My summary was maybe too short. If you can defer the election, only apply the first part of 3.� which produces a ranking, thus a winner, for possible advices. If the result is impossible do not use the latest part of 3.� , just defer the election.
If you cannot defer the election, apply 3.� integrally. Condorcet supposed it would lead to a winner either the obtained advice was possible or impossible... The following part of 4.� is just another way to explain what is a possible advice. Repetition to make sure others understood was very common a that time (to my knowledge). 4.� In the case when (where) one will not have to elect, and when one could defer (the election), one will examine the probability of regrouped advices that give the preference to A, to B, to C, et al. (and company) and one will admit the election only when it results in favor of one of the Candidates with a probability greater than 1/2; this cannot take place in the case when the result of votes (voices) leads to one of the 2^(n*(n-1)/2)-n*(n-1)*...*2 impossible (absurd) advices, and takes place in the case of the n*(n-1)*...*2 other advices, when each of the n-1 propositions A > B, A > C, et al. that form essentially the advice in favor of A, for example, are those that regroup the most votes; Markus Schulze a �crit : > Dear Steph, > > you wrote (13 Feb 2003): > > In summary: defer the election if you obtain an impossible > > advice. Use method 3.� only if you really need a result. > > Are you sure that he suggests to use method 3.� and not > method 4.� in these situations? > > Markus Schulze > > ---- > For more information about this list (subscribe, unsubscribe, FAQ, etc), > please see http://www.eskimo.com/~robla/em ---- For more information about this list (subscribe, unsubscribe, FAQ, etc), please see http://www.eskimo.com/~robla/em
