Case Study 2: The Whistleblower

The Situation

Priya has worked as a clinical trials coordinator at Vandermeer Pharmaceuticals for four years. She is competent, well-regarded, and careful. Her job involves organizing and monitoring data from clinical trials for experimental drugs — a role that requires meticulous attention to detail and a clear-eyed relationship to the data.

Three months ago, Vandermeer's flagship product — a new class of blood pressure medication called Veloxathrin — entered Phase III clinical trials. The results so far have looked good: blood pressure control is significantly better than the control group, side effects are within acceptable ranges, and the company has already begun preparing its FDA submission. There is significant financial and reputational pressure to get Veloxathrin approved. The company's stock has risen on the strength of projected approval.

Priya notices something disturbing while auditing data reports. A subset of patients in the trial — roughly 8% of participants, clustered in the highest-dose group — show an elevated rate of cardiac arrhythmia episodes. The signal is subtle; it would not be visible in a summary table and is only apparent when she cross-references individual patient records with adverse event logs. The statistical significance is borderline — the p-value is 0.06, just above the conventional threshold of 0.05.

Priya brings her concern to her supervisor, Dr. Marcus Ellis, the trial's principal investigator. He looks at her analysis, listens to her reasoning, and responds: "Priya, you're looking at a borderline result in a small subgroup. We can't extrapolate from 8% of one trial to the whole population. The overall trial data is clean. If we raised a flag every time we got a p-value of 0.06 in a subgroup analysis, we'd never approve anything. This is within normal variation." He tells her to focus on her assigned work and not to include the finding in the next status report.

Priya goes to HR, noting that the borderline arrhythmia finding feels like something that should at minimum be disclosed to the FDA as part of the full adverse event record. HR speaks to Dr. Ellis, and the response comes back: this has been reviewed by the clinical team, the finding is not statistically significant, and HR is not in a position to second-guess clinical experts.

Priya is now facing a decision. She believes she has discovered something real. She is being dismissed. And she is aware that if she pursues this further — particularly through formal regulatory whistleblower channels — it will almost certainly end her career at Vandermeer and may make her reputation in the industry.

Framework Analysis

Epistemic Injustice: Is This Testimonial Injustice?

The most immediate epistemological issue is what Miranda Fricker calls testimonial injustice: Priya's testimony is being given less credibility than it deserves. But is this an accurate characterization?

To be testimonial injustice in Fricker's sense, the dismissal must be traceable — at least in part — to a credibility deficit arising from Priya's identity. Let's examine what's happening here.

Priya is a clinical trials coordinator, not a principal investigator. Dr. Ellis is a credentialed physician-researcher. The institutional hierarchy grants authority to Dr. Ellis and assigns Priya a subordinate epistemic status. Is this purely about expertise, or is there something else going on? It's worth noting that Priya's role is data management — this is literally her job. The arrhythmia signal she found was in the data she is trained to analyze. Her dismissal cannot be reduced to a simple appeal to expertise, because she is being dismissed precisely in the domain of her expertise.

But the picture is more complex. If Priya is a woman, if she is a person of color, if she is younger than Dr. Ellis — any of these factors may compound the credibility deficit she faces. Institutional hierarchies do not operate in a vacuum; they interact with broader social hierarchies. When an institutional dismissal ("she's not a physician") overlaps with social dismissals ("she's young, she's junior, she's a woman"), the resulting credibility deficit exceeds what expertise hierarchy alone would produce.

Fricker argues that testimonial injustice is an epistemic harm, not just an ethical one. When Priya's testimony about the arrhythmia data is dismissed, the institution fails to track a piece of information that might be clinically significant. If Veloxathrin is approved and the arrhythmia risk is real, patients will be harmed. The epistemic failure and the ethical failure are connected.

Standpoint Epistemology and Institutional Knowledge

Sandra Harding's standpoint epistemology provides additional analysis. Priya occupies a particular position in the institution: close to the data, with direct access to individual patient records, but without the institutional authority of a principal investigator. This standpoint may give her a kind of knowledge that Dr. Ellis does not have — he works with summary reports, while she works with granular records. Her position within the institution gives her access to information that those above her in the hierarchy cannot easily see from their vantage point.

