U$,
you got it right: when the Republicans are in power, then it’s every man for himself in fighting for a turn to suck up some swill out of the trough of power.
But when they are in opposition, they tend to put differences aside and work together to get back to their favorite position of being able to fight each other again…

Posted by: ralphieboy | Nov 9 2006 7:11 utc | 4

I wonder, has anyone seen a large increase in the paper shredding business in DC since it was announced the Dems took both houses?
Its too early for spring cleaning, but I recall a certain photo a few weeks back headed to Satan’s house. I suppose the company is planning an expansion of its home service department right now.

Posted by: Fiat Lux | Nov 9 2006 9:18 utc | 6

A Come-to-Daddy Moment
MoDo nails it.

Poppy Bush and James Baker gave Sonny the presidency to play with and he broke it. So now they’re taking it back.
Poppy Bush and James Baker gave Sonny the presidency to play with and he broke it. So now they’re taking it back.
In a scene that might be called “Murder on the Oval Express,” Rummy turned up dead with so many knives in him that it’s impossible to say who actually finished off the man billed as Washington’s most skilled infighter. (Poppy? Scowcroft? Baker? Laura? Condi? The Silver Fox? Retired generals? Serving generals? Future generals? Troops returning to Iraq for the umpteenth time without a decent strategy? Democrats? Republicans? Joe Lieberman?)The defense chief got hung out to dry before Saddam got hung.

Times Select is free this week, so click for the rest.
This kind of satire reaches Americans, in my view, and is effective.
Bill Maher on Larry King Live:

KING: So what’s your take on the election?
MAHER: Well, it’s nice to finally win one. I felt like Black people at the O.J. Simpson trial.

Maher and Dowd satirize political events in ways that don’t patronize the public but make it acceptable and cool to be critical. Enough cannot be said in praise of latter-day Mark Twains – Colbert, Stewart, et al., in shifting American public opinion.
At the top of the LKL hour Biden discussed Gates’ role in Iran/contra (most of which I missed) – so that nasty piece of history is out there too.
Can’t believe I dare to hope.

Posted by: Hamburger | Nov 9 2006 11:40 utc | 8

Hamas Leader Calls Off Israel Cease-Fire

Hamas’ exiled leader on Wednesday called off a cease-fire with Israel and militants threatened to attack Americans after 18 members of a family, including eight children, were killed in an Israeli artillery barrage on a densely populated Gaza neighborhood.

Once again the Israelis jumpstart the cycle of violent revenge in “the Holy Land”.
How often do they have to do this before we comatose Americans finally notice that the Israelis want war at any cost?
And that the costs are measured in Palestinian lives and American (deficit) dollars… and lives as well since 9/11?
Forget about “the Democrats growing a spine”.
Do you think that we ordinary Americans might grow enough of a spine to order the Demoplicans to stop funding the murderous Middle Eastern Wars in Iraq and Palestine?

Posted by: John Francis Lee | Nov 9 2006 12:18 utc | 10

Thanks for that at #11 JFL…
Stasi spy chief Markus Wolf dies

“I hoped that after the Nuremberg Trials, there would be a time without war, aggression or crimes against humanity”

Posted by: Anonymous | Nov 9 2006 13:13 utc | 12

new senators Tester (Montana) & Brown (Ohio) seem to have a separate script from the Democratic party establishment.
also, they seem more in tune with regular peeps.
lets hope they keep the spirit.
the Dem establishment should be worried about types like Tester & Brown. These guys seem like they actually do really want to lead.
unlike Hilary & Obama who want to find a “sweet spot” in the political center. And stay there.

Posted by: jony_b_cool | Nov 9 2006 15:39 utc | 14

maybe its not a bad thing for the Dem establishment to continue to remain AWOL on the big issues of the time. It may actually be a very good thing.
because it will have the effect of firing-up inspired Dem candidates who have a feel for both the people & the moment and who are also not intimidated by either party’s bullyism, cashism or weirdism.

