III: The Battleground · Chapter 6

Discrimination

Discrimination hides behind neutral language. Your job is to decode it.

9 min read||

What You Will Learn

The three essential elements of a discrimination claim: a protected class, an adverse employment action, and a connection between the two.
How to prove a case without a "smoking gun" using circumstantial evidence to show the employer's stated reason is a lie (pretext).
The three-step "seesaw" of proof, called the burden-shifting framework, that courts use to analyze discrimination claims.

The 30-Second Answer

To prove discrimination, you must show you are in a protected class (e.g., race, age 40+), you suffered a negative job action (like being fired), and there is a link between the two. You do not need a confession. You can win by building a wall of circumstantial evidence (like patterns of bias or better treatment for others) so high that the employer's stated reason for their action looks like a lie.

The Story

Maria Johnson could not let it go. The phrase "fresh perspective" echoed in her mind. After twenty-two years of stellar performance reviews, being replaced by someone twenty years younger felt like more than just a "business decision." She started digging. Using LinkedIn, she looked up other managers who had been "restructured" out of the company since the new management took over. A pattern began to emerge. The head of accounting, 58 years old, replaced by a 32-year-old. The head of sales, 61, replaced by a 35-year-old. The director of operations, 55, replaced by a 29-year-old. In every case, a long-tenured employee over 50 was let go, and their replacement was significantly younger. Maria created a simple spreadsheet, listing the name, age, and tenure of each person who was fired, and the name and approximate age of their replacement. The pattern was undeniable. This was not about a "fresh perspective." This was a systematic purge of older workers. Armed with this spreadsheet, Maria realized she did not just have a feeling anymore. She had evidence. Discrimination hides behind neutral language. Your job is to decode it.

The Action Plan

1Confirm Your Protected Class. Are you over 40? Are you a member of a specific race, religion, or gender? Clearly identify the basis for your claim.
2Define the Adverse Action. What exactly happened to you? Were you fired? Demoted? Denied a promotion? Be specific about the negative action.
3Become a Pattern-Seeker. Like Maria, look for evidence of a pattern. Who else has been fired or demoted? What do they have in common? Who was hired or promoted instead? Create a simple spreadsheet to track this data.
4Hunt for "Comparators." Find a specific person who is not in your protected class but was in a similar situation and was treated better. This is your "comparator," and they are the most powerful evidence you can have.
5Document Suspicious Comments. Write down every "joke," "stray remark," or "code word" related to your protected class. A single comment might not be enough, but a pattern of them can be powerful evidence.

The Details

The Three Elements of Discrimination

At its core, a claim of Disparate Treatment, the most common type of discrimination, boils down to three simple elements. You have to prove three things:

You are a member of a protected class.

You suffered an Adverse Employment Action. There is a Causal Connection between your protected class and the adverse action. The Burden-Shifting Framework: The Seesaw of Proof Since employers rarely admit to discrimination, courts have developed a three-step process to analyze these cases, known as the McDonnell Douglas burden-shifting framework. Think of it as a seesaw. Step 1: You (the Employee) Go First. First, you must present a basic (prima facie) case. You show that you are in a protected class, you were qualified for your job, you were fired, and someone outside your protected class was treated better. The seesaw tilts in your favor. Step 2: The Employer Responds. The burden shifts to the employer. They must provide a legitimate, nondiscriminatory reason for their action. They will say, "We fired her because of poor performance," or "We eliminated his position in a restructuring." They do not have to prove it is true at this stage, just state a legal reason. The seesaw tilts back to their side. Step 3: You Prove It Was a Lie. The burden shifts back to you. You must now prove that the reason the employer gave was a lie, a pretext for discrimination. You can do this by showing that the reason is factually wrong, that it was not applied equally to everyone, or that the company has a pattern of bias. This is where Maria's spreadsheet is so powerful. The company will say they fired her as part of a restructuring. She will then show that the "restructuring" only seemed to affect older workers. This makes their reason look like a pretext for age discrimination. THE TOOLKIT The Toolkit EVIDENCE WHAT IT IS EXAMPLE TYPE Direct Evidence A "smoking gun" statement that directly proves discrimination. "We need to get rid of the old guys around here." Circumstantial Evidence Evidence that suggests discrimination without directly proving it. A pattern of firing older workers; promoting a less-qualified younger worker. Comparator Evidence Showing that a similarly situated employee outside your protected class was treated better. A white employee who made the same mistake as you was not fired. Pattern and Practice Statistical evidence showing a pattern of discrimination. A spreadsheet showing that 90% of recent hires are under 30. Remember: P.R.E.T.E.X.T. A mnemonic for proving the employer's stated reason is a lie: LETTER STANDS THE ACTION FOR P Pattern Is there a pattern of bias against people in my protected class? R Remarks Have managers made biased comments or "jokes"? E Experience Was the person who got the job less qualified than me? T Timing Did the adverse action happen right after I revealed my age, pregnancy, etc.? E Equal Application Are the company's rules applied equally to everyone? X Explanation Has the company's explanation for their decision changed over time? T Truth Is the company's stated reason factually untrue?

Five Real-World Scenarios

Scenario 1: The Shifting Explanation. A company first tells an employee she is being laid off due to a restructuring. Later, in a legal filing, they claim she was fired for poor performance. The changing story makes their reason look like a pretext.

Scenario 2: The Lone Exception. A company has a strict policy against personal use of company computers. A Black employee is fired for one violation. A white employee who committed the same violation multiple times was only given a warning. This kind of disparity is powerful comparator evidence that may support a discrimination claim.

