Whistleblowing is usually analyzed in terms of balancing the duties the whistleblower has to the employer and to the public. But this approach isolates the whistleblower as the sole responsible actor. In actuality, whistleblowing takes place in a system. The employer, the employee, and outside agencies are played off against each other. Those who attempt to do the directing include, of course, the whistleblower, but also the employer, who will use both their own power and recruit that of others to strengthen their own position.

To help frame the use of power in this case, it will be useful to distinguish various sources of power. Raven (1993) lists seven different types of power, each with some overlap into others. This classification helps us to recognize both the sources of power that actors in the case have and the fact that power can be exercised even by the party that seems objectively "weak" in a situation.

The seven different types of power that Raven (1993) outlines are:

1. Reward. The ability to give or to withhold rewards from someone. Goodearl's placement in the testing operation was a reward to her for her earlier good performance.
2. Coercive. The ability to compel action under threat of punishment. Certainly Hughes used its ability to punish to attempt to influence Goodearl and Ibarra. Goodearl and Ibarra blew the whistle to recruit an outside agent (the Inspector General of the Department of Defense) who had the power to compel action by Hughes.
3. Referent. Power based on people desire to emulate an admired person. Classic examples are religious leaders, but in Hughes, both Donald LaRue and Frank Saia had some referent power because of their long service at Hughes. They were respected employees.
4. Legitimate. Socially sanctioned power, usually held because the person occupies a role that has responsibilities and associated power. Because of its position as the employer, Hughes had power to organize its affairs and to structure the jobs of its employees. The Inspector General had legitmate power because of its establishment by congress.
5. Expert. Power based on expertise. Because of both his experience and his education, Frank Saia had expert power. "The girls" doing the testing on the shop floor knew only their own station's routine, and so had little expert power.
6. Informational. Power based on information to which one has access. One can have informational power without being recognized as an expert in an area. For instance, some of "the girls" had information about how fraud was committed in the testing process (because they witnessed it, and had enough knowledge about the process to recognize it). But in its criminal trial, Hughes attempted to defend LaRue's actions based on his expert knowledge about what chips needed what sort of testing.
7. Connectional. Power based on who one knows. Goodearl and Ibarra were exercising connectional power when they called on the Inspector General to investigate Hughes.

People usually think of power only in the first two senses (reward and coerce). But all the different types of power can be seen operating in the Hughes case, and for each use of power we can ask the question "is this use of power ethical?"

To ask this question appropriately, we need criteria and procedure. The computingcases.org approach is to use relatively straightforward tests such as:

• Harm/Beneficence: Does it do less harm or more good than the alternatives?
• Publicity: Would I want this choice published in the newspaper?
• Reversibility: Would I think was a good choice if I were among those affected by it?
• Code of Ethics: How does this choice stand in relation to the professional ethical standards of my profession?
• Feasibility: Can this solution be implemented given time, technical, economic, legal, and social considerations?

For detailed guidance in using each of these tests, see the General Tools discussion on Ethical Analysis in Cases. Each test can help understand a different facet of the ethical issues in a case, and the systematic use of the tests requires some knowledge of the structure and philosophy behind each test. In this presentation, we will refer to the tests assuming the presentation of them in the document referred to above.

To do each of these tests on each of the uses of power for each actor in this case would be prohibitive here. But we will selectively use some of the tests from the perspective of each major actor in the case, and on several levels of social analysis (individual, group, national).

Individual level

So where did they go wrong? The contract specified many things that needed to be done, and sometimes in what order. But it did not specify, for instance, that the "gross leak" needed to be done at a specific time, and Saia was free to move it around. But when they began omitting tests or falsifying tests, these actions clearly crossed the line. They did not simply go beyond the legitimate power. They actually disregarded the legitimate power (in their obligations in the contract) that others had over them. This clearly fails the reversibility test, since we want others to keep their promises to us, and so we should keep the promises we make to others. It also alerts us to the fact that we are dealing with obligations to respect the relationships and roles we have arranged with each other. So, they were overstepping the bounds of their own legitimate power and doing things that did not respect the legitimate demands of others. They were doing things (changing procedures) that looked a lot like things they had legitimate power to do, but they were imbedded in a web of obligations, and got the balance among them wrong.