This is a structural version of the standpoint insight: those lower in institutional hierarchies often possess knowledge that those above them lack, precisely because of their position. The dismissal of Priya's finding may represent not just testimonial injustice but a structural epistemic failure — the institution's hierarchy is actively preventing knowledge from flowing upward.

Patricia Hill Collins's "outsider-within" concept is relevant here: Priya may be positioned in such a way that she sees the institution's data practices from a vantage point that both insiders (with full authority) and outsiders (with no access to the data) cannot easily occupy. This position is epistemically productive — it generates a kind of knowledge — but it is also institutionally vulnerable.

Virtue Epistemology: What Virtues and Vices Are on Display?

Let's audit the epistemic character of the actors in this situation.

Priya is demonstrating intellectual thoroughness (careful cross-referencing of data), intellectual courage (bringing a finding that contradicts the institutional narrative), and the beginnings of epistemic justice (taking seriously data that pointed toward patient harm, even when it was inconvenient). Her challenge now is whether she will exercise intellectual integrity and continue to advocate for her finding under social and professional pressure, or whether she will capitulate to an institutionally comfortable silence — which would be a form of epistemic cowardice.

Dr. Ellis may be demonstrating epistemic arrogance — dismissing a finding without adequate engagement with the specific data Priya presented, relying on his institutional authority rather than engaging with the epistemological substance. His statistical argument (the p-value is 0.06, which is above the threshold) is a legitimate scientific point, but it is also true that borderline significance in a subgroup analysis of an important safety signal warrants further investigation, not dismissal. There is also a possible element of motivated reasoning: Dr. Ellis has significant personal and professional investment in the success of Veloxathrin. The finding that threatens the trial is one he may be motivated to dismiss.

The institution demonstrates systemic epistemic vice: the structure that allows a principal investigator to suppress a safety signal, and that gives HR no tools to evaluate clinical claims, is a system designed to prevent certain kinds of knowledge from circulating.

What Should Priya Do — Epistemically and Ethically?

The epistemic and ethical questions are intertwined here.

Epistemically, what is Priya's epistemic obligation? She has found a signal that she believes is real. She has brought it to the appropriate internal channels and been dismissed. The dismissal was not accompanied by a compelling rebuttal of her actual analysis — Dr. Ellis addressed the general problem of subgroup analysis without specifically engaging with the data she presented.

Virtue epistemology suggests that epistemic cowardice would be a serious failure here: remaining silent to preserve professional comfort, accepting an institutional dismissal without adequate grounds, allowing a belief she has good reason to hold (that there is a safety signal worth investigating) to be suppressed by institutional pressure. Intellectual courage and epistemic integrity point toward pursuing the finding through whatever legitimate channels remain.

But epistemic virtue doesn't translate directly into a prescribed action. Whether Priya should use regulatory whistleblower channels depends on factors beyond epistemology: the strength of the signal (borderline p-value is genuinely uncertain), the potential harm (arrhythmia can be fatal, but we don't know this signal is real), the costs to Priya herself, and the available channels.

What philosophy can clearly say: dismissing Priya's concern without genuine engagement with the evidence she found — rather than with her status in the institutional hierarchy — is epistemically unjust. The standard that should govern evaluation of her claim is the quality of her evidence, not the prestige of her job title.

Discussion Questions

1. Is Dr. Ellis's dismissal of Priya's finding necessarily testimonial injustice, or is there a legitimate argument from expertise that justifies deferring to the principal investigator? How would you determine which it is?

2. Priya's p-value was 0.06 — just above the conventional significance threshold of 0.05. The choice of 0.05 as a threshold is itself somewhat arbitrary. What epistemological issues does the conventional significance threshold raise? Should Priya weight her finding differently because it barely misses statistical significance?

3. Apply the concept of "epistemic cowardice" to this case. Who might be guilty of it, and how? What would the opposite virtue — intellectual courage — look like in practice for each person in this scenario?

4. This case involves a potential conflict between institutional epistemic authority (Dr. Ellis's expertise and hierarchical position) and distributed epistemic access (Priya's direct access to the granular data). How should institutions design knowledge systems to prevent the suppression of safety-relevant information?

5. If Priya eventually reports to the FDA and the arrhythmia signal turns out not to be real — if further investigation shows it was statistical noise — was she epistemically justified in reporting it? Does the correctness of her belief affect the justification of her action? (Consider: is justified belief independent of whether the belief is true?)