New book "Approval Voting" by Guy Ottewell

Guy Ottewell is one of the three modern re-discoverers of the excellent, simple Approval Voting method.
He has just revised his classic book on the topic, which used to be titled “The Arithmetic of Voting”.

I’ve just ordered it, and look forward to seeing the new chapters:

– A selection of examples of the acute need for Approval Voting – elections in which sides were split and voters suffered the “voter’s dilemma,” often resulting in a less-approved winner.
– Countries and times where Approval Voting was actually used, though it didn’t get that name until 1977. There are more and larger instances than I had realized.
– A comparison with the more complicated reform that was rejected in the British referendum of 2011.
– People in America who are campaigning for Approval Voting.
– Some examples of how exit polls reveal the way people would vote if they were allowed to vote by this freer rule.


5c()®£ \/οτ1N9 ι$ βe+tEr †[email protected] αρPr0∨∀|_ v◯t|ηG b3CÅ∪sE S©oRε is |\/|()rE e×ℙ℟ℇ5sɪ∨e

I would argue Approval is likely rediscovered every month or so. When I first got into this I rediscovered it myself before I found any references to cardinal systems.

Perhaps “three people who independently re-discovered and wrote about approval voting” would be more accurate. The point being that many people learned about it via their writings.

It is amazing to me that even in the 1950’s, many people still had this idea that non-ranking methods (perhaps except for FPTP) were somehow unworthy in the voting arena.

This is 100% due to lack of computers. IRV works by passing around ballot and counting. Score requires an actual computation of an average. In the 1950s an adding machine would not give the right number if it was not oiled properly so they were not super trustworthy. Even approval would require the splitting up of ballots to count. This is maybe more susceptible to tampering. Honestly, if I was organizing a vote for a large number of people but only had access to pen and paper I would likely choose IRV. If you read Hare’s original proposal or John Stuart Mill’s summary it is pretty clear that these issues were constraints. Should also note that he did not really propose IRV as we would understand it today.


Ironically, I found Approval in a textbook but then reinvented range voting (with that exact name) as a way to avoid the inexpressiveness of only 2 levels.

FPTP is a ranking method. The candidate with the most first-place votes is elected; all other ranks are ignored.

1 Like

Re the influence of computers: well put.

I’m curious what in particular you’re referring to about Mill’s summary. Hare’s work is referenced on 11 pages of that volume.. And note that Hare’s original proposal was for STV, not IRV. STV has a lot more to recommend it than IRV, I think.

Yes that is the summary which I am referring to. I do not think that Hare’s method was really IRV or STV since he preserved single member district but you could vote in multiple. Maybe the closest modern incarnation is Local PR. Either way his method has some flaws so it is not really worth considering. My point was that both Mill and Hare talk about the practicality of an election in the 1800s more than any concept of computation.

Thanks for sharing this. I’m excited to read it too.

FPTP is a degenerate case of many other systems