency] in the last month and a half, I’ve gone to his 5:30 meeting which is a meeting of George and his principle officers that are involved in clandestine services . . That meeting, I would guess, is the closest you can come to an operational meeting where somebody says I want to do the following, go do it, talk to somebody in this country and tell them not to do this tonight, because we’re worried about it, do something else. It is an operational role . that kind of operational level detail. Now I don’t know how to replicate that, and at what level you do it, but I know it needs to be done. And quite honestly, I’ve talked to some people about this and they’d say, we ll, you talk about micromanagement, that’s the ultimate in micromanagement. Now, can someone else do that? It ’s not obvious to me that someone more distant than George Tenet from the people who do real stuff, who send real messages . if you get much fa rther away from that ability I think you’re in a fair amount of trouble . . It does sound like micromanagement to me but from my perspective that’s really damn good micromanagement, that’s doing the hard work at the right place . . Now, how do you institutionalize that and make that a part of the

PAGE 159

151system? (Kerr, 4.62-63) Although Thompson can’t determine wh ether Tenet is a micro-manager, his description of Tenet’s hands-on, engaged management style su ggests a process for building system consciousness that, if replicated system wide, can cohere discrete and distant activities in th e system. Tulgan (2004) agrees. Hands-on management is not the same as micro-management: There has been so much emphasis on empowerment in recent years, and fear of micro-management, that managers ha ve moved too far afield of providing direction. If you want to empower dire ct reports, you must define the terrain in which they have power. (Tulgan, 2004, p. 7). While the articulation of these underlying principles is important, Wheatley (1994) warns, “It’s only half the task” (p. 55): Creating the field through the dissemination of those ideas is essential. The field must reach all corners of the or ganization, involve everyone, and be available to everywhere. . We need to imagine [leaders] as broadcasters, tall radio beacons of information, pulsing out messages everywhere . stating, clarifying, discussi ng, modeling, filling all of space with the messages we care about. If we do that, fields develop—and with th em, their wondrous capacity to bring energy into form. (pp. 55-56). Hands-on leaders in organizations guided by the spirit of autopoiesis engage in saturating the organization with self-referential information about its identity, for the purpose of guiding and cohering the variou s activities throughout the system. The activity of these sensemakers, resemb ling broadcasters, pulse out messages everywhere.

PAGE 160

152 The Role of Leaders: Designing Heedful Interrelating Designing heedful interrelating Engaging a greater number and diversity of people Loosening controls on engagement, participation, and action In mechanistic organizations, lead ers attempt to harness variety and complexity by regulating inputs from the environment and following standardized processes. This can be seen in the narrow de finition of national security intelligence that was in place before September 11, 2001. National security intelligence, according to various participants in the 9/11 Public Hearing Transcripts, was the information collected, analyzed, and disse minated by the Intelligence Community. Information that did not originate from formal intelligence production processes did not qualify as actionable intelligence. Th is approach fragments information, drawing lines between what (and who) informs the system. Similarly, with its focus on standardization, mechanistic organization s tend to regulate, standardize, and homogenize the training and development of staff—that is, people develop the knowledge, skills, and abilities needed to work within the current system. In contrast to mechanistic organization s, organizations that are guided by the spirit of autopoiesis respond to environmental complexity by developing the variety and complexity of the system’s knowle dge base. No one has a more sweeping opportunity to accomplish this than leaders. That is, leaders not only have a unique opportunity to saturate the organization wi th principles that cohere and coordinate discrete and distant actions, they also are uniquely positi oned to develop the people, processes and structures in ways that develop complex responses to complex environments. The goal is what Weick and Roberts (1993) call heedful interrelating. The concept of heedfulness designates behaviors su ggestive of “’noticing,

PAGE 161

153taking care, attending, applying one’s mind, concentrating, putting one’s heart into something, thinking what one is doing, alertness, intentness, studying and trying’” (Ryle, 1949, p. 136; see also Weick & Robert s, 1993, p. 361). It is about seeing, taking note of, being attentive to, competen ce, and know how, and requires a kind of learning and training that is not of ten found in mechanistically-structured organizations: Heedful performance is not the same thing as habitual performance. In habitual action, each performance is a replica of its predecessor, whereas in heedful performance, each action is mo dified by its predecessor (Ryle, 1949: 42). In heedful performanc e, the agent is still learni ng. Furthermore, heedful performance is the outcome of training and experience th at weave together thinking, feeling, and willing. Habitual performance is the outcome of drill and repetition. (Weick & Roberts, 1993, p. 362) Heedfulness, according to Weick and Robert s (1993) is “a kind of capacity in an ongoing activity stream [that] emerges in the style with which activities are interrelated” (p. 365). The unique opportunity for leaders is the actualization of this capacity by developing both the “know how” that individual s contribute to the system and the patterns of interrelated activities among members of the system. Discussion of the kinds of leadership practices th at enable creative, responsive, heedful responses to emergent environmental demands follows. Leadership Practice: Engaging a Grea ter Number and Diversity of People The ESR sensemaking sequence, a mode l that continues to inform this discussion on implications for future organi zational practices, defines the system’s relationship to the environment in terms of “reciprocal exchanges between actors (Enactment) and their enviro nments (Ecological Change)” (Weick et al., 2005, p. 414). With this model, we can visualize ho w reality is revealed “through an active construction in which we pa rticipate” (Prigogine & Stengers, 1984, p. 293; see also