Posted by: jony_b_cool | Nov 9 2006 15:59 utc | 16

Agreed, Bea. Today’s Democracy Now! was really good. Robert Parry on there too.
fwiw- Gates and Hamilton have a good old boy connection b/c of their asso. with a university. Hamilton, of course, is always the go to guy for covering up republican malfeasance.
I once wrote to Hamilton asking him to honestly investigate this current mess so that his legacy wouldn’t be one of covering up for the Bush family yet again. LOL. Obviously that made a big impression on him. But that will be his legacy, as he has chosen. He bends over as a matter of course. That should be the full body portrait of him for posterity, in fact.
Fwiw- Pelosi will not go for impeachment, as she said. This is a good move, in fact, because she would be third in line and such a move would be driven into the ground by the repukes. But that doesn’t mean all sorts of investigations into corruption cannot go forward.
also, it would be better to bring criminal charges after these current crooks are out of office. Just remember, Pinochet and his crew are finally being held account. This coincides with a left power surge in Latin America, no doubt made easier because the corporate political complex are busy ruining ppls lives elsewhere.

Posted by: fauxreal | Nov 9 2006 17:46 utc | 18

edwin’s link.

Posted by: beq | Nov 9 2006 17:57 utc | 20

Sic Semper Tyrannis.
But, yeah, of course Rall is right that we aren’t out of the woods when it comes to eluding tyranny in this country. I think, however, that the Bush regime is thoroughly discredited at this point, and without the support of the military, the sort of Pro-Cheney-Admin. coup he envisions is not in the cards. The real fascists are likely to arise not this year, but the year or two after that, if and when the Democrats take no forceful action to avert the multiple disasters that now loom (fiscal collapse, energy crisis, and possible loss of the Army itself that is in Iraq) and the Democrats are themselves in turn discredited by this inaction. The discrediting of the entire two-political party structure, while perhaps richly deserved, could actually serve as the opening for a real tyrant—meaning a strongman promising “stability” and willing to take large-scale and overt advantage of the dictatorial powers Bush has invested in the Executive—to take over.

Posted by: heatkernel | Nov 9 2006 18:42 utc | 22

WaPo takes up Billmon style

Posted by: b | Nov 9 2006 19:22 utc | 24

From Counterpunch (see link):
Pelosi:
“This spring, I was in Israel as part of a congressional trip that also took us to Egypt, Lebanon, Jordan, and Iraq,” said Pelosi. “One of the most powerful experiences was taking a helicopter toward Gaza, over the path of the security fence. We set down in a field that belonged to a local kibbutz. It was a cool but sunny day, and the field was starting to bloom with mustard. Mustard is a crop that grows in California, and it felt at that moment as if I were home. And then we were told that the reason we had to land in that field, as opposed to our actual destination, was because there had been an infiltration that morning, and they weren’t sure how secure the area was. And that point alone brought us back to the daily reality of Israel: even moments of peace and beauty are haunted by the specter of violence.”
(…)
“One thing, however is unchanged,” Pelosi added. “America’s commitment to the safety and security of the State of Israel is unwavering. America and Israel share an unbreakable bond: in peace and war; and in prosperity and in hardship.”
“The greatest threat to Israel’s right to exist, with the prospect of devastating violence, now comes from Iran. For too long, leaders of both political parties in the United States have not done nearly enough to confront the Russians and the Chinese, who have supplied Iran as it has plowed ahead with its nuclear and missile technology.”
link

Posted by: Noirette | Nov 9 2006 19:48 utc | 26

Chavez: Bush should get death penalty

Posted by: annie | Nov 10 2006 1:20 utc | 28

Gaza hit was ‘technical failure’

Israeli PM Ehud Olmert has said an army artillery barrage that killed 18 Palestinians in the Gaza Strip was the result of a “technical failure”.
But military operations against suspected Palestinian militants would continue, he added, admitting that further mistakes “may happen”.

“We’ll Keep Killing” – Ehud Olmert, Israeli Prime Minister

I admit it, I’m parphrasing. Although I’m not off by much. Israel keeps killing Palestinians in Gaza. Eighteen women and children were killed by IDF shells yesterday. Olmert can call their deaths a “technical error” if he wants, but that isn’t how the rest of the world sees it.
Israeli soldiers shot and killed two Palestinian women last week. The women were part of group acting as human shields. They were helping men holed up in a mosque escape from the soldiers. The men had guns and they were shooting, but they didn’t kill any Israelis. The women, the ones who got shot, were unarmed.
By the way, every shot fired, every shell that falls, has Uncle Sam’s name on it. Israel insists on putting a bulls eye on every American’s back.

