Division of Research School of Business Administration February 1990 HANDLING THE HEAT: OF ETHICS DURING OIRPORATE NEOTIATING THE MEANING THE IMPLEMENTATION OF ETHICS PROGRAMS Working Paper #634 Debra Meyerson The University of Michigan and Maureen Scully* Stanford University *Authorship is alphabetical and reflects a fully collaborative effort. The authors would like to acknowledge Kirk O. Hanson who provided guidance throughout the course of this project and the informants who gave their time and insights. However, the authors take full responsibility for the claims made in this paper. This research was supported in part by a grant from the Walter and Elise Haas Foundation, by a National Institute on Aging graduate award to Debra Meyerson, and by a National Science Foundation graduate award to Maureen Scully. FOR DISCUSSION PURPOSES ONLY None of this material is to be quoted or reproduced without the expressed permission of the Division of Research Copyright 1990 University of Michigan School of Business Administration Ann Arbor, Michigan 48109-1234

i I i I

Handling the Heat: Negotiating the Meaning of Ethics during the Implementation of Corporate Ethics Programs Ethics: Recently a hot topic in Congress, corporations, and the media. But what does "ethics" actually mean in these contexts? The media has used "ethics"as a justification to report anything in the private or public sector that has the slightest shade of wrongdoing, sometimes opportunistically heating up an incident simply by labeling it a matter of "ethics." Congress, responding in part to the media, has used "ethics" as a justification to investigate the personal lives of government officials and the inner workings of private corporations. Corporations, reacting to Congress and the media, have used "ethics" as a codeword to talk about everything from difficult value laden issues to straightforward concerns about specific business practices. To some corporate leaders, "ethics" means a focused concern about business practices — complying with specific rules and regulations regarding "fraud, waste, and abuse." For example, from this perspective, a corporate ethics program would deal with making sure labor time cards are filled out in pen instead of pencil, returning gifts to suppliers, and re-pricing $700 hammers. To other corporate leaders, "ethics" means a broad concern about values -- treating people "fairly" and with "dignity and respect." Defined in these broad, moralistic terms, corporate ethics may encompass such controversies as when plant closings and layoffs are warranted (and ethical), how much participation employees should have, and who should and should not make such decisions. For still other corporate leaders, "ethics" means both a discourse about morality and justice in the corporation and a rhetoric to enforce specific rules and regulations. For all corporations, the only thing that seems certain about "ethics" is that it is fraught with ambiguity and controversy. 2

Corporate ethics programs can have explosive, unintended consequences because of this ambiguity and controversy. Since the meaning of "ethics" or the scope of a corporate ethics program is ambiguous, heated value laden questions can be raised about what is "fair" and "just" and what, if anything, needs to be changed to become "more ethical." When different groups have different answers to these questions, ethics programs surface controversies that may previously have been suppressed. For example, corporate ethics programs often provide new vehicles, such as "ethics hotlines," for employees to raise concerns about "ethical violations." However, the concerns employees raise often go beyond the scope of ethics intended by top managers. Employees can justify raising these concerns by arguing their relevance to "ethics," broadly defined. Where top management talks about "ethics" as a matter of "treating all employees, customers, and suppliers with respect and dignity," callers to the hotline can appropriate this moralistic language of "respect and dignity" to complain credibly about a broad range of issues such as perceived discrimination and harassment, unfavorable performance reviews, favoritism, or layoffs. Employees can essentially put management in the position of appearing hypocritical if they do not live up to their own definition of ethics, a dilemma which adds a new legitimacy to employees' concerns. Whether or not management intended these other issues to be within the scope of "ethics," it can become difficult for them to ignore such calls and still maintain the credibility of their ethics program. In this way, ethics programs can provide the legitimacy, language, and structures that enable employees to participate on a wide, and often unintended and controversial, set of issues in organizations. For this reason, one corporate chairman claimed that his ethics program had "planted dynamite." In this paper we argue that whether or not the dynamite explodes depends in large part upon how companies initially define "ethics" and then subsequently defend or retract this definition. When corporations must implement an ethics program, as in our cases where the government mandated such programs for the defense industry, they must chose a 3

definition of ethics and maintain or renegotiate it as the program is implemented. Whether a corporation initially chooses a broad, value-based or narrow, rule-based definition, or equivocates between the two, can signify to their constituents -- government, customers, employees, public -- whether they see the program as a vehicle for change or for more superficial compliance. From their initial definition, corporations may be challenged to redefine ethics by constituencies with different interpretations. The process we discuss is one where at every stage the meaning of ethics and the scope of the corporations' ethics programs are shaped by previous definitions and interpretations and by the possible emergence of contradictions among definitions or contradictions between various definitions and actions. This paper traces this negotiated process through five case studies of corporate ethics programs in the defense industry. These five companies had all agreed to the same mandate to create ethics programs drawn up between forty defense contractors and the government. From our studies of how these companies responded to this mandate, we develop a process model of how three very different initial definitions of "ethics" lead to very different changes in definition, changes in practices, and changes in mechanisms and levels of opportunity for employees to participate on a wide range of issues. We argue that, although each company in our study developed its program in response to the same industry mandate, the ambiguity and controversy inherent in "ethics" allows the process of negotiating its meaning to produce very different programs. In addition to drawing from the data from our case studies, we draw from and contribute to two literatures. First, we draw from research on justifications in organizations. A corporation's initial definition of ethics and subsequent explanations of why definitions or practices are changing or not can be regarded as justifications that respond to or try to preempt real or anticipated challenges and that try to maintain legitimacy for the ethics program and those making the justifications. Much of the previous research 4

examines the effectiveness and appropriateness of various types of justifications: when justifications are and are not accepted by their intended audience and when certain types of justifications should and should not be offered (e.g., Bies, 1989). Most research assumes a static model of justifications; justifications are offered and either accepted or not In contrast, we suggest that whether or not a given justification is accepted or not depends upon where it fits in an ongoing process and whether it contradicts or supports previous justifications and actions. Justifications may not always have the intended effect of placating employees but can have the unintended -- and overlooked -- effect of making controversial issues salient and inviting questions, challenges, and even dissent. In such cases, there is a need for what we call "second order justifications" (Meyerson and Scully, 1988), which refer back to the initial justification and try to resolve any challenges or contradictions it provoked We also draw from and contribute to research on the problems of individual and organizational sense making under conditions of ambiguity (e.g., Eisenberg, 1984; March, 1976; March and Olsen, 1976; Meyerson, 1989; Weick, 1979). For example, March and colleagues argue that many common individual and organizational processes, such as decision making and learning, can be understood quite differently when we consider the possibility of ambiguous preferences or histories. Weick examines sense making as a process of reducing equivocality (a concept akin to ambiguity) through the ongoing action and interaction of pairs of individuals. Eisenberg views ambiguity as a condition that strategic actors can use to manipulate others' perception of reality; for example, ambiguity allows multiple interpretations of the same event to coexist, and thereby reduces the potential for intergroup conflict. Much of this research suggests that ambiguity is enabling because it encourages experimentation or provides discretion. In contrast, ambiguity is thought to be confusing and paralyzing in the research on role conflict and stress (Kahn et al, 1964; Rizzo et al, 1970). 5