PAGE 162

154Wheatley, 1994, p. 65). A dditionally, taking into account Buckley’s (1967) observation that “only variety can regulate variety “ (p. 495), the relationship between the numbers and diversity of indivi duals engaged in the interpretation of, and response to, the environment is made clear. According to Wheatley (1994), the environment is a potential relationship that can be realized by way of participation, and the more participants we engage in th is relationship, the greater the opportunity for evoking multiple interpretations and building system heedfulness (p. 64) This was a dominant observation also made by witnesses during the 9/11 Public Hearings. At the core of a call fo r “a systematic rethinking about the way we do business” was the realization that the kind of counterterrorism that we are facing in the 21st century is a task that is much broa der than any nation al security issue previously experienced by the United States (Steinberg, 4.34). As a result, it became clear to a number of witnesses that the informatio n and expertise needed to fight the war against terrorism was “not going to be picked up in traditional collection systems” (Gannon, 4.52). To this end, a number of witnesses called for “an approach that reflects this far flung, high ly decentralized universe of critical actors” (Steinberg, 4.34). Suggestions for being more inclusive in the engagement against terrorism included unprecedented in volvement from the following: 1. Local level police departments (R oemer & Jenkins, 1.1.302-303). 2. Front line employees who need to us e information to screen people who come to the United States (Fine, 1. 2.13) and need to use initiative and their brains to do the job they ar e trying to do (Dzakovic, 2.1.139). 3. Linguists (Harman, 2.1.41). 4. Individuals with the wo rst kinds of backgrounds, the worst kinds of criminal and humanitari an deficiencies (Deutc h, 4.33), bad guys who know a lot about what is going on in their respective communities (Thompson, 6.18).

PAGE 163

1555. The private sector (Brill, 1.2.126). 6. Academia (Jouejati, 3.58). 7. The public (Gunaratna, 3.25). System resilience, according to 9/11 Public Hearing witnesses is not just a task of government. Participation “seriously done” (Wheatley, 1994, p. 64) provides the multiplicity of viewpoints, perspectives, an d interpretations that are needed to make sense of the world and build system heedfulness. Design ing structures that foster inclusiveness, participation, and diverse contributions is an opportunity uniquely held by leaders and one for which leadership should be held accountable. Leadership Practice: Loosening Controls on Engagement, Participation, and Action Mechanistic organizations draw boundaries everywhere. Examples include boxes on organizational ch arts, chains of command, job titles, and position descriptions. Each of these distinctions aims to limit individuals’ participation, responsibility, authority, and accountability. With each designation, workers are further fragmented. The accountant, for example, is to bring his analytic mind to his work but, for the most part, his emotions and intuitions are to be restrained, if present at all, in the workplace. Those who step out of their designated places risk a range of disciplinary actions, includin g reprimands, loss of merit increases or opportunities for promotion, and dismissal. This select ive recruitment of parts of human beings to perform clea rly specified tasks robotizes the work place. In this distinctly anti-human context (Senge, 1990; Wheatley, 1994), it is not unusual or surprising that individuals are unwilling to look beyond the performance of their own jobs, to the survival and well bein g of the whole organization. The survival and growth of organization s guided by the spirit of autopoiesis, in contrast to mechanistic organizations, depend on increased levels of participation from individual system members. The conc ept “Think globally; act locally” captures the essence of these organizati ons. Not to be confused with previous misconceptions

PAGE 164

156of empowerment (i.e., indivi dual workers were to be set loose to do what they thought was best for the customer or the organization), these organizations build resilience by: 1. Encouraging autonomy, constrained on ly by the guiding principles or identity of the organization. 2. Holding individuals accountable for the survival and viability of the organization. In line with the holistic perspective that system thinking offers, the focus is on honoring, appreciating, and engaging th e totality of the individuals. In this sort of individual-organization relationship, leaders foster ownership. By inviting the whole indivi dual to engage in circumst ances greater than the day-today routine tasks of a job, leaders foster personal emotional attachments to the well being of the whole organization. This ki nd of participation and ownership can be seen in the performance of two witnesses who appeared before the 9/11 Public Hearing. Mr. Melendez Perez, the immigr ation inspector at the Orlando airport who persistently questioned a Saudi traveler despite the possibility of reprimands and disciplinary action, may have prevented a tw entieth 9/11 hijacker from entering the country. And General Arnold, risking court martial, scrambled aircraft on September 11 without obtaining required clearances. In each case, the heroism of these individuals on 9/11 looks different from the heroes of a mechanistic system where heroism is defined in terms of following the rules, and individuals such as Mr. Melendez Perez and General Arnold would be perceived as, if not reprimanded for, being loose, uncontrollable cannons. Not only do mechanistic organizations restrict the kinds of knowledge, skills, and talents that individuals are to contribute to the organization, they also keep a tight rein on interactions among individuals. This can be seen clearly in organizations with silos and rigid chains of command. For example, protocols in