How Gaza Offends Us All

An opened jaw with yellowed teeth gaped out of its bloodied shroud. The rest of the head parts were wrapped in a plastic bag placed atop the jaw and nostrils, as if to be close to the place to which it once belonged. The bag was red from the pieces that were stuffed inside it. Below the jaw was a human neck slit open midway down: a fleshy, wet wound smiling pink and oozing out from the browned skin around it, the neck that was still linked to the body below it. Above him, in the upper freezer of the morgue lay a dead woman, her red hennaed hair visible for the first time to strange men around her. More red plastic wrapped around an otherwise absent chin. She was dead for demonstrating outside a mosque in Beit Hanoun, northern Gaza where more than 60 men sheltered during the artillery onslaught by Israeli tanks and cannons.
Most of the others still had their faces intact. They lay on their silver morgue trays stiffly as unthawed frozen food. One man had a green Hamas band tied around his head; he looked like a gentle shepherd from some forgotten, pastoral age. Another’s white eyes were partially opened, his face looking out in horror as if he’d died seeing it coming. Then a muddy, grizzled blob on the bottom left tray, black curls tangled and damped into its rounded head and blessedly shut eyes. A closer look revealed a child, a boy of 4: Majed, out playing his important childhood games when death came in like thunder and rolled him up in a million speckles of black mud. The other dead had already been taken away.

Between Resistance and Deception

The Israeli regime unleashes racist brutality that by far outstands the crimes of the previous apartheid regime in South Africa. It imprisons an entire people behind ghetto walls, kills them and submits them to an economic blockade that has brought communities to the verge of starvation. Yet, while exactly 30 years ago the UN General Assembly called for comprehensive sanctions against apartheid in South Africa, Palestinians are reminded on a daily basis that the Zionist Occupation can still count on the blindness of the world to its atrocities and crimes. Until when?

U.S. Democrats re-embrace independent Lieberman

WASHINGTON (Reuters) – U.S. Senate Democrats have begun to re-embrace Joseph Lieberman and seem ready to give the former Democrat turned independent a chairmanship in the new Congress that convenes in January, party aides said on Thursday.
Lieberman won reelection to a fourth Senate term on Tuesday, running as an independent in Connecticut following his loss in a Democratic primary to anti-war activist Ned Lamont.
Many Senate Democrats respected the outcome of the primary and backed Lamont in the general election. But many also said they still consider Lieberman a friend and a Democrat.
Senate Democratic Leader Harry Reid, set to become majority leader of the new Democratic-led Senate, “talked to Lieberman on Wednesday and told the senator he expects him to be the next chairman of the committee” on Homeland Security and Governmental Affairs, said Reid spokesman Jim Manley.

Posted by: John Francis Lee | Nov 10 2006 2:57 utc | 30

I do’nt think an escalation is exactly what the electorate had in mind, last Tuesday.

Posted by: anna missed | Nov 10 2006 9:29 utc | 32

Rumsfeld is on his way out and Ehud Olmert is about to arrive. …
…
Iran will top the agenda. …The military option seems unlikely. A senior American diplomat figured this week that Israel does not have the military capability to attack the nuclear facilities in Iran. “The worst thing would be if you try and don’t succeed,” he said. He did not mention any possibility of America bombing Iran’s facilities, but it’s hard to see Bush taking such a risk at this point. He has enough troubles already with Iraq. And anyway, his intelligence people are telling him that there is still time for maneuvering.
John Negroponte … toured the Middle East last week, marking the first visit by such a senior figure in that community. In talks with his Israeli colleagues, he discussed ways of weakening Tehran’s growing influence. Maybe it would be possible to do something with ethnic minorities like the Azaris and the Kurds, who have ties to communities outside Iran? The Americans aren’t talking about toppling the regime of the ayatollahs, but about exerting pressure that would lead to a change in policy. link

Posted by: jj | Nov 10 2006 10:07 utc | 34

Fade @ #25
I almost choked after reading that…lol priceless!
Very trenchant, but quite on the mark.
Made my moring, thank-you.

Posted by: Uncle $cam | Nov 10 2006 15:05 utc | 36

And right on cue… Al Qaeda gloats over Rumsfeld

Abu Hamza al-Muhajir, also known as Abu Ayyub al-Masri, said in the recording posted on the Internet on Friday that the group had 12,000 armed fighters and 10,000 others waiting to be equipped to fight U.S. troops in Iraq.
“I tell the lame duck (U.S. administration) do not rush to escape as did your defense minister…stay on the battle ground,” he said.
He said his group would not rest until it had blown up the presidential mansion in Washington.
“I swear by God we shall not rest from jihad until we…blow up the filthiest house known as the White House,” the voice on the recording said.