We do not consider ambiguity to be consistently enabling or paralyzing, but instead explore changing responses to ambiguity, which can enable or constrain different constituencies at different stages in the process. Our case studies suggest that corporations initially respond in one of three ways to the ambiguity of ethics -- with a narrow, broad, or equivocal definition -- consistent with Martin and Meyerson's (1988) discussion of how organizational cultures reflect denial, channeling, or acknowledgement of ambiguity. Although Martin and Meyerson's typology offered a useful approach to our data, their analysis is static in that they do not describe the processes through which each of these responses to ambiguity is maintained or negotiated. We found that from the same initial definition, different contingent processes could arise. In this paper we illuminate these processes of negotiating the meaning of ethics and describe how these processes open up or close off opportunities for various levels of employee participation. However, before elaborating these processes, following the logic of induction, we will begin with an explanation of our methods. Methods Five Cases: Corporate Ethics Programs in the Defense Industry These days, few organizations could be as much under the gun to justify their practices, and even their very existence, as defense contractors to the United States government. These defense contractors are being questioned by the Department of Defense (DOD), Congress, and the public about their "ethics." In 1986, forty defense contractors signed the Defense Industry Initiative (DII). The DII was an agreement with the DOD that companies would scrutinize their own ethics by putting into place internal ethics programs, which involved such tasks as, appointing a corporate Ethics Officer, writing and distributing a corporate code of ethics to all employees, establishing ethics hotlines, conducting training sessions, voluntarily disclosing to the government any findings of wrongdoing. (See Appendix I for a list of the principles and audit criteria of the DII.) 6

While the DII was supposed to break down the ethics project into specific and manageable components, it remains open-ended, particularly as to how each corporation defines ethics. The 18 criteria of the DII are fairly open-ended and allow each company some latitude for how to put their program in place. Each year for three years (1987, 1988, 1989), companies would be audited on whether and how they have complied with each of the 18 criteria; the auditors give yes/no answers for each criteria. Each company must meet minimal criteria for compliance with the DII, but can decide whether and how far to go beyond that minimum in addressing "ethics." Because these programs were new and not yet institutionalized, and because the ethics requirements were so open-ended, they afforded an excellent opportunity to study how corporations cope with the ambiguity of "ethics" and with the ongoing process of justifying and re-justifying their ethics programs. Sample Our interview sample included 56 senior executives and 22 middle and lower level employees at five companies in the defense industry. (See Table 1.) This population consisted of at least eight senior managers or officers at each company, including the Corporate Ethics Officer (or a comparably titled position) and drawing from among the CEO, President, Corporate Senior or Executive Vice Presidents of Human Resources, Finance, Legal, Public Relations, Operations and Manufacturing as well as senior executives from the companies' operating divisions. We also interviewed 11 middle managers or engineers and 11 non-exempt employees. However, our sample best represented top management in each company. These are essentially the individuals who must justify their ethics programs to internal and external constituencies. Insert Table 1 about here 7

Thus, most of the challenges to justifications we report are from accounts offered by senior managers. These may be real or imagined. Only one of the five companies allowed us to conduct in depth interviews of lower level employees. Two other companies provided limited access. We therefore have limited and unsystematic data on employees' actual reactions to the ethics programs, but have enough to draw some fairly strong inferences. Moreover, because we are interested in the process of justifications which may occur in response to either real or imagined challenges, this problem is not a major one. Our assessment of the opportunities available for participation, therefore, rely on limited data from lower level employees, senior management's verbal reports, and structural indicators (e.g., hotlines, surveys). In addition to the informants from the five defense contractors, prior to conducting our investigation we interviewed a number of other informants who in many ways shaped our initial questions. For example, we interviewed a Chief Counsel for the U.S. Defense Contract Administration Services (Western U.S.). This informant was responsible for the other side of this enterprise: purchasing services from the defense contractors and enforcing codes of conduct sanctioned by the Defense Department We also interviewed and participated in discussion with a number of academic and industry personnel who participated in the Bay Area Ethics Consortium, an organization devoted to discussing various academic and practical aspects of corporate ethics. Methods Between September of 1987 and March of 1988 we conducted in depth, semistructured interviews with top managers at the five defense contractors. This time period was when the programs were being implemented and when companies were preparing for the first DII audit. Since the theory emerged through an iterative process of fieldwork and analysis, the interviews became increasingly structured as the project progressed. Both of 8

the authors participated in all of the interviews. Typically, one of the authors was responsible for asking questions from the interview guide while the other author listened for and pursued new topics or possible contradictions that arose in the course of discussion. All interviews were tape-recorded and professionally transcribed. Although the structure of the interviews sharpened, from the project's onset all interviews tapped at least the following issues: 1) how top managers initially justified the ethics program, 2) how they imagined and experienced employees' reactions to their justifications, 3) how they may have had to modify their initial justifications and provide a series of "second-order justifications" -- in response to real, imagined, or anticipated challenges, 4) how they justified other related actions of the company brought into the spotlight by the ethics program, and 5) what policies they actually changed as an intended part of the program or as an unintended result of employee challenges2. Analysis As largely an inductive process of theory generation, our analysis was ongoing (e.g., Glaser and Strauss, 1967; Miles and Huberman, 1984). However, a few distinct stages can be identified. The first stage of informal analysis occurred after we had interviewed managers at the first three companies. This stage simply consisted of reviewing the transcripts for themes and patterns so that we could follow up on relevant topics in subsequent interviews. From this first analysis emerged a rough skeleton of the three general strategies of negotiating the meaning of ethics outlined below. 2 Our method required that we infer longitudinal processes from managers' retrospective accounts and from archival sources. Archival sources, such as press releases and company announcements, that we had across time periods provided us with traces of the actual language used to define ethics and justify the ethics programs. These traces enabled us to locate discrepancies between documented and verbal accounts of the processes and, importantly, identify managers justifications of these schisms. 9

The second stage of analysis took place after we had interviewed in all five companies. In this stage we compared the companies using summary transcripts from the tape recordings of our daily debriefs. These summaries were compiled and then used to identify important dimensions, distinguishing characteristics, and general typologies. Using a general coding template that helped classify the ethics programs along several diminsions and categorize the types of responses to various sources of ambiguity (and contradiction), we coded a subsample of matched positions (totalling approximately two thirds of the interview transcripts) from each company. We also reviewed archival sources (e.g., company reports, media clippings, ethics codes, training materials and literature). From this analysis, the authors wrote up company summary reports and sent them to each company for review. We met again with one Ethics Officer for detailed feedback on his company and the anonymous comparative information we provided on the other four companies. These reports guided the final distinct round of analysis. We analyzed each transcript for 1) how top managers defined "ethics" to various constituencies; 2) how they initially introduced (and justified) the ethics program to employees (including the language and structures they used); 3) how they perceived employees' response to ethics and the ethics program; 4) how they in turn answered to these real and imagined responses; 5) how they coped with the various ambiguities in ethics. Although we coded the interview transcripts for responses to these and other issues, we decided that standard statistical significance and reliability tests would be inappropriate because of the unstructured nature of our data. Since the interviews were not identical in form or content, we felt it would be inappropriate to conduct or report, for example, that 40% of managers said X. As described earlier, the difference in interview format resulted from our following up on unexpected topics and new lines of inquiry as we learned more about the issues. Here, our aim was to develop, rather than to confirm, theory (Eisenhardt, 1989; Glaser and Strauss, 10

