More than a decade after its publication, Farquharson's Theory of Voting (1969) is still often cited and seems to have established itself as a classic volume, all, no doubt, because of Farquharson's remarkable insight that 'sophisticated' voters would act quite differently from 'sincere' voters. Since students are still struggling through Farquharson's book, it seems advisable to collect in one place all of the errors, misleading statements, and other assorted problems found in this slender text. This list may not be exhaustive, but it should make use of the volume much easier. (The errors noted on the errata sheet distributed by Yale University Press are included here for convenience.)
- Pgs. xi-xii. 'The particular results of Appendix I ... can readily be extended to cover any desired number of either [individuals or alternatives].' Anyone who has worked with Farquharson's reduction procedure knows that this method is not easily applied to situations with any number of persons and alternatives. Fortunately, McKelvey and Niemi (1978) (see also Miller, 1977) define a new procedure for ascertaining sophisticated outcomes under binary voting processes, and Gretlein (forthcoming) shows that this procedure yields the same outcome as Farquharson's method.
Niemi and Frank's (1982) definition of sophisticated voting for the plurality procedure appears to have properties superior to Farquharson's and is clearly easier to work with.
If one wishes to use Farquharson's procedure, Niemi and Frank show a much simpler layout than Farquharson's three-or-more dimensional matrices.
-
P. 14, fn 8. Entries under col. FFF should be, from top to bottom: 00001100.
-
P. 33. Add italicized portions to this sentence: 'If there is one subset which is weakly separated from every other subset, and if all elements of all * Points 2, 16, 22, and 30 were pointed out to me by Dr.