And if you listen very, very carefully to the tape, you can hear George the Younger in the background saying “Now let’s get Karl ready for that new beheading video!”

Posted by: Monolycus | Nov 10 2006 19:59 utc | 38

Creating Consistency
Along with Bolton’s nomination, Bush said he would like to move forward on legislation to retroactively authorize the National Security Agency’s domestic surveillance program. CNN today.
“We also believe that the authorization to use force, which was passed by the Congress in the days following the attacks of September 11, constituted additional authorization for the president to engage in this kind of signal intelligence.” – President’s Radio Address, Dec. 17, 2005
“So, consistent with U.S. law and the Constitution, I authorize the interception of international communications of people with known links to al Qaeda and related terrorist organizations.” – National News Conference, Dec. 19, 2005
“The terrorist surveillance program is both necessary and lawful,” Gonzales said. “Accordingly, the president has done with this lawful authority the only responsible thing: use it.” – Jan. 24, 2006

Posted by: Anonymous | Nov 10 2006 22:23 utc | 40

@John Francis Lee:

Instant Runoff Voting is better than plurality voting (the system used in the U.S. national elections at present), but it isn’t really very good, and isn’t really suited to most elections if you’re doing it right.

Voting is a topic mathematicians like to discuss. (Some of ’em, at least.) And it’s usually easy to come up with examples, so it’s easy to talk about it with non-mathematicians.

There are five obvious mathematical conditions which a good voting system should fulfill. (There are a number of others, actually, but there are five obvious ones. The others start to get a little abstract, and tend to be less likely flaws in a real-world election than the obvious ones.)

The Majority Criterion: if choice A is the first choice on a majority of the ballots, choice A will win.

The Condorcet Winner Criterion: if choice A would beat any other choice in a one-on-one election, then choice A will win.

The Condorcet Loser Criterion: if choice A would lose against any other choice in a one-on-one election, then choice A will lose.

The Pareto Criterion: if choice A is ranked lower than choice B on all ballots, choice A must lose.

The Consistency Criterion: if the ballots are divided into two groups, and choice A wins within both of these groups, then choice A still wins if the ballots are counted as a whole.

A fully-implemented Instant Runoff Voting (IRV from now on) system passes the first four tests (although it fails the last one). However, in a fully-implemented system, there can be no cutoffs. All ballots must rank all the choices. (If there are five candidates, then all five must be listed on all ballots; or, at least, list four, since the fifth would then be implied.) Otherwise, IRV ceases to satisfy anything beyond the Majority Criterion, which puts it on a par with plurality voting, which is what you’re trying to escape. Except where the number of ballots is very small — the low two digits — it is possible to concoct a possible (and even plausible) scenario where a cutoff would lead to a winning choice which not only received the fewest first-choice votes, but also would lose in a one-on-one election against all other choices. (I’m not going to do it now, because I’m sleep-deprived enough to start feeling a little dizzy, but if there’s interest and nobody else does, I’ll write it up tomorrow night.)

This leads to a problem, since in many elections with more than 2 choices, many of the voters do not know enough about the choices to give a serious ranking of all of them. (The problem is even more pronounced in small- to mid-sized groups, since it is often the case that a majority of the voters do not even know the candidates.)

There is, however, another problem with even a fully-implemented IRV system: the winner will not be the “worst” choice (the Condorcet Loser), but it quite possible that IRV will result in the second-worst choice; that is, IRV can result in the selection of a choice which, in one-on-one elections, would only beat the Condorcet Loser.

There is also the famous second-guessing game, which is already rampant, and which is not stopped by IRV: “I prefer A to B, but B seems more likely to win than A, so I will claim on the ballot to rank B before A in order to make sure C (or any other choice) does not win.” This is what kills third-party initiatives in the U.S., and IRV would fail to address the problem.

(IRV is also occasionally vulnerable to sabotage by deliberate spoiler candidates in the middle in a heavily polarized election, although this is rare and requires that the smaller group be more dedicated to their choice.) (Which, of course, is likely to describe the Republicans.)