1967). Moreover, the theory we set forth below consists of a set of processes to be compared overall, in context, rather than a set of discrete independent and dependent variables with generalizable properties distinct from their contexts (Mohr, 1982). Thus, instead of trying to isolate and measure a set of variables, our aim was to illuminate some processes, each of which contains interdependent stages. Negotiating the Meaning of Ethics All five companies that participated in our study had met the minimal compliance requirements of the mandate, yet some went beyond strictly the letter of the law. For example, they all had a corporate ethics program director, yet the visibility, title, authority of the director varied. All companies had implemented ethics training programs, although these varied in scope (the range of issues covered), length (size of commitment), and in coverage (whether or not hourly employees were included). All five had a corporate ethics hotline as well as division ethics hotlines, but the kind of call that would be accepted over the hotline varied. For example, in each company the callers to the hodines had a variety of concerns, including reports of perceived fraud (e.g., "My co-worker has charged his time to the wrong contract"), requests for information (e.g., "Can I accept a calendar from a supplier?"), and complaints about human resource issues (e.g., "My boss hasn't let me see my performance review"). According to the letter of the mandate, the hotlines were implemented to encourage the first and second kind of call —allegations of "fraud, waste, and abuse" and preventative measures regarding these issues. However, all companies found that human resource calls predominated by far. Whether or not these human resources calls, which often dealt with quite heated issues, would be considered within the 11

domain of ethics, and thus answered on the ethics hotline or passed along, varied across the companies. Yet the issue of how to handle these calls seemed to be an important one at all companies. Accepting these calls would imply a broadening in the definition of ethics and the scope of the ethics program to include HR concerns, like issues of "fairness." Disclaiming them would reaffirm a narrow definition and focused ethics program. Which choice they made, therefore, would have the effect of defining or redefining "ethics" in the corporation, defining or redefining the scope of the ethics program, and ultimately opening up or closing off opportunities and justifications for employee participation. For this reason, our discussion below considers in particular how initial and subsequent definitions of ethics invited or blocked opportunities for employees to participate on a variety of issues of'ethics." Three Initial Definitions of Ethics Our study revealed three initial definitions of ethics, employed by the five corporations as shown in Figure 1. Each initial definition, and then subsequent changes in that definition, suggested certain consequences for employee participation. Insert Figure 1 about here Stages in the Justification Process Our data also suggests that the process of negotiating the meaning of ethics occurred in four distinct stages3. It should be noted that while we take the initial stance of companies on ethics as the starting point from which a process developed, a structural account, such as a resource dependence explanation (e.g., Pfeffer and Salancik, 1978), would usefully 3 It is possible and likely that additional stages unfolded in each or some of the companies since the time this study took place (1987-1988). 12

suggest why each company developed its initial stance. For example, the company most dependent on government contracts took the ethics program most seriously, striving to fulfill the spirit as well as the letter of the mandate, and the company least dependent on government contracts provided the most minimal and grudging compliance. What resource dependence would take as its endpoint, we use as our starting point Our consideration of the process that ensues does not necessarily make different predictions about the final position of each company, but does allow for this possibility. As corporations respond to constituencies in addition to the government, such as their employees, their initial stance could change, and did in two of our cases. Stage 1: Justification of Initial Definition. Each company started with some stance toward the DII mandate. Two aspects of their definition seemed to vary. First, what is the scope of the program? Top management at the companies defined their programs in terms of rules and compliance, or in terms of broad values, or they equivocated between the two. Second, is this program a change from the status quo or not? Top managers either justified the program as externally imposed and not a change (e.g., "we have always behaved ethically"), or as internally motivated and a real change (e.g., "we need to behave more ethically"). Alternatively, they equivocated between the two, trying to explain how the program was a safe move in the present climate, but not warranted by specific problems. These possibilities are displayed in Table 2. Insert Table 2 about here Stage 2: The Need for Further Justifications. The appearance of contradictions prompted top managers to provide further justifications. First, two justifications might contradict, particularly in the cases where there was ambiguity as to whether the program 13

was about values or rules and whether it represented a real change in behavior or not. Second, there might be acontradiction between a justification and a particular policy in the ethics program For example, the statement, "All issues of employee conduct should be regarded as ethics issues" contradicts a policy of not handling Human Resource (HR) complaints on the ethics hotline. Finally, a justification and a policy outside the ethics program may be contradictory. For example, the statement, "Treat all customers and colleagues with respect," and a policy of closing plants without notice might appear contradictory. These possibilities are illustrated in Table 3. Insert Table 3 about here These contradictions demand a resolution if the program is to remain credible both to those who run it and to its internal and external audiences. Moreover, these contradictions must be responded to by the company to prevent employees from voicing their concerns directly to the Department of Defense (through their ethics hotline) or to the press. To demand a response from the company, such a threat may be real or imagined. Stage 3: Second Order Justifications.. As shown in Table 4, top management can respond to a contradiction between two justifications by favoring one of the competing justifications over the other, or alternatively, by providing an overarching justification that attempts to integrate the two. Insert Table 4 about here Top management can try to resolve a contradiction between a justification and an ethics or outside policy in one of two ways: 1) they can change their policy (e.g., from small steps, like starting to accept calls about HR concerns, to major steps, like 14

reconsidering a plant closing), or 2) they can change their justifications (i.e., redefine what ethics means and how it will be implemented) so that their actions are consistent with the new definition. Finally, they might also try to deny the contradiction, again by providing an overarching justification of how the statement and policy do not contradict For example, they might justify a plant closing by explaining that it does not represent disrespect for employees but actually implies respect for employees who will be retained by keeping the company solvent. Stage 4: Ongoing Justifications. The above responses may not always be sufficient In particular, if top management changes its definition of the program midstream, employees might regard the program as a sham. To defend against such perceptions, further justifications -- or further changes in course of action -- may be required. As depicted and summarized in Figure 2, the cycle of justifications and responses can continue. Insert Figure 2 about here Five Cases of Justifying Corporate Ethics These four stages can describe the process by which each company initially and subsequently defined and justified ethics and its ethics program. In their effort to justify and thereby define what they mean by "ethics," these companies had to cope with the ambiguous and controversial concept of "ethics" and the ambiguous mandate of the DII. In this way, each process represents a distinct means of responding to a critical ambiguity. Below we contrast how each process of responding to the ambiguous mandate results in different definitions of "ethics," different scopes for the ethics program, and quite different 15

mechanisms and levels of opportunity for participation. Figure 3 contrasts the companies' processes. Insert Figure 3 about here I. A Narrow Definition: Denial of Ambiguity Curbs Participation Top management at two of the companies chose a narrow, rule based definition of ethics as "business practices" from the beginning. That is, they focused on how to comply with the specific details of procurement, labor charging, and other regulations. One of these companies maintained this circumscribed definition and the other eventually broadened its definition so that ethics was about values. The former continued to thwart participation on the more controversial topics while the latter eventually began to entertain a broad range of employee concerns within the domain of ethics. Case 1. Alpha Company: Maintaining a clear definition. Stage 1: Justification of Initial Definition. Several informants at Alpha felt that the DII represented a"witch hunt" against defense contractors. They felt that they need only comply with the letter of the DII until the negative mood of the public and Congress changed. They described their stance simply: Management has to get into this ethics business,compliance business, I should say. We have to keep people's attention on compliance. Ethics violations are in fact rules violations. 16