PAGE 165

157some 21st century organizations require Empl oyee A to transmit a message to Employee B by having it travel “up” A’s chain of command and then down B’s chain of command, until it reaches Employee B. In organizations guided by the spirit of autopoiesis, the issues are not controls on the contributions of its memb ers and flows of information, but dynamic connectedness (Wheatley, 1994, p. 23). With its focus on continuous learning and change, these organizations stri ve to make connections that previously did not exist. In language frequently used in the 9/11 Public Hearing Transcripts, these organizations strive to connect the dots (Armitage, 8.2.185; Ashcroft, 10.1.142, 166, 166; Baird, 1.2.146; Ben-Veniste, 11.1.146, 12.1.114; Clark, 8.2.121; Cleland, 2.1.18; Fetchet, 1.1.218; Fi elding, 10.1.19; Gorton, 4.89; Hamilton, 8.2.189, 199; Kerr, 4.61; Kerrey, 11.1.147, 11.2.108; Lehman, 12.1.117, 12.2.48; Lieberman, 2.1.15; Reno, 10.1.62; Rice, 9.20; Ro emer, 10.1.110; Rumsfeld, 8.1.133). However, in the case of autopoietic system s, the focus is not on collecting sufficient dots of data from the environment in order to act with certainty. The dots to be connected are the dots of information held by individuals within the various parts of the system. “These unseen connections between what were prev iously thought to be separate entities,” accordin g to Wheatley (1994) “are th e fundamental elements” of new knowledge (p. 10). As a result, with circular flows of information throughout the system, systems can connect various understan dings of small fluctuations that vary from the norm (Wheatley, 1994, p. 19) an d increase the number and complexity of its interpretations of the environment. Th ese circular flows of information can be likened to a train on a toy railroad, circling through its recurring journey (Senge, 1990, p. 76). Coursing through th e system in recurring loops, the sense that has been made of bits or dots of informat ion can become more complex. That is, sensemaking “grows in strength ” (Wheatley, 1994, p. 19) as it continuously informs, and is informed by, those who are participat ing in the conversation. These circular

PAGE 166

158flows of information, according to Senge ( 1990), are “reciprocal flows of influence” (p. 75). In this way, systems can co nnect dots and increa se the variety and complexity of its knowledge base. That is, they can build “the variety that regulates the variety in the environment” (Buckley, 1967, p. 495)—that is, the variety that is fundamental to system heedfulness, resiliency, and viability. Such opportunities can be found in the establishment of interdisci plinary programs, cross-functional teams, task forces, and cross training. Fundamen tal to the success of these efforts is the organization’s ability to fully engage its workers. Also fundamental to the development of the system’s knowledge base is the development of its members. Foremost among the knowledge, skills, and abilities that are needed for autopoietic processe s are: ongoing development of general knowledge and skills, in contrast to the precise knowledge or skills that are needed for one’s current position; the development of a system consciousness, one that focuses on the whole of the organization rather than the innumerable fragments that we have been enculturated to beli eve and see in our or ganizations; and the development of skills (e.g., facilitating gr oups and listening to others) that build strong relationships and foster lear ning (Senge, 1990; Wheatley, 1994). Organizational leaders hold a unique opportunity for developing people, processes and structures that develop system-wide heedful interrelating. Opportunities for building heedful interrelating can be found in: 1. Designing organizations that more fully engage the knowledge, skills, and abilities that individuals bring to their positions. 2. Creating opportunities for circular flows of information through the organization so that its members ca n continuously contribute to the variety and complexity of the system’s stock of knowledge (Berger & Luckmann, 1966) and learn from others.

PAGE 167

1593. Expanding the potential of the organi zation by expand ing the knowledge and skill base of its members beyond the requirements of their current positions. The Role of Leaders: Legitimizi ng Ambivalence in the Workplace Legitimizing ambivale nce in the workplace Admitting, “We don’t know.” Asking, “What is new?” The two previous sections have fo cused on more commo nly discussed topics in organizational develo pment, including providin g guiding principles for organizational action, fostering inclusiveness and diversity, and more fully engaging workers in the workplace. The focus of these discussions has been a realignment of these frequently discussed practices to the newer view of organizations as autopoietic systems. This se ction continues to explore the application of autopoietic concepts to organizational practices. In th is case, the focus is not on re-defining old labels, but on introducing a concept that is, for the most part, novel to current organizational prac titioners. This is the concept of “legitimizing doubt” (Weick, 2001, pp. 91-102) or ambivalence in the work place (Weick et al., 2005, p. 414). Sensemaking models, such as the ES R Sensemaking Sequence, highlight the significance of ambivalence for systems whose environm ents are dynamic, complex and ever changing. What these models reve al are the potential for system learning and autonomy that occurs during system-e nvironment exchanges. These exchanges are made meaningful by the system when it retrospectively notices, brackets, labels, and categorizes these experiences, using ment al models acquired over time to make sense of the experiences. However, only with ambivalent use of previous knowledge are systems able both “to benefit from le ssons learned and to update either their actions or meanings in ways that adapt to changes in the system and its context”

PAGE 168

160(Weick, Sutcliffe, & Obstfeld, 2005, p. 414). For systems that reside in complex environments, ambivalence is fundamental to continued learning, resilience, and viability. Mechanistic organizations view the wo rld as stable and knowable. Thus, in these sorts of organizations, we find rout ines, standardized practices, and searches for actionable intelligence—th at is, a search for sufficien t amounts and specificity of information in order to act with certainty, accuracy, and effici ency. In contrast, organizations guided by the spirit of autopoiesis view the world as unfolding events that require an ongoing updating of the sy stem’s experiences wi th the environment. In this updating of its expe riences, the system builds th e variety and complexity of its stock of knowledge. The focus of thes e organizations is not the day-to-day application of what is known, but an ambivalent use of its stock of knowledge, honoring both what is known and what must also be created. This is the crux that distinguishes mechanistic an d autopoietic views of orga nizations. One assumes a stable environment, and relies, for the most part, on what is already known, and the other sets forth as a priority the ongoin g development of plausible explanations (stories) of its environment. In autopoietic organization s, “Newness [is played up]. . What’s new is the context . . What’s new is a premium on updating [our explanations or stories of the system’s context or environment]” (Wei ck, 2001, p. 94). Through circular flows of information, previous experiences wi th the environment ar e validated, and the organization benefits from le ssons learned, or previous experiences are reworked or updated “in relation to unanticipated ideas that are conceived, shaped, and transformed under the special conditions of a current performance” (Weick, 2001, p. 97). This flexible treatment of previous experiences is about “making something [new] out of previous experi ence, practice and knowledge ” (Weick, 2001, p. 97), and can only be accomplished through ambivale nt treatment of current experiences.

