Abstract
Selection rules in the form of conservation laws are sometimes
postulated in order to explain the non-observation of certain
reactions in elementary particle physics. One way to devise selection
rules is by induction: data on observed and non-observed reactions are
used to induce phenomenological selection rules directly from
experience. This article describes a systematic procedure to carry
out direct induction of selection rules and quantum properties from
reaction data. The procedure has been implemented as a computer
program and applied to historical data on the discovery of
strangeness, with interesting results. Experience with the program
also led to the conjecture and proof that, given any reaction data
whatsoever that can be accounted for by selection rules of
conservation, exactly one selection rule is sufficient, unless
additional constraints are stipulated.
full paper