In addition to violating their obligations in the contract, they also may well have been putting lives in jeopardy because they were allowing inadequately tested chips to be shipped out to be used in military hardware. This clearly fails the harm/beneficence test. They would have been wrong to skip the tests under most any circumstances, but especially wrong when the skipping could result in unsafe systems and potential loss of life. Goodearl and Ibarra cite their concern about safety as the primary reason for their whistleblowing. There were less upset because LaRue and Saia were merely bending the rules, but were driven to action by the safety implications of those departures from the rules.

LaRue and Saia also used the coercive power that came with their position to punish Goodearl because she was not being what they called a "team player." Surely this fails the reversibility test, and also the publicity and harm/beneficence tests. It is important to recognize why it fails these tests. Every organization wants employees to be team players. This expectation is reasonable, and it may be necessary to sanction or even dismiss those who do not meet it. But the code of ethics test helps us to see how the balance was missed. The ACM Code of Ethics. Section three of that code specifies organizational leadership imperatives: the responsibilities of those who are leaders in organizations that deal with computing. Item 3.1 reads:

ACM member and an organizational leader, I will articulate social responsibilities of members of an organizational unit and encourage full acceptance of those responsibilities.

So, one of the responsibilities of an organizational leader, according to the code, is to make clear to employees their social responsibilities and to support employees in accepting those responsibilities. But LaRue and Saia were only emphasizing loyalty to the organization, to the team. And not only were they not "encourag[ing] full acceptance of [social] responsibilities" they were actively punishing Goodearl and Ibarra for attempting to fulfill their social responsibilities.

Finally, LaRue and Saia used their expert power to claim a right to selectively test chips. On the surface, this seems reasonable and perhaps even laudable. If, in fact, their use of their position as experts on testing to make sure the chips were make the line more efficient or even to simply make more money for Hughes while maintaining the required quality, all would be well. But in fact, they misused their expert power either to claim (falsely) that no harm was being done or to minimize the harm being done by the fraud they were committing. In either case, this clearly fails the publicity test. The publicity test focuses on personal character (what would publicity reveal about the kind of person you were?) and were are thereby alerted to the possibility that their misuse of expert power is wrong because it highlights the lack or a virtue (honesty) or the presence of a vice (greed, perhaps, or cowardice). Since honesty and deception is one of the main ImpactCS categories, we will deal in more detail with this angle of LaRue and Saia's unethical behavior in that section.

Goodearl and Ibarra. In her initial attempts to reform the testing procedure from the inside, Goodearl attempted to use her informational/expert power to convince the organization that it needed to follow the testing protocol. She was stymied in these attempts and punished for her efforts. If we look at the IEEE guidelines for ethical dissent, we can see that Goodearl blundered ahead through the steps of dissent without taking time to make her case. Thus, though morally right, her tactics were flawed. The IEEE guidelines suggest that a careful case be put together, based on the best information, that helps to maximize the goals of both the dissenter and of the organization. Goodearl spent most of her time simply telling other that they had to follow the rules. She was correct, but tactless.

Here the IEEE guidelines and the ethical tests we propose converge. The feasibility test (the last item in our list of ethics tests) ask questions about how best to proceed in attaining the ethically desired goal. Can it be attained at all? If a thing cannot be done, one is usually not help responsible for not achieving it. In the criminal trial, Hughes was found guilty of fraud, but LaRue was exonerated because the jury felt her was put in a position in which he had few options except to do the organization's bidding. One the other hand, one of the jobs of a thoughtful ethical dissenter is to form coalitions with others in the organization who can help to support the dissenting case.

Goodearl in fact attempted to do this by joining forces with Ibarra in quality control. Ibarra has expert power in the organization, and could help to strengthen her case. But although it appeard that Ibarra had power, the organizational reporting schemes required her to route all her complaint through Don LaRue, the very person who was committing the fraud. So, it appears that the organizational setup was one that gave the appearance of independent review, but actually gave no power to the independent reviewers.