PAGE 169

161 The variety and complexity of stories th at are available to the system are its resources for managing flux in the enviro nment, resilience and viability. The challenge, however, is this new way of working for workers who have been enculturated in the mechanistic wellspring of American culture. In mechanistic organizations, managers think and wo rkers do. According to Tannen (1998), Americans regard doubt as “synonymous with intellectual inquiry, a sign of intelligence” (p. 273). However, rigid roles in the workplace “disable” and “cow” many; paralyzed by roles, many do no participate; fearful, many offer only the meekest agreement (Senge, 1999, xviii-xix). As a result, a role for which organization al leaders should be held accountable is establishing ambivalence as a system-w ide priority and a new way of working in the complex, dynamic environment of the 21st century. The goal, in the words of Weick et al. (2005), is system-wide “actin g thinkingly” (p. 412); that is, members simultaneously use trusted frameworks to interpret experience, “yet mistrust those very same frameworks by testing new fr ameworks and new interpretations” (p. 412). By “legitimizing doubt,” or circulat ing an attitude of ambivalence in the workplace (Weick, 2001, pp. 91-102), lead ers position organi zations for ongoing sensemaking, in contrast to the set-in -stone decision ma king that typifies mechanistic organizations. Leadership practices that foster system-wide ambivalence, sensemaking, creative ness, and resilien ce include: 1. A mixture of puzzlement, ambivalence, and honesty. Autopoiesis assumes a world that is unknowable and unpredictable. In this kind of world, a world that unfolds, leaders approach life “from a creative as opposed to reactive view” (Senge, 1990, p. 141). 2. Admitting when they don’t know. Ta lking about ambivalence “may open people’s minds somewhat, but actions always speak louder than words” (Senge, 1990, p. 173). There’s nothin g more powerful that leaders can

PAGE 170

162do to legitimize doubt, foster acting thinkingly, and nurture ambivalence than admitting, “We don’t know.” 3. Saturating the organization with the question What is new? by articulating it as a guiding principle for interpreting and responding to the environment; designing processes that engage a diversity of perspectives; and developing people on this new wa y of interpreting and responding to the environment. Discussion and Conclusions The terrorist attacks on Septembe r 11, 2001, and other major human-made national disasters (for example, the Columbia and Challenger accidents) have been attributed to institutional failures, with no explanation of the ways in which individuals were resp onsible and accountable for thes e events. This approach is easier to understand when we take into account the mechanistic view that informs every aspect of our lives, including how we think about and manage organizations. Mechanistic organizations standardize the way in which individuals are to contribute on a day-to-day basis. Through practices such as policies, procedures, position descriptions, and disciplinary and reward programs, behavior is habitualized, thickened, and hardened (Berger & Luck mann, 1966, p. 59), attaining such a firmness in practice and consci ousness that it is assumed that this is the only way things can be done. This way of doing things “becomes real in an ever more massive way and it can no longer be changed so readily” by individual s (Berger & Luckmann, 1966, p. 59). In our institutions, a world of habituated thoughts and actions, it appears that only the massive workings of the institution can be held accountable for failures when they occur. Mechanistic practices, however, are be ing challenged. Assuming they reside in stable environments that have minimal impact on day-to-day organizational operations, mechanistically-structured organizations have institutionalized routine

PAGE 171

163practices as a way of fosterin g organizational stability and viability. Increasingly, however, there is a recognition that techno logy is creating dynamic, ever-changing organizational contexts that cannot be su fficiently theorized, or responded to, by mechanism. With Bertalanffy’s application of an op en system’s approach to organizations in the mid-1950s, many developments in orga nizational theory have been built on the principle that organizati ons, like organisms, must ac hieve appropriate relations with their environments if they are to survive. To date, however, many of the advances have occurred in the theorizing of organizations as autopoietic systems, while practical applications of open system s theory continue to focus on mechanistic controls. We find this, for example, in the 9/11 Public Hearing Transcripts. The two dominant recommendations by witnesses for deterring future terrorist attacks on the U.S. homeland focused on controlling perf ormance by reorganizing the Intelligence Community and implementi ng a rigid performance management system. Advocates of autopoiesis encourage a di fferent view of the world, one that recognizes that organization al resilience requires not on ly a view of the dynamic environments within which 21st century organizations resi de, but also a new way of working within this context. Suggestions that align practices with a view of organizations as autopoieticall y-structured systems include: 1. Saturating the organization with self-referential principles to guide action that supports the viability of the systems identity and integrity. 2. Loosening the controls on engagement, participation, and action and holding individuals accountable for viability of the system. 3. Fostering ambivalence in the workplace. These suggestions, moreover, are no t new concepts to organizational practitioners who follow organizational theory. Efforts to implement these suggestions in organizations include the development of mission vision, and value

