STRATEGY 



advice as for promoting new ideas. Also discussion helps one to 

 gain that clear understanding of the problem, which is so essen- 

 tial. 



Another thing to try when one is up against an impasse is 

 to go back to the beginning and try to find a new Hne of 

 approach by looking at the problem in a different way. It may 

 be possible to collect more data from the field or clinic. Fresh 

 field or clinical observations may also be useful in prompting 

 new ideas. As a result of trying to reduce the problem to an 

 experimental inquiry, the worker may have selected a sterile and 

 erroneous refinement of the problem. When the crude problem 

 is seen again he may select some other aspect for investigation. 

 Sometimes it is possible to resolve the difficulty into simpler 

 components which can be tackled separately. If the difficulty 

 cannot be overcome, perhaps a way around it can be found by 

 using an alternative technical method. It may be helpful to look 

 for analogies between the problem presented and others that have 

 been solved. 



If, after persistent attempts to resolve the difficulty, no advance 

 is being made, it is usually best to drop the problem for a few 

 weeks or months and take up something else, but to think and 

 talk about it occasionally. A new idea may arise or a new devel- 

 opment in other fields may occur which enable the problem to be 

 taken up again. If nothing fresh turns up, the problem will have 

 to be abandoned as being insoluble in the present state of know- 

 ledge in related fields. It is, however, a serious fault in a research 

 worker to be too ready to drop problems as soon as he encoun- 

 ters a difficulty or gets seized by enthusiasm for another line of 

 work. Generally speaking one should make every effort to com- 

 plete an investigation once it has been started. The worker who 

 repeatedly changes his problem to chase his newest bright idea 

 is usually ineffectual. 



As soon as a piece of work is nearing completion it should be 

 written up as for publication. It is important to do this before 

 the work has been brought to a close because frequently one 

 finds gaps or weak points which can be remedied while the 

 materials are still at hand. Even when the work is not nearing 

 completion, it is as well to write up an investigation at least 

 once a year, because otherwise when one writes up work from 



135 



