A substantial fraction of ordinary people will inflict apparent harm when instructed by a figure of authority.
In the best-known revised-baseline condition, 65% of subjects continued to the maximum shock level, and obedience varied sharply across Milgram's many variations. Archival work by Perry and others shows more experimenter improvisation, and more doubt among subjects about whether the shocks were real, than the published account implies.
In practice
- A finance clerk telephoned by someone claiming to be head office moves the money, and the tone of authority did more than the instruction.
- The finding that survives is narrower and still uncomfortable: proximity to the victim, and the physical presence of the authority, moved compliance enormously.
- Possible explanations
- The explanation Milgram offered is the agentic state, in which a person under instruction attributes responsibility to the authority. Later accounts favour engaged followership, in which subjects continue because they believe in the experiment's purpose.
- In its favour
- The basic result has been repeated in modified and ethically acceptable forms, in several countries, and the variation across his conditions, obedience falling with proximity and rising with institutional prestige, is informative in itself.
- Against it
- Archival work shows more experimenter improvisation and prodding than the published account implies, the reported rates depend on which conditions are cited, and a substantial share of participants said afterwards they had doubted the shocks were real.
- Standing
- Archival work shows more variation and experimenter improvisation than the published account implies.
- Subject
- Psychology & Behavior
- Essay
- A century and a half of measuring the mind