Case Study 2: The Workplace Disagreement

The Situation

Cornerstone Analytics is a data consulting firm with about eighty employees. For the past year, the company has been implementing a "performance-tiered compensation" policy: employees are ranked quarterly into three tiers based on client feedback scores, project delivery metrics, and billable hours. Those in the top tier receive a bonus and preferred project assignment. Those in the bottom tier receive a "performance improvement plan" and, if they remain in the bottom tier for two consecutive quarters, face potential termination.

Two senior consultants have reached an impasse over this policy.

Priya (38, senior consultant, ten years at the company) argues the policy is discriminatory in practice. She has looked at the data: employees with caregiving responsibilities — disproportionately women and parents of young children — are overrepresented in the bottom tier. She has also noticed that certain client sectors (non-profit clients, smaller firms) generate lower satisfaction scores regardless of consultant quality because the clients themselves are less resourced to give timely feedback. She believes the policy systematically disadvantages people based on factors outside their control.

Daniel (42, senior consultant, twelve years at the company) argues the policy is fair. Every employee knew the criteria when they were hired or when the policy was introduced. The company's survival depends on client satisfaction and delivery quality. If certain employees are consistently underperforming on those metrics, the company has an obligation to its clients and its other employees to address that. He does not believe the policy discriminates; he believes it measures what it says it measures.

They have had three conversations about this. Each one has ended in a standoff. The rest of the team is affected: two junior consultants have told Priya privately that they agree with her but are afraid to say so; one senior consultant has told Daniel privately that Priya is "making everything political." The team lead, Hana, has told both of them that they need to resolve this between themselves or she will have to get HR involved.

Analysis

What kind of disagreement is this?

Applying the taxonomy from this chapter, this is not a single type of disagreement — it contains multiple types simultaneously:

A factual disagreement: Is the policy actually producing discriminatory outcomes? This is in principle a resolvable empirical question: look at the data. Who is in the bottom tier? Is there a statistically significant overrepresentation of employees with caregiving responsibilities, or employees assigned to particular client sectors? Priya has begun this analysis; Daniel has not engaged with it. This is a factual claim that deserves factual engagement.

A conceptual disagreement: What does "fair" mean? Daniel uses "fair" to mean procedurally equal — the same criteria apply to everyone. Priya uses "fair" to mean substantively equitable — the criteria do not systematically disadvantage protected groups. These are two different and philosophically well-established concepts of fairness. The debate in political philosophy between formal equality and substantive equality (going back at least to Rawls and to critiques of Rawls) is precisely this debate. Daniel and Priya are not simply disagreeing about a fact; they are applying different frameworks for what fairness requires.

A values disagreement: Even if both the factual and conceptual questions were resolved, there might remain a genuine values conflict: How much weight should be given to individual performance metrics vs. structural barriers? At what point does a "fair" process become unjust because of what it produces? These are genuine values questions, and reasonable, well-informed people disagree about them.

💡 Key Concept: Compound Disagreements Most significant real-world disagreements are compound: they combine factual disputes, conceptual disputes, and values disputes in ways that make them appear simpler than they are. Identifying which type of disagreement is operating at each level of the conversation is essential for making progress — because factual disputes respond to evidence, conceptual disputes respond to definition and clarification, and values disputes respond (if at all) to different approaches entirely.

Applying steelmanning:

Before either party can have a productive conversation, they need to steelman each other:

The steelman of Priya's position: Companies that use performance metrics without examining structural factors embedded in those metrics are producing discriminatory outcomes even without discriminatory intent. If women and caregivers are systematically scoring lower because of factors like client sector assignment (not consultant quality) and reduced availability for extra-hours work (not effort or talent), then the metric is a proxy for a protected characteristic. This is the basis of disparate impact analysis in employment law, and it is a well-established legal and ethical standard. Ignoring the distributional outcome of a "neutral" policy because the policy is procedurally equal is not philosophical neutrality — it is a values choice to privilege formal equality over substantive equity.

The steelman of Daniel's position: A consulting firm depends on client satisfaction and delivery quality. If those metrics are not reliable indicators of performance — if they're contaminated by factors outside consultant control — then the answer is to improve the metrics, not to abandon performance-based evaluation altogether. Moreover, a policy that treats employees who perform well the same as employees who perform less well, regardless of reason, is a form of unfairness to high performers. If caregiving responsibilities are genuinely affecting performance in ways that the firm should accommodate, the appropriate response is a caregiving accommodation policy, not the abandonment of performance measurement.