Scenario 3: The "Stray Remark." A manager once made an off-hand comment about a female employee's "aggressive" communication style. While a single "stray remark" is usually not enough to prove a case, it can be used as part of a larger wall of circumstantial evidence.

Scenario 4: The Perfect Pretext. A company fires a 58-year-old employee for being late three times in one month. The company has a clear, written policy about tardiness, and they have consistently fired every other employee, regardless of age, who has violated it. While the employee is in a protected class, the company has a legitimate, non-discriminatory reason that it has applied consistently. This is a difficult case to win.

Scenario 5: The Statistical Case. A large company lays off 100 employees. An analysis shows that 80% of those laid off were over the age of 50, even though they only made up 30% of the workforce. This statistical evidence can be used to show a pattern and practice of age discrimination. DECODER RING: Pretext A false reason given to justify an action. In discrimination law, it means the legitimate-sounding reason the employer gives for firing you (like "poor performance") is just a cover story for the real, illegal reason (like your age or race). THE ALGORITHMIC SHADOW The Algorithmic Shadow When an algorithm is the problem, the legal process takes a different turn. Discovery is not just about emails and memos; it is about getting access to the algorithm itself. This is a hard-fought battle, as employers and their AI vendors will fight relentlessly to protect their proprietary systems as trade secrets. But the stakes are too high to accept algorithmic opacity as a defense, especially as these tools now cast a shadow over your entire career trajectory, not just the hiring decision. The landscape of liability is expanding. In a landmark 2024 decision, Mobley v. Workday, a federal court ruled that an AI vendor could be sued as an agent of the employers who used its allegedly discriminatory screening tools. The court's reasoning was a direct shot across the bow of the tech industry: "Drawing an artificial distinction between software decisionmakers and human decisionmakers would potentially gut anti-discrimination laws in the modern era." This precedent means that the creators of biased AI can no longer hide behind their corporate customers, opening up a critical new front in the fight for algorithmic accountability. This fight extends far beyond hiring. AI is now deeply embedded in managing careers and reputations. Predictive analytics are used to flag employees as "high-risk" for attrition, potentially stalling their careers. AI-driven surveillance tools monitor every keystroke and conversation, creating opaque "productivity scores" that can lead to discipline or termination. The EU's groundbreaking AI Act, for instance, directly bans the use of emotion recognition technology in the workplace, recognizing its profound threat to worker dignity. Similarly, Colorado's AI Act (SB 24-205) became the first comprehensive state law in the U.S. to impose affirmative duties on companies to prevent algorithmic discrimination in employment, signaling a new era of regulatory scrutiny. Statistical evidence remains the backbone of these cases. The EEOC's four-fifths rule provides a clear, quantitative framework for demonstrating disparate impact. If an AI tool selects 60% of male applicants but only 40% of female applicants, the selection rate for women (40/60 = 67%) falls well below the 80% threshold, creating a strong presumption of adverse impact. But proving the case requires more than just the numbers. It requires a legal team that can depose the vendor's engineers, demand the system's training data and validation studies, and work with data scientists to dissect the code and expose its biases. The central question is no longer just what the decision was, but why the algorithm made it. Your career and reputation can be silently shaped, or shattered, by these automated systems. The plaintiff-side insight is this: you must become the auditor of your own algorithmic career path. Document every interaction with an automated system, from application portals to performance reviews. Question the basis of any automated decision that feels unfair or inexplicable. The law is catching up, but it is the courage of individual workers challenging these black box systems that will ultimately ensure that technology serves humanity, not the other way around. CLOCKED IN Clocked In Build a Wall of Evidence: You do not need a smoking gun. You can win by building a wall of circumstantial evidence so high that the employer's story crumbles. Find Your Comparator: The most powerful piece of evidence is a person who is just like you but not in your protected class, who was treated better. Pretext is the Goal: Your job is to prove the employer's stated reason for their action is a lie. In the next chapter, we turn to harassment , a closely related but legally distinct battlefield with its own rules of engagement. Before You Move On ▢ ▢ ▢ ▢ ▢ CH A P T E R 6 A C T I O N C H E C K L I S T I have identified my protected class(es) I have identified the adverse employment action I have looked for at least one comparator (someone outside my protected class treated better) I have documented any suspicious comments, code words, or pattern of bias I have started building my "wall of circumstantial evidence" SECTION SUMMARY This chapter explains how to prove workplace discrimination. To win a case, you must show three things: you belong to a protected class (based on race, age, etc.), you suffered an adverse employment action (like being fired), and there is a connection between the two. You don't need a confession from your employer; you can build a strong case with circumstantial evidence, such as patterns of bias. The goal is to prove that the employer's stated reason for their action is a pretext for discrimination.

Resumen en español

Este capítulo explica cómo probar la discriminación en el lugar de trabajo. Para ganar un caso, debe demostrar tres elementos: que pertenece a un grupo protegido (por motivos de raza, edad, etc.), que fue objeto de una acción laboral adversa (como el despido) y que existe una conexión entre ambas cosas. No es necesaria una confesión de su empleador; se puede ganar un caso construyendo un caso sólido con pruebas circunstanciales, como la existencia de patrones de sesgo. El objetivo es demostrar que la razón que la empresa da para justificar su acción es un pretexto para ocultar la discriminación. Aprenderá a identificar patrones, a encontrar empleados en situación similar (compañeros de trabajo que recibieron un trato más favorable) y a documentar comentarios sospechosos para fundamentar su demanda.

Educational publication

This chapter provides general information, not advice about a specific situation. Do not submit confidential facts through the book reader.