There is, unfortunately, no system which manages to beat all the possible objections at once. However, there is a system which behaves similarly to IRV in most cases, and often provides more reasonable selections when IRV fails: Approval Voting.

In Approval Voting, voters rank each choice as either “Acceptable” or “Unacceptable”. There is no relative ranking, and a voter may give more than one choice “Acceptable” status. The winner is the choice which receives the most “Acceptable” votes. Under most conditions, this leads to the same selection as IRV, but in certain circumstances, it can result in a winning choice which has majority support, which would not have it with IRV. It also adds the benefit of removing the second-guessing (no rankings, just yes-or-no) and does not require voters to be able to give solid rankings between extremely similar choices, thus allowing similar options to coexist without spoiling.

Sorry for the soapbox. It’s rare that I can bring anything to the discussion, so I always try to go into the subject a bit.

Posted by: The Truth Gets Vicious When You Corner It | Nov 11 2006 8:54 utc | 42

And I yours, mark. A form of instant runoff voting was recently explained and narrowly defeated here in british columbia.
We have a provincial government that owns all seats but one or two, and four years ago held all. The only criticism I’ve heard is that IRV leads to fractured coalition governments.
In Canada we’ve seen great progress with such coalitions, one of which led to our national healthcare system.
To reiterate I am interested in your take on IRV in Ireland, one of the only places I recall using this radical election device.

Posted by: jonku | Nov 11 2006 11:05 utc | 44

Radical?????? Gosh!

Not getting at you – Now that you say it, I can see how it could be seen that way. It’s just that it’s never ocurred to me before that it could be seen that way. There y’are now you’ve made the Gaelic Gorilla scratch his sloping brow in wonderment.

Anyway I’m asked to explain this so often that I’m making this response to you as a file in my cut and paste library here goes:

In Ireland Proportional Representation – Single Transferable Vote (PR-STV) is the method used to elect members of the Dáil, the lower and more powerful chamber of our bi-cameral parliament the Oireachtas.

Under the Irish system constituencies are multi-seated. That is each constituency is represented by more than one TD (Teachta Dála).

To be elected a candidate must reach a quota of votes. The quota is determined by the number of electors for the constituency on the electoral register. The system is designed in such a way that it is mathematically impossible for more candidates to reach the quota than there are seats for the constituency.

Voters rank the candidates in order of their preference. Their ballot is transferred down from higher to lower preference as candidates are eliminated.

Because we have multi-seat constituencies political parties run a number of candidates. Generally they run more candidates than the number of seats they are statistically likely to be win in that constituency.

It can take several counts of ballot papers done in public and under intense scrutiny both by the public and by “tallymen” from the various parties before all the seats for a constituency are filled.

The advantages:

It’s very accurate – candidates are elected on the basis of the levels of support for their party and for them personally in their constituency.

Unlike in the party list system candidates aren’t elected solely on the basis of party policy, or loyalty to a party leader. The candidate must also have personal support within the constituency.

The combination of party loyalty, PR-STV, and multi-seat constituencies has the very desirable effect of making politicians vulnerable to public demands. They cannot afford to alienate their constituencts who can and do make demands on them without a guarantee that they will vote in return.

They also punish politicians who they deem unresponsive to their local duties. More than one party leader has either come close to losing his seat or has actually lost it. On one particularly famous occasion [Frank Closkey] the seat in question was considered to be one of the “safest” in Ireland and the party in question [Labour] had to find a new leader fast.

In short politicians need to be heavily involved with their community involvement to develop the contacts with first with people whose support will help create the personal link between them and their voters and thus with the electorate as a whole.

Let me give a personal example:

My mother was heavily involved in pioneering community run pre-school and primary school montessori method schools. The community montessori pre-school she founded was the first in Ireland. We lived within easy walking distance of a pretty deprived part of Dublin called Ringsend. Local women leaders in the commnunity approached her to help them set up the school run it and see to it that they got trained (and qualified) them in montessori techniques and the management skills for such things as keeping track of the budget, scheduling, etc. They saw no reason why their kids shouldn’t have the same advantages as middle class kids and were prepared to work to get it. Of course she said “yes.”