Stage 2. The Need for Further Justifications. Even though this stance was straightforward, Alpha Company was nonetheless faced with providing further justifications about what scope of issues was appropriate to the ethics program. First, there was the issue of whether the program was a change from their previous behavior or not They stated from the beginning that they saw the impetus for the program coming from the external environment and the climate of "contractor bashing." Despite the public's negative sentiments, they maintained staunchly that their "long tradition of integrity" was not in question. To make this position credible, they had to justify the introduction of an ethics program as not a penance for wrongdoing, but merely an adherence to government regulations that keeps the government, as a customer, satisfied. Second, there was the scope of the program. Even though the company defined ethics narrowly, many employees greeted the introduction of an ethics hotline as a channel for airing HR complaints. Some callers hoped to find an advocate for a broad array of problems (e.g., how their boss treated them) that they considered logically to fall in the domain of "ethics." This unexpected wave of calls prompted top managers to re-justify the focused scope of the program. Because at Alpha the program was never intended to go beyond the strict letter of the law and therefore not defined to include HR issues, these second order justifications were simply reaffirmations of their initial stance. Stage 3. Second order justification. First, regarding whether the program represented significant change, the ethics program was re-justified as simply a continuation and promotion of their past practices, not a correction of them or departure from them. Management wanted explicitly to emphasize that no substantive change was needed because they were "already ethical." As the Corporate Ethics Officer stated it: This is not new. We're simply beefing up the message and trying to get it out to more people, or more frequently trying to provide them with more examples. 17

While the awkward language suggests that this is a delicate topic, the predominant message was that this program was not supposed to represent a change. Second, regarding the program scope, top management reinforced its position that ethics is about a specific set of "business practices," not controversial HR issues. The Ethics Officer made it clear that his program was not an invitation for employees to voice their HR concerns, nor was he a new form of ombudsman. Articles and interviews appeared in the company newspaper to this effect. In his words, "in no case do I go and advocate the employee position." The definition of the ethics program remained consistent in these further justifications. Since management had not included within the scope of ethics anything more than business practices, employees did not have the grounds on which to demand more. Though employees could have tried to demand that the current definition of ethics be changed, this would have been much harder than trying to make sure that the existing definition be fulfilled. Management was not put on the spot. Top managers did make some effort to insure that their circumscribed approach to ethics made sense to people. Many jokes and stories circulated about the picayune features of the DOD rules with which the company had to comply. They "blamed" this onerous compliance program on an external mandate. For example, people laughed at how, in an effort to avoid a "conflict of interest," an employee could not buy an admiral a 25c cup of coffee. Instead of paternalistically telling employees what is and is not really ethical, some managers simply agreed that the rules were absurd, but that the reality was that everyone must comply to the letter or risk losing their job. Compliance with a specific set of rules, and nothing more, was required. Stage 4. Ongoing Justifications. Since Alpha Company had effectively preempted the emergence of problematic contradictions and carefully circumscribed the 18

definition of ethics to exclude the heated HR concerns, they had no need to make further justifications. The use of hotlines, training programs, and other structural mechanisms of the ethics program were to be used for the reporting of ethical violations, strictly defined. In this way, Alpha Company managed to preclude participation on the more value laden issues (e.g., of "fairness") and preempt tough challenges on the company's ethics, broadly defined. They had carefully maintained their narrow definition of ethics and reinforced it with further, consistent justifications as required. This process and its outcomes contrasts with the companies described below. Case 2. Beta Company: Expanding the definition of ethics Stage 1. Justification of Initial Definition. Like the Alpha Company, the Beta Company began with a circumscribed definition of ethics as "business practices." They also emphasized that, inasmuch as they had integrity all along, this program was not a correction or a change. Initially, the program applied only to the approximately 25% of their business that related to the defense industry. Stage 2. Need for Further Justifications. Just prior to our study, Beta's 'Director of Integrity and Efficiency" (the manager in charge of defense industry ethics) was moved to the corporate office to oversee ethics in the defense division and the other divisions. This structural reorganization was the contingency that prompted the broadening of the ethics program and some philosophical reflection. After the reorganization top managers were seriously reconsidering whether it made sense to implement a corporate-wide ethics program for all the divisions, not just defense. One manager described the rationale for the decision to broaden the program's coverage: A recognition by our corporate leaders that you cannot segment the corporation when you're dealing with something such as ethics which does (touch) everyone. 19

Yet they did not want the inclusion of the divisions to suggest that an ethics program was suddenly necessary, especially since before the reorganization they had justified that they were already sufficiently ethical. Without the obvious impetus of the DII, it was not completely clear why the non-defense divisions should have this program. Furthermore, an ethics program embracing the whole corporation would obviously not focus on DOD regulations, but would have to be broadened in scope to include issues that were relevant to the other divisions. Stage 3. Second order justification. Since there was still a great deal of discussion and confusion about how to announce the corporate-wide ethics program, management needed to provide a second order justification. But their stance was reluctant and consequently confusing: We're not coming out and saying that you have not had an ethics program before, and you're going to have one now, and this is it. We're taking a subtle approach. Regarding the scope of the program, some managers realized that broadening it beyond rules for "how we do business with the government" would require teaching employees how to exercise more discretion in making ethical judgments on a range of issues. We ran more of a compliance program rather than what you might call an ethics program. Ethics is more of the personal decision point, that you allow people more of a freedom to make personal decisions. We are starting to address that, but historically, we've been in more of a compliance program. In addition to providing more discretion, broadening the scope of the program seemed to create a more open atmosphere where employees throughout the company were encouraged to speak up about their "ethical concerns." Even though what was meant by "ethical concerns" was still ambiguous at Beta, the broadening of the program provided employees with more structural channels through which they could participate. For 20 N.