PAGE 172

164statements; the implementation of cross tr aining and job rotations; and the use of task forces. But the temptation to control people and activities is difficult to resist. For example, in one organization where a cross-functional team was established to implement an institution-wide financial system, members were assigned full-time to the project but were given du al reporting relationships to both the project manager and the director of their fu nctional areas. These repo rting relationships made it difficult throughout the life of the project for members to support decisions that were in the best interest of the institution, rather than their functional areas. This spotlights the current challenge of 21st century organizations. To tap into the potential that autopoiesis offers organization al practitioners, new views of the world, organizations, and leadership roles and responsibilities are needed. Given the fear of relinquishing control, a fundamental concept of mechanism, it is unlikely that managers are willing to convert immediately to a new way of managing that involves relinquishing familiar controls. However, advances in building organizational resilien ce can be achieved by focusing on lessening the controls that dominate current practices an d infiltrating current practices with those that are informed by autopoiesis. This can be done by holding leaders accountable for identifying and realizing opportunities for developing system-wide consciousness, loosening controls, and fostering ambivale nce. Questions that can be used by leaders to infiltrate organizations wi th autopoietic practices include: 1. How, and where, can I make ever-prese nt a few self-refer ential principles for guiding individual actions? 2. Where can I pilot a program that broadens the knowledge base and participation of organizational members? 3. Where can I facilitate circular flows of information that allow for updating of experiences and diversity of perspectives?

PAGE 173

1654. On what projects, with what teams, ca n we instill ambivalence as a part of our processes? Are we always asking: What don’t we know? What mental models are constraining our perspective? What other mental models can we apply to this experience? What is new? What is unusual? What seems unimportant? Wh at doesn’t feel right? Leaders, actively engaged in the creation of system-wide con sciousness, “heedful interrelating” (Weick & Roberts, 1993, pp 357-381) and “acting thinkingly” (Weick et al., 2005, p. 412) ca n leverage efforts to build system-wide accountability and organizational resiliency. A challenge for leaders, however, is to step outside their current sensemaking processes and structures. One way to ad dress this challenge is to interject alternative perspectives and language into the organization’s sensemaking processes. For example, mechanistic organi zational structures such as functional silos and rigid performance systems are constructed in feminist theories as masculine structures that serve to homogeni ze the workforce, centralize control and power, and constrain the participation and contributions of many of its employees. Injecting the perspective and voice of femi nist theory into sensemaking processes can foster an ambivalent view of our cu rrent interpretations and responses to the environment, as well as our processes for generating this stock of knowledge. Diverse perspectives, when they are pa rt of an organization’s sensemaking processes, create opportunities to build the variety of knowledge, skills, and abilities that is requisite for systems that reside in complex, dy namic environments. It is important to keep in mind, however, that the autopoietic sensemaking model presented in this dissertation is not a transmission model of communication. Therefore, the intent is not a one-way street of communication where various theories inform organizational practice. True to the theory of autopoiesis, the intent is to inject the voice and perspective of scholarship (e.g., feminism, autopoiesis) into

PAGE 174

166organizational sensemaking pr actices where, through circul ar flows of communication and heedful inter-relating, each informs and is informed by the other. For example, research on autopoietic syst ems from the field of biology informs, and is informed by, the sensemaking practices of organizations guided by the spirit of autopoiesis. In this way, learning opportunities are cr eated that enrich indi viduals, the system, and the scholarship informing them. Creating a complex, dynamic, intelligent system, for the purpose of bu ilding system-wide accountability and resilience, is the unique responsibility of lead ers. For this, leaders should be held accountable.

PAGE 175

167References Albert, S., & Whetten, S. ( 1985). Organizational identity In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (pp. 263-295). Greenwich, CT: JAI Press. Asch, S. E. (1952). Social psychology. Englewood Cliffs, NJ: Prentice Hall. Axley, S. R. (1984). Manageri al and organization al communication in terms of the conduit metaphor. The Academy of Management Review, 9, 428-437. Bantz, C. (1993). Understanding organizations Columbia: University of South Carolina Press. Berger, P. L. (1979). Sociology of knowledge In R. W. Budd, & B. D. Ruben (Eds.), Interdisciplinary approaches to human communication (pp. 155-173). Rochelle Park, NJ: Hayden. Berger, P. L., & Luckmann, T. (1966). The social construction of reality. New York: Doubleday. Berkowitz, B. D. (1996). In formation age intelligence. Foreign Policy, 103, 35-50. Berkowitz, B. D., & A.E. Goodman (1989). Strategic intelligence for American national security. Princeton, NJ: Princeton University Press. Berlo, D. K. (1960). The process of communication San Francisco: Rinehart Press. Blumer, H. B. (1979). Symbolic interact ion. In R. W. Budd, & B. D. Ruben (Eds.), Interdisciplinary approaches to human communication (pp. 135-153). Rochelle Park, NJ: Hayden. Bruemmer, R. J. (1992). Intelligence community reorganization: Declining the invitation to struggle. The Yale Law Journal 101, 867-891. Buckley, W. (1967). Sociology and modern systems theory. Englewood Cliffs, NJ: Prentice-Hall, Inc. Capra, F. (1996). The web of life New York: Anchor Books.