Notice that both steelmanned positions are stronger than the positions as the parties have actually been presenting them. Each person has been arguing against a weakened version of the other's view.

How to structure a conversation that might actually move:

The first task is separating the three types of disagreement and addressing them in sequence:

Step 1: The factual layer. "Before we talk about what the policy should be, can we agree on what the policy is actually producing? Can we look at the data together — not separately, but together?" This is collaborative empirical inquiry, and it requires genuine openness on Daniel's part to the possibility that the data might show what Priya says it shows. It also requires Priya to accept that the data might not show what she thinks it shows, or might show a more complicated picture.

Step 2: The conceptual layer. "We seem to be using 'fair' differently. Can we try to clarify what we each mean?" This is the philosophical move of defining terms before arguing about applications. The procedural vs. substantive equality distinction is not obscure — it is a distinction that educated people can engage with — and getting it on the table transforms what appears to be a dispute about policy into a recognizable philosophical debate with a literature behind it.

Step 3: The values layer. "Even if we agree on what the data says and on our definitions of fair, we might still disagree about which kind of fairness should take priority here. Can we talk about that as a genuine disagreement, not as one of us being right and the other being wrong?" This framing — acknowledging genuine values disagreement — does several things at once: it reduces the sense of moral judgment (Daniel is not "wrong" for prioritizing formal equality; he is making a different values choice), it opens space for negotiation (if it is a values choice, it can be discussed and potentially compromised on), and it creates the possibility of what Rawls called overlapping consensus — finding policy language that both can live with, for their different reasons.

The philosophical virtues required of both parties:

Priya needs: intellectual courage to present her data analysis rigorously rather than rhetorically; intellectual humility to acknowledge if the data is more complicated than she has presented it; practical wisdom to choose conversations strategically and frame them in ways that are more likely to produce genuine engagement.

Daniel needs: genuine openness to the factual question — the willingness to engage with Priya's data analysis rather than dismissing it; the intellectual honesty to acknowledge if his concept of "fair" is the procedural rather than the substantive version, and to defend that choice rather than pretending the distinction doesn't exist; epistemic humility about structural factors he may not have considered.

Both parties need: the recognition that this is a compound disagreement with multiple layers, and that resolving one layer (the factual) does not automatically resolve the others. Progress on the factual question may still leave the conceptual and values questions intact, and those deserve genuine philosophical engagement rather than dismissal.

⚖️ Philosophical Tension: Advocacy and Inquiry Both Priya and Daniel have a stake in the outcome of this debate. Priya believes the policy is harming colleagues; Daniel believes its elimination would harm the firm's clients and high performers. Genuine philosophical inquiry requires a willingness to follow the argument where it leads, even when it leads somewhere uncomfortable. But both parties are also advocates, not just inquirers. The philosophical question is whether advocacy and inquiry are compatible — and the answer is: yes, but it requires holding the advocacy lightly, remaining genuinely open to revision, and being willing to update your position if the evidence demands it.

The Role of Hana:

The team lead's instruction that they "resolve this between themselves" may be unhelpful if the disagreement is genuine and compound. Some disagreements between reasonable people with genuine stakes in an outcome cannot be resolved by the parties alone — they require a structure, a mediator, or a formal process. Hana might do better to help them structure the conversation: identify the different types of disagreement, agree on a fact-finding process, establish what decision-making authority each party has, and create a process for the values questions that doesn't require one person to simply capitulate to the other.

Conclusion

The workplace disagreement between Priya and Daniel is a useful case because it is genuinely hard. Neither person is obviously wrong. Neither person is engaging in bad faith. They disagree about facts (which are at least in principle resolvable), about concepts (which require philosophical clarification), and about values (which require genuine dialogue rather than debate).

The philosophical tools — the taxonomy of disagreement types, steelmanning, Habermasian communicative ethics, the identification of validity claims — do not dissolve the disagreement. But they change what the disagreement is about and create conditions in which genuine progress is possible. They also give both parties a shared vocabulary for engaging with a problem that has, so far, been producing only heat.

The philosophical virtues — intellectual courage, intellectual humility, and practical wisdom — are what make the application of the tools possible. Frameworks alone are inert. They require the disposition to use them honestly.