example, until the ethics program was extended to the non-defense divisions, few employees used the "open door" policy. One manager claimed that ethics provided the excuse for employees to take more advantage of this structural mechanism. Stage 4. Ongoing Justifications. Beta Company had extended the boundaries of its program and had raised the possibility of a broader definition of ethics, but had not committed to the broad definition entirely since managers were concerned that giving employees more discretion might undermine the effort to ensure compliance within the DII framework. However, because they had already introduced this definition, returning convincingly to a definition of ethics as just a limited set of business practices would have been problematic. Moreover, broadening the boundaries of inclusion seemed to make more structural opportunities for participation available to employees. Summary. In the first case, using a narrow definition initially seemed to be a stable solution for Alpha Company, but one that may be vulnerable to challenges that might push toward a broader definition. This would in part depend on how long their narrow definition can continue to preempt employee participation. At Beta Company, maintaining a narrow definition and denying the ambiguity of ethics ceased to work in part because a structural reorganization affected how ethics could be defined. Enlarging the boundaries of inclusion seemed to have the dual effect of broadening the definition of ethics and scope of the ethics program and encouraging employees to use more discretion in their decisions. In addition, the broadening of Beta's ethics program seemed to provide the excuse for employees to use preexisting structures for participation. II. A Broad Definition: The Ambiguity of Ethics Provokes Participation Only one company in our sample started with a broad definition of ethics. Managers at Gamma Company were rather overwhelmed with the number and range of issues that employees raised in response to their initial value based definition. They 21

answered employees' concerns either by responding with real changes in their actions or by justifying why change was not needed and thereby redefining what ethics did and did not include. This re-definition often invited another round of challenges. Thus, their initial broad definition and subsequent re-definitions seemed to provoke participation and dissent on a wide range of issues. Case 3. Gamma Company: Managing unexpected reactions. Stage 1. Justification of Initial Definition. The Gamma Company started with the approach that ethics was about a broad set of issues that concerned values. This company had been hit hard with indictments and was prepared to admit that major internal change -- changes in the way they think and act -- was needed. Interestingly, part of the rationale for this initial stance was that a set of encompassing values would illustrate the spirit of the mandate and include the specific rules. The code of conduct booklet included statements about all employees treating one another with "respect and dignity." This booklet made the broad values salient Subsequent discussions and training sessions about ethics continued to use language oriented toward these values. The values approach seemed best to signify that this company was working on really overhauling the way it does business and improving its image as an ethical company. Stage 2. The Need for Further Justifications. The moralistic approach and language of Gamma's ethics program looked to employees like an invitation to take their own agendas and raise them as ethical concerns. A huge volume of calls (676 in ten months of 1987) flooded the ethics hotlines, including some quite heated HR calls. Top managers expressed ambivalence about how to respond to these calls. On the one hand, they wanted to respond to the HR calls partially as a way to stimulate internal improvements. On the other hand, they did not want to get distracted by every dispute or have to make concessions on every controversial issue. They realized that any reluctance to 22

handle the HR calls would look like a contradiction to employees: they could not tout a broad, value based definition of ethics (e.g., "fairness" and "respect) and refuse these HR calls. Stage 3. Second order justification. They saw two ways to avert this contradiction. They could rejustify and narrow their definition of ethics. Or they could take action on the HR issues. Narrowing the program too readily might make it appear a sham. So top managers chose to entertain employees HR concerns. For example, they received a call complaining that it was unfair to make employees turn over their "frequent flier" points while executives flew first class; this represented a "double standard" over cost-cutting. In response, management decided that all employees could keep their frequent flier points and that no one could fly first class. More significantly, in response to the wave of calls that the ethics program provoked, managers set up a network of 29 hotlines. In addition, other structural mechanisms were established to respond to these calls, including establishing an ombudsman in each division and a comprehensive employee attitude survey administered to everyone in the organization. These structural mechanisms, the strong, moralistic language used to talk about ethics, and the apparent contradictions that invited challenges further legitimated and broadened the range of employees' claims. Stage 4. Ongoing Justifications. The scope of employee requests broadened into more controversial areas. Through the ethics program, employees used the strong moralistic language and apparent contradictions to challenge management about organizational fairness. For example, the union leader in one division wrote an impassioned letter to the CEO stating that the company's decision to lay off several hundred employees after only the fourth day of their strike over a wage cut was in sharp contradiction to the value of fairness underlying the ethics program. 23

Management's response to this charge could not simply be to hire these people back. Earlier actions in response to employee concerns had been less costly and controversial than the layoff issue. Thus, the CEO, generally the most vocal champion of the ethics program, worried that the ethics program had planted "dynamite" in the corporation. The response to these more controversial issues began to take several forms which led to a need for yet further justifications. First, there was some discussion of narrowing the scope of the ethics program and emphasizing "business practice" issues as the core issues. Second, the hotlines began to function to "cool out the mark" -- there was sympathy but no promises for the callers' complaints. Third, there was some attempt to come up with "overarching explanations." For example, the CEO explained how maintaining a broad definition of ethics did not mean caving into every party that chose to disagree with a company policy. However, the language of this kind of explanation began to sound a bit contrived and convenient. Since none of these responses completely quelled the more controversial concerns, Gamma Company remained embroiled in a cycle of justifications and responses. Summary. For Gamma Company, a broad, value based initial definition snowballed and had the effect of creating a language and a set of structures with which employees could air their grievances and voice their requests. Moreover, management's moralistic language provided a new legitimacy for employee claims. For this reason, we suggest that a broad definition encouraged participation on a range of issues and may have even provoked dissent. The implications and the future of this participation and dissent are not yet clear. To the extent that hotlines vent frustrations and "cool out the mark," this dissent may not result in substantive changes and may even lessen the impetus for dissent through other channels. Since this participation is conducted in the moralistic language of ethics, it may have had the unusual feature of being a form of participation that perpetuates 24

itself, that demands a response, and that cannot be ignored without endangering the credibility of the ethics program itself. III. Equivocating over Definition: Ambiguity Causes Confusion about Participation Companies that followed this strategy were unable to take an initial clear stance on the ethics program; they permitted the contradictory positions to remain in tension. The following cases consider some of the consequences of this process. As stated earlier, the two contradictions that recurred are: 1) "This program represents a change in our behavior" versus "This program does not represent a change in our behavior" and 2) "Ethics is about values" versus "Ethics is about specific rules." Case 4. Theta Company: Ambiguity creates confusion. Stage 1. Justification of Initial Definition. From the beginning, Theta's interpretation of ethics and the ethics program was ambiguous. Because of its history of wrongdoings, Theta already had the rudiments of an ethics program in place. In fact, Theta's management was quite proud of how ethical the company had become, and Theta's chairman was among the most visible spokespersons for corporate ethics programs in the defense industry. The messages at Theta were clear, but contradictory, along the lines of, "We are already ethical," and also, "We need to become more ethical." This confusing stance led management to try to explain whether Theta's program represented a change or not. However, in doing so, they only confused the issue further: We're no stranger to the ethics business, but what we did was reinstituted some of the initiatives that we didn't have. 25

