Magistrates let accused people know that confession, repentance, and naming someone else could change what happened next.
The Salem proceedings created a brutal incentive. Massachusetts court historians note that magistrates let accused witches know they could avoid execution by confessing, repenting, and placing blame on someone else. During the core 1692 outbreak, confessors were not among the nineteen people hanged, while people who maintained their innocence were among those executed.
That pattern is safer to state than a single official theology explaining every decision. Religious ideas about repentance mattered, but the surviving record most clearly shows the practical result: a confession could postpone or avert execution, while an unyielding denial did not provide the same protection.
Confessions also generated names. Accused people who confessed were questioned about accomplices and other supposed witches, and those names could become new accusations. Tituba's early confession helped establish that pattern: she described other people and supernatural figures, survived the executions of 1692, and remained imprisoned until her jail fees were eventually paid.
This does not mean every confession was a calculated lie, or that every confessor understood the system the same way. Some may have believed what they said; others were frightened, pressured, or trying to survive. What the procedure did was reward testimony that confirmed and expanded the court's premise.
Modern plea bargaining operates inside a very different legal system, but it offers a limited structural comparison: the outcome attached to admitting guilt can be materially different from the outcome attached to contesting a charge at trial. That difference does not make Salem confession and a modern plea bargain equivalent. It does show why incentives matter when a system asks an accused person to choose between maintaining innocence and reducing danger.
The system rewarded the answer that kept producing more cases.