Volunteers kept increasing the shocks they thought they were giving

The switch was real; the shock was not

Stanley Milgram's 1963 paper described volunteers being pressed to administer increasingly severe electric shocks to another person. The person receiving the supposed punishment was part of the experiment. The shocks were simulated, but the volunteers' conflict and distress were central observations.

In the reported condition, 26 of 40 participants continued to the highest switch, labelled 450 volts; 14 stopped earlier. Those numbers are striking. They describe behavior in a particular staged situation, however, not a timeless percentage of humanity willing to kill.

An authority without an ordinary threat

The experimenter pressed participants to continue within what appeared to be a learning study. The apparatus, laboratory and repeated instructions gave the situation an organized structure. The paper also records protests and visible tension among people who continued.

This is why obedience should not be confused with enthusiasm. Continuing an action, believing it is justified and feeling comfortable about it are different things. A person can express distress while complying. Counting only the final switch loses that conflict; counting only the protests can hide the fact that compliance continued.

The denominator matters

Forty participants provide a defined denominator for the reported result. They do not constitute a random sample of every society, circumstance or kind of authority. Changing what participants believe about the situation would also change what their actions mean.

The strongest reading starts with the procedure rather than the famous percentage. What did the participant think the task was? What information was available about the other person's condition? What did the experimenter say when the participant hesitated? An answer about human behavior needs those conditions attached to it.

Stopping is also part of the result

The fourteen participants who refused are not statistical debris around the famous twenty-six. Their behavior establishes that the same staged authority did not produce an identical outcome in everyone. The paper records several points at which participants broke off, rather than a single uniform moment of refusal. A useful interpretation must therefore explain variation as well as compliance. It should also resist describing every person who continued as calmly endorsing what was happening. The observed distress makes that shortcut particularly poor. Read the distribution and the participant remarks together: the result concerns a conflict expressed in different ways, not an absence of moral concern in every obedient participant.

The experiment inside the experiment

Deception made the procedure possible and also made it ethically contentious. A study of pressure placed its own participants under pressure. That is not an incidental irony: the knowledge being sought and the experience imposed on the volunteer were closely connected.

The paper remains useful when treated as evidence about a constructed encounter with authority. Read the results together with the participants' objections and the author's discussion of the setting. The unsettling conclusion is that distress did not reliably interrupt compliance in this condition. It is a narrower claim than a verdict on human nature, and a more precise starting point for asking what allows people to refuse.