Stage 2. The Need for Further Justifications. The general pattern, as at other companies, was an influx of HR calls to te ethics hotline. But it was not this overall pattern that suggested to top managers the need for furtherjustifications. Instead, some top managers focused attention on particular problematic cases and approached in a more ad hoc fashion the use of the ethics hotlines and the related issue of what definition of ethics should be employed to increase or restrict the use of the hotlines. For example, some employees at Theta, who feared they were to be let go, decided to call the ethics hotline to make a complaint. Having done so, they could not be easily let go under the provisions of whistle-blower protection, but they were considered difficult employees who might well have been let go before making their call. This example suggested to top managers that there may be many exceptions or special cases to consider with respect to the enactment of ethics; however, any attempt to define those exceptions might look like a deviation from or convenient manipulation of their initial definition of ethics. Stage 3. Second order justification. This response, which management hoped would clarify the initial position, only made it more confusing. In a sense, Theta was locked into Stage 1, because any attempt to advance and provide the needed rejustifications resulted in the same initial confusion. For example: I don't really see this as a change effort in this company, except in the area of openness and communication. I think we've got a cultural change that has to be involved here. As far as the encouragement of reporting and fixing of wrongdoing, we have always had that kind of thing. We have always had the open door. Stage 4. Ongoing Justifications. Attempts to resolve the ambiguities about ethics and the ethics program simply generated more ambiguities. The ambiguities made it difficult to determine what ethics meant and thus it became difficult for either management or employees to make forceful demands of each other under the rubric of ethics. Insofar as the ambiguity about ethics confused people and led to a number of opportunities for 26

employees to voice their confusions and concerns, participation was allowed, but seemed ineffectual. Case 5. Zeta Company: Ambiguity allows creative slack. Stage 1. Justification of Initial Definition. Because of the company's tradition of divisional autonomy, top management at Zeta was reluctant to pass down a central mandate about how the ethics program should look. They allowed each division to interpret the DII in its own way. Consequently the divisions received mixed messages about the meaning of ethics at Zeta: At some divisions of the corporation, 'ethics' is the word. Because that is the word that management wants to use. They see this as a program about ethics. And personal ethics and personal values and those things become company ethics and values. Other divisions and the corporate office have taken the approach that it is a bit too emotionally loaded and that what we really are talking about here anyway is the company's practices, and individual employee's practices in conducting their business relationships. That to me is what business practices means. Stage 2. The Need for Further Justifications. Given that ethics was at least partially defined to include broad values, employees could legitimately bring forth concerns about fairness, particularly fairness of various HR policies. At the same time, given that ethics was defined by specific business practices, managers could legitimately decide to emphasize only those aspects of the program that were about compliance. Take the following example: Over 50% of the stuff coming in on the open line is personnel related, employee relations oriented. 'I don't like my supervisor,' or, 'Do you know what that character is doing,' or whatever it may be. And those things are automatically funnelled directly to the Human Resources element in that division or wherever it might be. Or, 'I was supposed to get an increase and, by God, they didn't give it to me, and that is unethical.' Everybody's view of what is ethical and what is unethical is in the eye of the 27

beholder, in many respects depending on what it is that the person is looking at. Stage 3. Second order justification. The company continued to define ethics to include both broad values and specific rules by offering a second order justification as to how these two interpretations could coexist For example: The higher you go up in the organization, where you have to make decisions, the more that values are being emphasized. When you get down in the organization, where it's just rote or where there's not major decision making, it's more compliance. Similarly, they offered a further justification of how the ethics program represented both change and continuity: I saw [the ethics program] as a real major challenge and the benefit of doing it well seemed to me went beyond just protecting the company liability. It is positive reinforcement of the things that make [Zeta] a good company. And I think [Zeta] is a good company, and I think it is an important task to continue to reinforce those things. Stage 4. Ongoing Justifications. Zeta's acknowledgement of ambiguity might have been a realistic recognition that values guide the interpretation of business practices, while at the same time, business practices enact or implement certain values. Indeed, the meaning and domain of ethics may be inherently grey. Rather than imposing one interpretation over the other, they kept alive a debate over the interpretation, and perhaps allowed the deepest and most controversial form of participation: discussion over the meaning and scope of ethics. The cost of permitting this dialogue may have been an unavoidable stage of sending mixed message to employees. The benefit was perhaps a more culturally specific and longer term commitment to the spirit of ethics. Summary. Acknowledging the ambiguity in the definition of ethics, as they did at Theta and Zeta, can have the cost of creating confusion, but the benefit of keeping a realistically confusing and complex issue open for discussion. In this way, keeping the ambiguity alive may allow employees to participate on a wide range of issues, including the 28

definition of ethics, but the effectiveness of this participation is questionable. A rich and nuanced definition of ethics may be difficult to reach, however, since it may sound circular or opportunistic. Yet a simple and focused definition of ethics may be reached perhaps only at the expense of false clarity. This leaves open an important question: Who ethically gets to decide what exceptions or conditions are to be made so that multiple definitions of ethics can be considered as partly right? Conclusion and Discussion At one level, our study showed that a impact and effectiveness of a justification depend in part upon the justification's context and place within a sequence of justifications. For example, the justification, "Ethics should be about business practices," appeared in three ways in our data, with three different consequences: 1) As an initial justification for Alpha, it enabled them to maintain a consistent approach. 2) As a "second-order justification" for Gamma, it threatened the credibility of the program, because it contradicted their initial justification, that "Ethics should be about values," and seemed too easy a retreat 3) As part of a pair of contradictory justifications, it created confusion at Theta, but allowed discussion about realistic tensions to surface at Zeta. We found that when top management received reactions to their justifications -- and particularly where contradictions among justifications or actions were pointed out -- they tried to respond with further actions and justifications, whether to strengthen a point or to resolve a contradiction. Most problematically for top management, it was sometimes necessary to make retractions, change course, grant concessions, or give overarching justifications in an attempt to explain away an apparent contradiction. Our research argues that justifications, like interpretations (Weick, 1979), cannot be taken as static and context free. Indeed, 29

considering them as such may be particularly problematic when the justifications are supposed to address issues that are controversial or ambiguous. An another level, we argued that ethics is controversial because the very mention of the topic can raise questions about what is "fair" and what is "right," and different parties have different ideas about fairness and rightness. Moreover, ambiguity allows different parties to have different legitimate notions about what lies within the scope of "ethics." Thus, because of the ambiguous and controversial nature of ethics, employees could credibly challenge the initial definition of ethics and, in doing so, set off a spiral of justifications and re-justifications. In this way, the process of defining ethics and delimiting the scope of an ethics program was a negotiated one, resembling the process of double interact as described by Weick (1979). However, the extent to which multiple constituencies actually and effectually participated in this negotiation varied across the companies. We found that the opportunities and the mechanisms available for various forms of participation seemed to vary across the companies. Whether structural and/or linguistic artifacts of the ethics program provided employees with the opportunity and credibility to participate seemed to correspond in part to how the companies initially talked about ethics. Defining ethics broadly, in its spirit, as at Gamma, provided employees with a language and a legitimacy with which to participate. When top managers talked about ethics as "treating each other fairly," but then claimed that complaints about inequitable compensation were "not a matter of ethics," their legitimacy and the credibility of the program was threatened by their apparent hypocrisy. In contrast, defining ethics narrowly, in its letter, as at Alpha, prevented employees from trying to use management's language of ethics to push for a response to their broader ethical concerns. Yet employees at Alpha still attempted (however unsuccessfully) to use the structural byproducts of the ethics program, like the ethics hotline, to voice these concerns. 30

