The prerequisite that was not one

A course listed a prerequisite that had been added in 1998 and had not been reviewed since. Removing it had no effect on results and raised enrolment by a third.

Prerequisites exist for a defensible reason: some material genuinely cannot be understood without earlier material, and allowing students to enter unprepared harms them more than exclusion does. The difficulty is that a prerequisite is cheap to add and expensive to remove. Adding one is a cautious act that nobody opposes. Removing one requires somebody to assert publicly that students will cope, and to be identifiable if they do not.

The asymmetry produces a predictable drift. Over two decades a syllabus accumulates requirements, each defensible when introduced, and the set as a whole is never evaluated because evaluating it is nobody's job. Meanwhile the content changes: the topic that needed the earlier course may have been dropped, or may now be taught from the beginning, and the requirement survives as a record of a syllabus that no longer exists.

The review that found this one used a simple method. For each prerequisite, name the specific topic in the course that depends on it, and the page or week where that topic appears. Requirements that cannot be traced to a topic are removed. Those that can be traced to a single topic are considered for a different fix: teach the topic from scratch in an hour, and open the course.

The method generalises to any accumulated set of rules. The question is not whether each rule is reasonable, which it usually is, but whether anyone can point to the thing it protects. Rules whose protected thing has disappeared are the commonest kind of unnecessary rule, and they are invisible to a review that asks only whether each rule makes sense.