Overview
In 1961, a few months after Adolf Eichmann’s trial began, the Yale psychologist Stanley Milgram set out to test whether ordinary people would follow an authority’s orders even when the orders went against their own conscience. The answer, that most people will go much further than anyone predicts, has become one of the best-known results in social psychology. It is usually read as a lesson about obedience, but it also shows how distance from the people we affect, and who we feel accountable to, shape moral behaviour.
The setup
Source: Leaders Eat Last: Why Some Teams Pull Together and Others Don’t
- Volunteers, recruited through a newspaper ad and told the study was about memory and learning, played the “teacher”. The “student” was an actor; a rigged draw made the volunteer believe the roles were assigned by chance
- The teacher sat at a console with thirty switches from 15 to 450 volts, labelled from “Slight Shock” up to “Danger: Severe Shock” and finally a red “XXX”, and was told to shock the student for each wrong answer
- No real shocks were given to the student. The actor first expressed discomfort, then screamed and pleaded to stop
- If the teacher hesitated, a scientist in a white lab coat answered with a fixed series of prods: “Please continue”, “The experiment requires that you continue”, “It is absolutely essential that you continue”, “You have no other choice, you must go on”. After the fourth protest the experiment ended
- Before the study, the scientists predicted that only 2â3 % would go to the end, and that those would be psychopathic
Findings
Source: Leaders Eat Last: Why Some Teams Pull Together and Others Don’t
- Most volunteers were visibly distressed (sweating, shaking), yet a majority carried on to the last switch
- Obedience depended strongly on proximity to the student. When the teacher had to place the student’s hand on the shock plate, about 70 % quit early; with the student in the same room, about 60 % quit; when the student could be neither seen nor heard, only about 35 % quit, so about 65 % went all the way to 450 volts
- Afterwards, the volunteers who obeyed mostly worried about their own responsibility, not the student. Some felt remorse; others blamed the scientist (“just doing as I was told”) or the student (“he deserved it”)
- Almost all who refused felt accountable to a higher authority than the scientist (a religious or moral code), which is a main reason why the experiment is used as an argument for purpose and values in leadership
Distance and abstraction
Source: Leaders Eat Last: Why Some Teams Pull Together and Others Don’t
- The harm we are willing to do rises as the person harmed becomes less visible. Numbers, spreadsheets, reports and chains of command work like the wall between teacher and student: they turn people into abstractions
- Modern versions of “I was only following orders” are “it is above my pay grade”, “within the law” or “shareholder value”. The pattern is not limited to bad people: the volunteers were ordinary, and the effect comes from the situation
- Seeing, hearing or knowing the people affected by a decision is a countermeasure, which is part of why a leader who stays close to the people in their care makes better decisions
- Fear also matters: fear of trouble, of losing a job or of falling in the pecking order pushes out the moral reflex. This is the cortisol side of obedience; where people feel safe (psychological safety) they can more easily say no
Who is the ultimate authority
Source: Leaders Eat Last: Why Some Teams Pull Together and Others Don’t
- The scientist in the lab coat was an authority in the situation, not an ultimate authority. Bosses, customers, shareholders and analysts are the same: they hold power in the moment but do not own the decision
- The volunteers who resisted answered to something above the person giving the order: God, a cause, a vision or a moral code. A strong sense of purpose gives people the courage to do the right thing, even at a short-term cost
- The same applies outside work: parents, teachers, coaches and friends who anchor their choices in values rather than in whoever is applying pressure at the time are more likely to hold the line
Caveats
- The numbers in this page follow Sinek’s retelling, which is simplified. Milgram ran about 20 variations, not four, and obedience varied from 0 % to 92.5 % depending on factors such as proximity, the setting and the authority’s legitimacy. A 2014 meta-analysis (Haslam, Loughnan and Perry) put the average across conditions at about 43 %
- Later work has questioned the experiment: Gina Perry’s research in the Yale archives argues that many participants suspected the deception and that the experimenter often went off script. Harder to settle is what the results show: obedience as such, or willing identification with the experimenter’s cause (science)
- The study is a standard example of ethics problems, and a full replication would not be approved today. A partial replication by Jerry Burger (2009) stopped at 150 volts and found a rate of continuing similar to Milgram’s at that point
- Applying the findings to firms (Sinek estimates 650 of the Fortune 1000 leaders could ignore the human impact of their decisions) is an extrapolation, not a result
Resources
- 2026-11-17 âĻ Leaders Eat Last: Why Some Teams Pull Together and Others Don’t (Simon Sinek, 2014) â Milgram’s setup, findings and the link to abstraction and purpose
- Stanley Milgram, “Behavioral Study of Obedience” (1963) and Obedience to Authority: An Experimental View (1974)