This essay has been on appreciative theorizing. It is my conviction that it is important to get appreciative theory reasonably well worked out, before one gets in the business of building formal theory. Otherwise there are few restraints preventing formal modelling from going amuck, and little that pulls the enterprise towards being about real phenomena. On the other hand, given the existence of a reasonable well worked out appreciative theory, formal theorizing can be a very helpful part of the intellectual enterprise.
There is centrally, the problem that the informality of appreciative theorizing makes it difficult to check out the logical completeness and correctness of the 'causal' arguments in that theory. The exercise and discipline of formalizing the arguments can reveal a lot about what is incomplete or problematic in the appreciative causal story. As someone who has played this game a number of times, I can attest that much is learned even before a formal model is fully developed and capable of serving as an analytic engine .....
Source: Richard Nelson in dosi_ea_1998, p. 332