Dear Lotfi, > RP is a not a contrived or isolated example. On the > contrary, it is representative of many situations in which causality is > a factor. Its deep structure, shorn of surface details, is the > following. I activate an event AO and observe a consquent event, B. My > question is, did AO cause B and, if so, to what degree?
This is not my main point, but "To what degree" is a question which was not in the original problem statement. Different theories will have different definitions of degree, which means that such a question does not have a unique, objective answer until you clarify which notion of degree is meant. In fact, I assumed you had deliberately omitted the notion of degree from the question for this reason. Should we take it that you mean to include this in the problem statement? If so, do you have a specific interpretation in mind, or is part of the question that a proposed solution has to define what is meant by degree? > I performed a > single experiment and have a single datapoint (AO,B). What is obvious is > that given just one datapoint ,no theory could be expected to answer the > question. Exactly. What you are saying here is that Mark Hopkins' answer is correct, if the problem is to be regarded as fully stated in its present form. > I have partial information about N. In this case, please state the partial information. The alternative is for a person answering the question to state assumptions about what this information is before answering the question. However, you could then simply reply that you disagree with the assumptions, which would defeat the purpose of the debate. My main point is that, until you have specified the partial information, you have not finished stating the question. > Based on my partial information about N, I may be able to > arrive at a conclusion like: it is likely that the 20% increase in > advertising played a significant role in increasing my sales by l0%. Certainly. In fact, once you have stated the problem, existing theories will do better than this: they will not only tell you _whether_ it is likely, but exactly _how_ likely it is (for a given definition of "significant role") or even what the best estimate is for the degree of causation. > The real question, then, is: can a theory of causality come up with > answers in this spirit? Clearly, most of the participants in this discussion believe the answer is yes. The reason I wrote my original message is that I get the impression list members are burning for you to finish stating the problem so that it can be answered. regards, Konrad
