ABSTRACT
In a sense the previous chapter has brought us full circle, by coming back to a problem that was already raising its head when Weber published De Tactu in 1834. A basic law, one of the few simple results in sensory processes, breaks down at extremes, and the extremes are readily encountered in a laboratory with versatile apparatus. There are four choices in science when this happens; make the theory more complicated, or scrap it (which needs courage), or start from another theory and show that odd things happen under limiting conditions, or write off the deviations as observational error. The fourth is quite popular, but here the third is to be preferred. The problem is, determining what constitutes limiting conditions, and how to represent them without proliferating parameters until the degrees of freedom almost match the number of anomalies originally found, as was noted at the end of chapter 11.
