REASON 



hypothesis with the fewest assumptions is to be preferred. (This 

 is known as the maxim of parsimony, or " Occam's Razor ". It 

 was propounded by William of Occam in the fourteenth century.) 



How easy it is for unverified assumptions to creep into our 

 reasoning unnoticed ! They are often introduced by expressions 

 such as "obviously", "of course", "surely". I would have 

 thought that it was a fairly safe assumption that well-fed animals 

 live longer on the average that underfed ones, but in recent 

 experiments mice whose diet was restricted to a point where their 

 growth rate was below normal Uved much longer than mice 

 allowed to eat as much as they wished. 



Having arrived at a clear understanding of the basis from 

 which we start, at every step in our reasoning it is essential to 

 pause and consider whether all conceivable alternatives have been 

 taken into account. The degree of uncertainty or supposition is 

 usually greatly magnified at each step. 



It is important not to confuse facts with their interpretations, 

 -tbatJs to say, to distinguish between data and generalisations. 

 Facts are particular observational data relating to the past or 

 present. To take an obvious illustration : it may be a fact that 

 when a certain drug was administered to rabbits it killed them, 

 but to say that the drug is poisonous for rabbits is not a statement 

 of a fact but a generalisation or law arrived at by induction. The 

 change from the past tense to the present usually involves stepping 

 from the facts to the induction. It is a step which must often be 

 taken but only with an understanding of what one is doing. 

 Confusion may also arise from the way in which the results are 

 interpreted : strictly the facts arising from experiments can only 

 be described by a precise statement of what occurred. Often in 

 describing an experiment we interpret the results into other terms, 

 perhaps without realising we are departing from a statement of 

 the facts. 



A difficulty we are always up against is that we have to argue 

 from past and present to the future. Science, to be of value, must 

 predict. We have to reason from data obtained in the past by 

 experiment and observation, and plan accordingly for the future. 

 This presents special difficulties in biology because, owing to the 

 incompleteness of our knowledge, we can seldom be sure that 

 changed circumstances in the future may not influence the results. 



87 