PAGE 176

168Carey, J. W. (1989). Communication as culture: Essays on media and society. Boston: Unwin Hyman. Chia, R. (2000). Discourse analys is as organizational analysis. Organization 7, 513-518. Cline, R. S. (1989-1990). Book review: Stra tegic intelligence for American national Security. Political Science Quarterly 104, 695-696. Cobb, J. C. (1926). Th e social sciences. The American Journal of Sociology, 31(6), 721-732. Columbia Accident Investigation Board. (2003). Report of the Columbia accident investigation board. Retrieved April 6, 2007, from Conant, R. C., & Ashby, R. W. (1970). Every good regulator of a system must be a model of that system. International Journal of Systems Science, 1, 89-97. Connors, R., Smith, T., & Hickman, C. (1994). The Oz principle: Getting results through individual and organizational accountability. Englewood Cliffs, NJ: Prentice Hall. Deutch, J., & Smith, J. H. (2002). Smarter intelligence. Foreign Policy 128, 64-69 Dombrowski, P. M. (1991). The lessons of the Challenger investigations. IEEE Transactions on Professional Communication, 34, 211-216. Eisenberg, E. M., & Goodall, H. L. (1993). Organizational communication: Balancing creativity and constraints. Boston: St. Martin’s Press. Eisenberg, E. M., & Goodall, H. L. (1997). Organizational communication: Balancing creativity and constraints. Boston: St. Martin’s Press. Eisler, R. (1987). The chalice and the blade. San Francisco: Harper. Fernstermacher, G. D. (1979). Educational accountability: Features of the concept. Theory into Practice, 18, 330-335. Feldman, M. S., & March, J. G. (1981). Information in organizations as signal and symbol. Administrative Science Quarterly, 26, 171-186.

PAGE 177

169 Flanagan, S. J. (1985). Managi ng the intelligence community. International Security, 10, 58-95. Goodman, A. E. (1984-1985). Dateline Lang ley: Fixing the intelligence mess. Foreign Policy, 57, 160-179. Goodman, A. E. (1987). Refo rming U.S. Intelligence. Foreign Policy, 67, 121-136. Gross, A. G., & Walzer, A. (1997). The Challenger disaster and the revival of Rhetoric in organizational life. Argumentation, 11 75-93. Haydon, G. (1978). On being responsible. Philosophical Quarterly, 28, 4657. Huber, G. P., & Daft, R.L. (1987). The information environments of organizations. In F.M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication (pp. 130-164). Newbury Park: Sage. Johnson, L.K. (1985). Legislative re form of intelligence policy. Polity, 17, 549-573. Katz, D., & Kahn, H. (1966). Social psychology of organizations. New York: John Wiley. Kearns, K.P. (1998). Institutional accountabi lity in higher education: A strategic approach. Public Productivity & Management Review, 22, 140-156. Krone, K.J ., Jablin, F. M., & Putnam, L. L. (1987). Communication theory and organizational communication: Multiple Perspectives. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational Communication (pp. 18-40). Newbury Park: Sage. Lighthall, F. F. (1990). Engineering management, engi neering reasoning, and engineering education. IEEE Xplore 369-377. Retrieved September 22, 2005 from Louis, M. (1980). Surprise and sensemak ing: What newcomers experience in

PAGE 178

170entering unfamiliar organizational settings. Administrative Science Quarterly, 25, 226-251. Luhmann, N. (1990). Why “systems theory”? In O. Thyssen (Ed.) Some basic notions in the systems theory of Niklas Luhmann. McWhinney, W. (1993). Of paradigms and system theories. Santa Barbara: Fielding Institute. Mead, G.H. (1967). Mind, self, and society. Chicago: The University of Chicago Press. Meier, A., Jones, S., & Moyers, B. (2004). Ci rcle of accountability Retrieved March 2, 2005 from ow/politics/moyersandjones.html Mills, J. H. (2003). Making sense of organizational change. London: Routledge. Morgan, G. (1997). Images of organizations. Thousand Oaks, Ca: Sage Publications. National Commission on Terrorist Attacks upon the United States. (2004). The 9/11 commission report. New York: Norton. 9/11 by the numbers Retrieved January 18, 2005, from Ogul, M. S., & Rockman, B. A. (1990). Over seeing oversight: New departures and old problems. Legislative Studies Quarterly, 15, 5-24. Perrow, C. (1999). Normal accidents: Living wi th high-risk technologies. Princeton, NJ: Princeton Univ ersity Press. Presidential Commission on the Space Shuttle Challenger Accident. (1986). Report of the presidential commission on the spac e shuttle Challenger accident. Retrieved April 6, 2007, from Prigogine, I., & Stengers, I. (1984). Orde r out of chaos. New York: Bantam Books. Ransom, H.H. (1970). The intelligence establishment. Cambridge, MA: Harvard.