Finally, leaving the definition of ethics ambiguous, as at Theta and Zeta, might have provided the language, structures, and legitimacy for employees to participate on a broad range of issues; however, the confusion about ethics at Theta seemed to make it difficult to ground this participation effectively. Both parties to a conflict could hold either of a pair of definitions, or subsequent justifications, to be true, as they found convenient. For example, management could claim that they had been saying all along that, "This program does not want to change our behaviors," which was partly true. Employees could claim that management had been promising all along that, "This program does hope to change our behaviors, " which was also partly true. The more powerful party could try to insist on its current definition, but the previous equivocation may have meant that they do so without much credibility. The result could be either a stalemate, or a realistic and productive dialogue. Indeed, we are not completely sanguine about the effects of this participation. There seem to be limits to its scope and effectiveness. As the participation grew into more extreme dissent, such as the protests that a plant not be closed, top managers were less able or less inclined to respond as the dissenters wished. In addition, as the mechanisms and language of the ethics program became increasingly institutionalized, they may have had less impact. Some top managers drew analogies with affirmative action programs, which were derived and implemented from a set of value based standards, but over time have become a set of routine procedures to follow, in letter if not always in spirit. We could describe similar fates for other, initially value laden organizational initiatives. There also seemed to be a paradoxical limit to participation. Even a broad definition of ethics, which encourages participation, remains a top-down definition of ethics. It may be most ethical to allow employees to participate in defining what "ethics" does and does not include and how it will be implemented. A lack of definition may have empowered employees by allowing them to participate in the meaning and scope of ethics, perhaps the 31

deeper and more controversial decision. Allowing the ambiguity about ethics to remain unresolved -- and allowing both sides of a contradiction to be partly "right" -- may be the only way to retain the relative, contingent, and open interpretation of ethics. However, this kind of truly open discussion of ethics looked very confusing in two of our cases. It is possible that more powerful parties may be better able strategically to use the ambiguity to further their own interests (Eisenberg, 1985). And, it makes it difficult to ground any claim about ethics very forcefully. This study leaves open the interesting question of whether an open-ended, contingent definition or a static broad definition of ethics is more enabling of effective participation and whether either solution is inherently more "ethical." Implications Because of the ambiguity and controversy inherent in the meaning of ethics, a corporate ethics program can also raise some more fundamental issues, such as the legitimacy of the authority structure. Such fundamental questions underlie the tensions over what the scope of the ethics program should be and who has the right to define what ethics does and does not include. For the most part, these deeper questions about who decides what is and is not "ethics" and what is and is not "ethical" remain suppressed. Instead, employees focus on symbolic, surface issues, like who gets to keep frequent flyer miles or who has a privileged parking space and ignore the deeper issue of which this is symbolic. But occasionally the more fundamental issues surface subtly. The ethics program begins with a focus on the "ethics" or legitimacy of certain practices and then occasionally escalates into a questioning of the legitimacy or "fairness" of more fundamental organizational arrangements and what issues are and are not open for participation. More fundamentally, who gets to participate in which decisions and whose 32

interests are represented in a definition of "ethics" underlie all of the more superficial questions about the meaning and implementation of "ethics." This research on corporate ethics is not unlike other organizational research that focuses on the proximate effects of issues that represent potentially deeper and more value laden tensions4. For example, researchers have considered consequences of labor/management grievance arbirtration (e.g., Karambayya and Brett, 1989), organizational deaths (e.g., Sutton, 1983), corporate decline (e.g., Cameron, Sutton, and Whetton, 1988), or the distribution of and stories about organizational perks (e.g., Martin, Feldman, Hatch and Sitkin, 1983). Most research that treats these discrete issues stops short of considering the underlying questions about hierarchy and authority that such conflicts, dislocations, or symbols could provoke. For example, culture research suggests that a company's decision about how to assign (or not to assign) parking spaces and the stories told about these decisions symbolize the organization's values about hierarchy. Yet in this body of research there is seldom discussion about how this subtext might be raised to explicit and provocative discussion about hierarchy and authority within the organization. In contrast, these deeper questions about authority, power, and the legitimacy of various forms and levels of participation have been the focus of critical theorists. However, with few exceptions (e.g., Buroway, 1979; Clegg and Dunkerley; 1980 ), these theorists have given less attention to how everyday issues in organizations can provide the entry point for getting to these deeper concerns. We suggest issues that are controversial in organizations, particularly those that are ambiguous and thus open to multiple interpretations, might provide good entry points into these more fundamental concerns. Ethics is ambiguous and controversial at deep as well as surface levels. The right to decide what is ethical as well as what is "ethics" -- what issues are to be included in and 4 We do not mean to imply that all issues are not value laden; rather, some issues more explicitly than others represent value tensions fundamental to the structure and ideology of an organization. 33

excluded from its domain and on whose interests should these decisions be based -- may themselves be matters of ethics. In this paper we have attempted to suggest this connection by examining how the introduction of an ethics program provokes questions about the scope, content, and legitimacy of employee participation that direct attempts at studying participation or hierarchy may not have captured. 34

Appendix 1 The Defense Industry Initiative (DI) All companies that signed the DII "pledged to abide by six principles": 1) Written code of business ethics. 2) Employees' ethical responsibilities. 3) Corporate responsibility to employees 4) Corporate responsibility to the government 5) Corporate responsibility of the defense industry 6) Public accountability. The 18 point questionnaire was designed to assess fulfillment of these principles. Public auditors determined the Yes or No answer to each question for each company. 1) Does the company have a written code of business ethics and conduct? 2) Is the code distributed to all employees principally involved in defense work? 3) Are new employees provided any orientation to the code? 4) Does the code assign responsibility to operating management and others for compliance with the code? 5) Does the company conduct employee training programs regarding the code? 6) Does the code address standards that govern the conduct of employees in their dealings with suppliers, consultants and customers? 7) Is there a corporate review board, ombudsman, corporate compliance or ethics office or similar mechanism for employees to report suspected violations to someone other than their direct supervisor, if necessary? 8) Does the mechanism employed protect the confidentiality of employee reports? 35

9) Is there an appropriate mechanism to follow up on reports of suspected violations to determine what occurred, who was responsible, and recommended corrective and other actions? 10) Is there an appropriate mechanism for letting employees know the result of any followup into their reported charges? 11) Is there an ongoing program of communication to employees, spelling out and reemphasizing their obligations under the code of conduct? 12) What are the specifics of such a program? a) Written communication? b) One on one communication? c) Group meetings? d) Visual aids? e) Others? 13) Does the company have a procedure for voluntarily reporting violations of federal procurement laws to appropriate governmental agencies? 14) Is implementation of the code's provisions one of the standards by which all levels of supervision are expected to bemeasured in their performance? 15) Is there a program to monitor on a continuing basis adherence to the code of conduct and compliance with federal procurement laws? 16) Does the company participate in the industry's "Best Practices Forum"? 17) Are periodic reports on adherence to the principles made to the company's Board of Directors or to its audit or other appropriate committee? 18) Are the company's independent public accountants or a similar independent organization required to comment to the Board of Directors or a committee thereof on the efficacy of the company's internal procedures for implementing the company's code of conduct? 36