For this reason, Goodearl and Ibarra both ended up using connectional power by establishing outside connections with the General Inspector's office. This outside agency had power that was independent of the immediate reporting chain in which Goodearl and Ibarra were stuck. It may have been possible for Goodearl and Ibarra to still stay inside the organization and find some ally outside their immediate reporting chain. In fact, Goodearl attempted to do so by speaking with the Personnel office. But again, the reports were quickly funneled back into the immediate reporting chain, this time to Frank Saia, LaRue's direct supervisor.

Still, it seems apparent that Goodearl (at least) and Ibarra (possibly) were less than strategic in their attempts to find an inside remedy the problems they saw. This is certainly excusable in that they were not professional computer scientists or engineers, and did not have the broad knowledge (or expert power) required to critique the system and play it against itself.

If we approach their decisions from the viewpoint of the ethics tests, we can see that they felt they were helping to reduce harm to military personnel and that this overweighed the harm they might do to themselves and their families or to the organization. They still claim that they would do the whistleblowing over again. But they did pay high personal price.

In terms of the reversibility test, were they treating their supervisors and Hughes with appropriate respect? Surely they were doing things that hurt Hughes and their supervisors, but they could be done in a way that respected their rights. In the criminal trial, it became clear that there were personal animosities between Goodearl and LaRue, and it seemed that they were asking "the girls" on the floor to pick sides in the ensuing fight. So, there is some evidence that Goodearl was not respectful of LaRue. But again, this may be excusable given the sever pressure she was under.

Certain Goodearl and Ibarra are celebrated as virtuous persons because of their whistleblowing, and this make the publicity test easy to pass. An important point to remember here is that Goodearl does not need to be sainted, to be perfect in virtue, in order for us to say that she showed courage. In fact, in the criminal trial, it became clear that Goodearl had a variety of character flaws (e.g. bragging about untrue exploits, making immodest claims about personal achievement) and the defense for Hughes used these flaws to their advantage in attacking her credibility. But when we step back from the case it is easy to say that both she and Ibarra showed extraordinary perseverance and courage in pursuing their whistleblowing claim.

Group Level

In thinking about the ethics of LaRue and Saia's actions we have already talked about Hughes legitimate power to set up a system that achieved their goals. There we mentioned that authoritarian atmosphere that Saia and LaRue maintained in the environmental testing unit. But the atmosphere was spread more widely throughout Hughes. When Goodearl attempted to report incidents to the Personnel office, her complaints were not taken seriously, but immediate funneled back to her superiors.

It seems, then, that although Hughes had legitimate power to structure its work environment so that it could achieve its corporate goals, it failed in its duty to "articulate social responsibilities of members of an organizational unit and encourage full acceptance of those responsibilities" (ACM Ethics Code 3.1). This broader failing worked, in the end, against even Hughes' own goals, since they were found guilty of fraud in a federal criminal court.

We can begin to construct a case here for the ethical responsibilities of organizations to encourage ethical behavior in their employees. We can base this case on the various ethical tests we have been using, as long as were are willing to say that an organization can show virtues, weigh harms and benefits, and show respect.

National Level

Congress certainly used legitimate power in its design of laws to encourage whistleblowing. The original law that supported Goodearl and Ibarra was passed during the civil war to encourage (and reward) those people who would sue, on the behalf of the government, perpetrators of fraud. These laws are under regular revision, and Congress sees the laws as supporting an environment in which employees are encouraged to think about the ethical issues associated with their behavior as agents of organizations.

The Inspector General's office is authorized to use coercive power (through lawsuits and criminal proceedings) to compel cooperation on the part of government contractors. In the Hughes case, most of use think this is a reasonable use of power. But if you go to the Inspector General's reporting hotline web page you will discover that the office has wide discretion about how much they will investigate anny particular allegation. Feasibility is surely one of the issues here – the office can only investigate so many incidents at a time. But in additon there is an issue of when they have enough evidence to justify an investigation or a suit. In Goodearl and Ibarra's case, the two employees were encouraged to collect evidence that the fraud they alleged was actually being perpetrated. The point here is that even the investigative powers of a legitimate authority have limits. Describing those limits takes us outside the scope of this analysis, but is a useful analysis. In attempting this analysis, remember not to reduce the moral and ethical responsibilities of the Inspector General to the legal requirements.

Next Issue