PAGE 179

171Ransom, H. H. (1975). Congress and the intelligence agencies. Proceedings of the Academy of Political Science, 32, 153-166. Ring, P. S., & Van de Ven, A. H. (1989). Formal and informal dimensions of Transactions. In A. H. Van de Ven, H. L. Angle, & M. S. Poole (Eds.), Research on the management of i nnovation: The Minnesota studies (pp. 171192). Romzck, B. S., & Dubnick, M. J. (1987). A ccountability in the public sector: Lessons from the Challenger Tragedy. Public Administraton Review, 47,, 227-238. Rosenbaum, E. W. (1971). Book review: The intelligence establishment. Military Affairs, 35, 114. Ruben, B. D. (1979). General systems theory. In R. W. Budd & B. D. Ruben (Eds.), Interdisciplinary approaches to human communication (pp. 95-118). Rochelle Park, NJ: Hayden. Ryle, G. (1949). The concept of mind. Chicago: University of Chicago Press. ? Schutz, A. (1967). The problems of social reality. The Hague: Martinus Nijhoff. Senge, P. E. (1990). The fifth discipline: The art an d practice of the learning organization. New York: Doubleday Currency. Senge, P.E. (1999). Fore word. In W. Issacs, Dialogue and the art of thinking together (pp. xv –xx). New York: Currency. Spretnak, C. (1997). The resurgence of the real. Reading, MA: Addison-Wesley. Steier, F., & Eisenberg, E. (1997). From records to relationships: Courting organizational dialogue at NASA. Cybernetics & Human Knowing 4 51-58. Tannen, D. (1998). The argument culture. New York: Ballentine Books. Thomas, J. B., Clark, S. M., & Gioia, D. A, (1993). Strategic sensemaking and organizational performance: Linkag es among scanning, interpretation, action, and outcomes. Academy of Management Journal, 36 239-270. Tompkins, P. K. (1984). The functions of hu man communication in organization. In

PAGE 180

172 C. Arnold & J. Bowers (Eds.), Handbook of rhetorical and communication theory (pp. 659-713). Boston: Allyn & Bacon. Tompkins, P. K. (1993). Organizational communication imperatives: Lessons of the space program. Los Angeles: Roxbury. Tompkins, P. K. (2005). Apollo, Challenger, Columbia: The decline of the space program. Los Angeles: Roxbury Publishing Company. Tulgan, B. (2004). H.O.T. Amherst, MA: HRD Press. Varela, F.J., Thompson, E., & Rosch, E. (1991). The embodied mind. Cambridge, MA: The MIT Press. Waterman, R.H., Jr. (1990). Adhocracy: The power to change. Memphis, TN: Whittle Direct Books. Watzlawick, P., Beavin Bavela s, J., & Jackson, D. (1967). Pragmatics of human communication. New York: Norton. Weick, K E. (1979). The social psychology of organizing. Reading, MA: AddisonWesley. Weick, K. E. (1985). Sources of order in underground systems: Themes in recent organizational theory. In Y. S. Lincoln (Ed.), Organizational theory and inquiry (pp. 106-136). Beverly Hills, CA: Sage. Weick, K. E. (1991). The vulnerable syst em: An analysis of the Tenerife air disaster. In P. J. Frost, L. F. Moore, M. R. Louis, C. C. Lundberg, & J. Martin (Eds.), Reframing organizational culture (pp. 117-130) New Park, CA: Sage. Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, California: Sage. Weick, K. E. (2001). Leadership as the legiti mation of doubt. In W. Bennis, G. M. Spreitzer, & T. G. Cummings (Eds.), The future of leadership (pp. 91-102).

PAGE 181

173San Francisco: Jossey-Bass. Weick, K. E., & Roberts, K. H. (1993). Collective mind in organizations: Heedful interrelating on flight decks. Administrative Science Quarterly, 38(3), 357381. Weick, K. E., & Sutcliffe, K. M. (2001). Managing the unexpected. San Francisco: Jossey-Bass. Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (2005). Organizing and the process of sensemaking. Organizational Science, 16 409-421. Wheatley, M.J. (1994). Leadership and the new science. San Francisco: BerrettKoehler Publishers, Inc.

PAGE 182

174 Appendices

PAGE 183

175Appendix A: Dates of 9/11 Public Hearings and URLS for PDF Records The first public hearing:3 March 31, 2003 ( ) April 1, 2003 ( http://www.9-11commission .gov/archive/hearing1/911Commission_Hearing_2003-04-01.pdf ) The second public hearing: May 22, 2003 ( ) May 23, 2003 ( ) The third public hearing on “Terrorism, Al Qaeda, and the Muslim World”: July 9, 2003 ( http://www.9-11commission .gov/archive/hearing3/911Commission_Hearing_2003-07-09.pdf ) The fourth public hearing on “Intelligence and the War on Terrorism”: October 14, 2003 ( ) The fifth public hearing on “Emergency Preparedness”: November 19, 2003 ( ) The sixth public hearing on “Security and Liberty”: December 8, 2003 ( ) The seventh public hearing on “Borders Transportation, and Managing Risk”: January 26, 2004 ( ) 3 Dates and titles for the 9/11 Public Hearings were taken from the official website of the National Commission on Terrorist Attacks upon the United States ( ). No titles were given to the first and second hearings.

PAGE 184

176Appendix A: (Continued) January 27, 2004 ( ) The eighth public hearing on “Counterterrorism Policy”: March 23, 2004 ( ) March 24, 2004 ( ) The ninth public hearing, with te stimony form Dr. Condoleezza Rice: April 8, 2004 ( http://www.9-11commission .gov/archive/hearing9/911Commission_Hearing_2004-04-08.pdf ) The tenth public hearing on “Law Enfo rcement and the Intelligence Community: April 13, 2004 ( ) April 14, 2004 ( ) The eleventh public hearing on “Emergency Response”: May 18, 2004 (http://www.9-11co May 19, 2004 ( ) The twelfth public hearing on “The 9/11 Plot” and “National Crisis Management” June 16, 2004 ( ) ` June 17, 2004 ( )