References Bies, R. J. (1989) The predicament of injustice: The management of moral outrage. In L.L. Cummings & B.M. Staw (Eds.), Research in Organizational Behavior (Vol. 9). Greenwich, CT: JAI Press. Buroway, M. (1979) Manufacturing Consent. Chicago: University of Chicago Press. Cameron, K., R.I. Sutton, & D.A. Whetten. (1988) Organizational Decline: Conceptual, Empirical, and Normative Foundations. Cambridge: Ballinger. Clegg, S. & D. Dunkerley (1980) Organization, Class, and Control. London: Routledge & Keegan Paul. Eisenberg, E.M. (1984) Ambiguity as strategy in organizational communication. Communication Monographs 51: 227-242. Eisenhardt, K.M. (1989) Building theories from case study research. Academy of Management Review 14: 532-550. Glaser, B.G. & A.L. Strauss (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine. Kahn, R.C., D.M. Wolfe, R.P. Quinn, J.D. Snoeke & R.A. Rosenthal (1964) Organizational stress: Studies in Role Conflict and Ambiguity. New York:Wiley. Karambayya, R. and J.M. Brett. (1989) Managers handling disputes: Third-party roles and perceptions of fairness. Academy of Management Journal., 32: 687-704. March, J.G. & J.P. Olsen (1976) Organizational choice under ambiguity. In J.G. March & J.P. Olsen (Eds.), Ambiguity and Choice in Organizations. Bergen, Norway: Universitetsforlaget. Martin, J., M. Feldman, M.J. Hatch, & S.B. Sitkin. (1983) The uniqueness paradox in organizational stories. Administrative Science Quarterly, 28: 438-453. Martin, J. & D. Meyerson (1988) Organizational cultures and the denial, channeling,] and acknowledgment of ambiguity. In L.R. Pondy, R.J. Boland & H. Thomas (Eds.), Managing Ambiguity and Change. New York: Wiley. Meyerson, D. (1989) The Social Construction of Ambiguity and Burnout: A Study of Hospital Social Workers. Unpublished doctoral dissertation, Stanford University, Stanford, CA. Meyerson, D. & M. Scully. (1988) A justification in time breeds nine: The case of corporate ethics programs in the defense industry. Paper presented as part of symposium, The Creation of False Peace: On the Strategic Use of Explanations to Reconcile Irreconcilable Differences," R. Bies & S. Sitkin (Chairs.) at the Annual Meetings of sthe Academy of Management, Anaheim, August. Miles, M.B. & A.M. Huberman (1984) Qualitative Data Analysis. Beverly Hills, CA: Sage. 37

Mohr, L.B. (1982) Explaining Organizational Behavior. San Francisco: Jossey-Bass. Weick, K. (1979) The Social Psychology of Organizing. Reading, MA: Addison-Wesley. 38.

Table 1 Interview Sample Alnha RBta Gamma Thrta 7eta R S I " a 6 %^ %a I I I ~ W,.^...v~ -. w Wi I Corporate Senior President CEO Ethics Officer Human Resources Public Relations Internal Relations Trininr. no\ferrn Officers*: x V X X ~ x x xL x Y x x x V X. X 2 X ~ r\ I\ I I X X ~ X Y I, l 1Iu1II CK uAVY IVviu w. Finance/Accounting X X X Internal Audit X Legal X X X Treasurer X Manufacturing 2 Contracting X X Division Sr. Officers or Sr. Manager***: General Mgr. X X Ethics Officer X X X X Human Resources X X 2 Training X X X Finance X Materiel/Contracting X X Manufacturing/Oper. X 2 Division Employees: Middle managers 2 4 Engineers/Scientists 2 2 1 Non-exempts 11 * Officers may have had different titles, but comparable positions. Individuals in corporate sample were most senior officer in the position unless designated otherwise. ** "X" represents one individual unless otherwise designated. *** These individuals were also officers or senior managers. **** CEO and President positions were held by the same individual in this corporation. 39

Table 2 Initial Definitions of the Ethics Programs Program represents a change Program does not represent a change Program does and does not represent a change Rule-based program Value-based program Rule and value based program Add new rules to increase our Rules will enforce and maintain We need to change our ethical standard so we can our "long tradition of integrity" rules and follow our "become more ethical." and maintain our high ethical old rules of conducting standard, business. We need to reevaluate our Our values have always reenforced We need to change our values and change the way we a high standard of ethical conduct. values and reenforce our think in order to behave more existing values. ethically. We need to change the rules of We need to rely on our values and We need to reevaluate our conduct and reevaluate our follow new rules to maintain our our values and rules values to become more ethical. ethical behavior. to become more ethical and to remain ethical as we have always been. 40

Table 3 The Need for a Response: Types of Contradictions Tvne Example 1. Justification 1~ Justification 2 2. Justification * Ethics Policy 3. Justification * Corporate Policy "We need to become more ethical" and "we must maintain our long tradition of integrity." Ethics is about employee conduct,yet we do not handle human resource issues on the ethics hotline. Ethics is about treating each other fairly, yet we close plants without notice. Table 4 Second Order Justifications Response Tvpes 1. Justification 1~ Justification 2 a. Favor one justification over the other. b. Provide overarching justification. 2. Justification ~ Ethics Policy 3. Justification ~ Corporate Policy a. Change b. Change c. Provide a. Change b. Change c. Provide justification. policy overarching justification justification. policy. overarching justification. 41

Figure 1 Responding to the Ambiguity of "Ethics" 1. Denying Ambiguity: A Narrow Definition Curbs Participation Impose a clear, circumscribed definition from the start. This narrow definition may be remain stable or may change. Case 1. Alpha Company Case 2. Beta Company 2: Encompassing Ambiguity: A Broad Definition Provokes Participation Impose a broad definition from the start. This broad definition may be modified as different constituencies respond with their own interpretation of this broad definition. Case 3. Gamma Company 3: Acknowledging Ambiguity: Equivocation over Definition Creates Confusion about Participation Equivocate over whether the definition should be broad or narrow. Continue to equivocate as the definition gets challenged. This ambiguity can either allow fruitful discussion about the meaning and implementation of ethics, or it can result in confusion. Case 4. Theta Company Case 5. Zeta Company 42

Figure 2 Cvcle of Justifications Perceived need for further justifications Perceived need for further justifications Justification of Initial Definition Second Order Ongoing Justification Justifications 43 ib

Figure 3 Neaotiatina "Ethics" in Five Cornorations Stage 1 Initial Definition Stage 2 Need for Response Stage 3 Second order Justifications Stage 4 Ongoing Justifications Narrow (rules) Alpha Beta Broad (values) aramma - D. Zeta Equivocal (rules & values) Theta I

IJ~~~~ ~~~ ~~~ ~~~ ~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ S~~~~ ~~~ ~~~ ~~~ ~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~