PAGE 185

177Appendix B: Members of the National Commission on Terrorist Attacks upon the United States Thomas H. Kean, Chair Lee H. Hamilton, Vice Chair Richard Ben-Veniste Fred F. Fielding Jamie S. Gorelick Slade Gorton Bob Kerrey John F. Lehman Timothy J. Roemer James R. Thompson Joseph M. Cleland (Hearings 14, resigned December 2003)

PAGE 186

178Appendix C: List of Witne sses Cited in this Document Richard Armitage Deputy Secretary of State4 Larry Arnold Commander, First Air Force and Commander of the Continental U.S. No rth American Aero-space Defense Command (NORAD) Region John Ashcroft Attorney General Zoe Baird Markle Foundation Stewart A. Baker former General Co unsel, National Security Agency William P. Barr former Attorney General of the Un ited States Michael R. Bloomberg Mayor, City of New York Steven Brill Author, After: How America Confronted the September 12 Era Daniel Byman Georgetown University Mike Canavan former Associate Ad ministrator for Civil Aviation Security, FAA Richard Clarke former National Counterterrorism Coordinator, NSC Jon Corzine Senator, New Jersey John M.Deutch former Director of Central Intelligence and Deputy Secretary of Defense Gerald Dillingham Director, Civil Aviati on Issues, General A ccounting Office Bodgan Dzakovic Civil Aviation Security Inspector, Transportation Security Agency Mamoun Fandy United St ates Institute of Peace 4 The titles listed for witn esses cited in this docume nt were taken from the 9/11 Commission Report : Final Report of the National Co mmission on Terrorist Attacks upon the United States (National Commission, 2004). Each title suggests the role that the Commissioners expected the witnes ses to assume during the Hearings, and the expertise each was to provide.

PAGE 187

179Appendix C: (Continued) Mary Fetchet Voices of September 11 Glenn Fine Inspector General, U.S. Department of Justice Cathal L. Flynn former Associate Admi nistrator of Civil Av iation Security, FAA John Gannon Staff Director, House Select Committee on Homeland Security Jane Garvey former fAdministrat or, Federal Aviation Administration Mark Gasiorowski Loui siana State University Porter Goss Representative, Florida Bob Graham Senator, Florida Rohan Gunaratna Institute for Defence and Strategic Studies, Singapore; author of Inside al Qaeda, Global Network of Terror John J. Hamre former Depu ty Secretary of Defense Jane Harman Representative, California Brian Jenkins RAND Corporation Murhaf Jouejati George Washington Institute Richard Kerr former Deputy Di rector of Central Intelligence Mindy Kleinberg, September 11 Advocates Randy Larsen Institute for Homeland Security Frank Lautenberg Senator, New Jersey Joseph Lieberman Senator, California John MacGafin former Associate Deputy Director of Operations, Central Intelligence Agency Claudio Manno Assistant Administrator for Intelligence, Transportation Security Administration James May Air Transport Association of America

PAGE 188

180Appendix C: (Continued) John McCain Senator, Arizona Mary O. McCarthy form er National Intelligence Officer for Warning Kenneth Mead Inspector General, Department of Transportation Jose E. Melendez-Perez Inspector, Customs and Border Protection, Department of Homeland Security Judith A. Miller former Gener al Counsel, Department of Defense Appendix C: (Continued) Norman Mineta Secretar y of Transportation Robert Mueller Director, Federal Bureau of Investigation Nancy Pelosi Representative, California Stephen Push Famili es of September 11 Magnus Ranstorp University of St. Andrews Janet Reno former Attorney General Condoleezza Rice National Security Advisor Marc Rotenberg Electronic Privacy Information Center Mary A. Ryan former Assistant Secretary for Consular Affairs Mary Schiavo former Inspector Gener al, Department of Transportation James R. Schlesinger former Director of Central Intelligence and Secretary of Defense Intelligence and Secretary of Defense Stephen J. Schulhofer New York University Charles Schumer Senator, New York Christopher Shays Repres entative, Connecticut Richard Shelby Senator, Alabama Craig Sincock United States Army (retired) Abraham D. Sofaer Hoover Institution

PAGE 189

181Appendix C: (Continued) James B. Steinberg The Brookings Inst itute and former Deputy National Security Advisor, 1996-2000 George Tenet Director of Central Intelligence Larry D. Thompson former Deputy Attorney General Peter F. Verga Principal Deputy Assistan t Secretary for Homeland Defense, Department of Defense Harry Waizer survivor, Cantor Fitzgerald, LP Michael Wermuth RAND Corporation Appendix (Continued) Randall Yim Director, Nation al Preparedness Team, General Accounting Office James Ziglar former Commissioner Immigration and Naturalization Services, Department of Justice Commission Staff cited in this Document Philip Zelikow (Executive Director) Susan Ginsburg John Raidt

PAGE 190

About the Author Sandra Cooper received a Bachelor’s Degree in English education from Florida Atlantic University in 1972 and a Mast er’s Degree in Communication from the University of South Florida in 1985. She has twenty-five years of experience as a teacher, communications trainer, and orga nization development consultant. Her experience includes teaching English in the Florida public school system, teaching English as a Second Language in Saudi Arabia and the Netherlands, conducting management training for public, private, an d not for profit organizations both in the United States and abroad. From 1989 – 2005, Ms. Cooper served as Director of Organization Development and Training at the Univer sity of South Florida (USF) where she enrolled in the Ph.D. progra m in the Department of Communication, College of Arts and Sciences. Since May 2002, her work ha s focused on directing, managing, and facilitating special assignments in the ar eas of organizational and human resource development at USF.