One of the local Td’s was Garret Fitzgerald. At that time he was Taoiseach (prime minister.) He was summoned to several meetings by those women. Who told him to be helpful or else see a drop in his vote. He wanted his over the surplus votes transferred to other Fine Gael candidates – he damn well did as he was told. This, was very typical of “New Ireland” a community group could and did make demands on a very senior politician, while offering very little in return. ALL they offered was a negative. “If you’re not helpful we’ll tell.”

I like a system where the prime minister and party leaders go in terror of their electorate. Persecuting civil servants might be their job persecuting them is ours.

Criticisms:

It leads to clientilism. This certainly used to be true. It’s far less so now. Mostly this critique is based on Professor Chubb’s work. His founding study was done in 1963 (‘Going about persecuting civil servants: the role of the Irish Parliamentary Representative’), with further work done in 1977, and a slight updating in 1982. When I started doing post-grad stuff Chubb was one of my tutors (“I am the bloody minded son of a Derbyshire coal miner and I like the Irish system because you lot are even more bloody-minded than I am” – you’ve got to like a professor with that attitude,) and was one of the first to point out that his work had been overtaken by changing Irish circumstances.

There are still the trappings of clientilism. But its effectiveness as a vote getter is now minimal throughout all but the most underdeveloped parts of the country. Which, thanks to our access to the EU and the ability that gave us to get out from under British economic domination are becoming thankfully rare.

It Leads to Weak and Fractured Coalition Governments:

Is this supposed to be a bad thing? Excessively strong central governments run by politicians who consider themselves safe from the electorate are the bad thing from my POV.

Our best governments have been coalitions. The fact that they’re elected in a way which represents the level of support for their party both locally and nationally means we get what we voted for. If we don’t like it we know how to change it.

I’m not alone in liking it either. Every single time politicians have tried to change it. It’s been massively rejected in the required referendum which has been followed by the party that proposed it getting the shite kicked out of them in the next election.

It’s also a difficult system to manipulate. Not impossible, but very very difficult. One of the things I really enjoy really really enjoy listening to on election night radio is listening to the chorus of interviews with would-be Cathall O’Róibhs, pollsters, political “consultants,” and “vote managers” bitching and whinging and whining that the “votes are going all over the place.” By which they mean the electorate are doing what the electorate wants and not what the politicians want. Well boo hoo.

What the “vote managers” really hate is that the voter can vote in such a way as to STOP a candidate from getting elected. How you do it is either by stopping transferability before you get to that candidate on the ballot paper. OR by transferring away from their party. That’s what I do. My personal politics are conservative but I vote tactically as follows:

The Party I support a moderate conservative party called Fine Gael. They get No:1 and maybe No: 2.but I vote in a 5 seater constituency in five seaters it’s the transfers that count so:

Labour get No: 3

Workers Party get No: 4

The Greens get No: 5

(The green pols get elected on the last count and scrape in on transfers without reaching the quota.)

That helps keep the conservatives honest. It also helps see to it that they don’t go charging off in the direction of libertarians (yuk!) / weird right wingers (double yuk!) / or towards various assorted loolahs such as “Mr. Abbey of the Holy Cross Fitzsimons” (oh please …….)

I then vote “all over the place” in such a way as to be absolutely sure that there isn’t a hope in hell that my vote will go the nearest thing we have to the neo-cons a party called the Regressive Autocrats oops sorry I meant to type “Progressive Democrats.” My bad. Or to the “natural party of government” a bunch of shaggers called “Fianna Fáil” Have I wasted my vote? No I haven’t 5 out 5 politicians representing my political priorities ain’t bad. Lots of Irish people vote this way. It comes of our being blood-minded. 🙂

If you’re interested in the mechanics Wikipedia have a good article here. The “go to” guy for all of this in an Irish context is Frank Litton in the Institute of Public Administration. He’s a helpful chap and does reply to letters. 🙂

Posted by: markfromireland | Nov 12 2006 0:23 utc | 46

@John Francis Lee, #45

Here is a single example of how IRV can fail to work the way you expect. Suppose that, tomorrow, the citizens of the U.S. were offered a ballot saying “What should be done about Iraq? A. Withdraw B. Continue as now C. Nuke every square inch of the place until it glows”. It is plausible (although not necessarily true) that this would break down as follows: 45% of Americans would like to get out, but would prefer the current chaos (which at least leaves some people alive) to nuking them all. 40% would like to nuke the country and be done with it (so they could declare “victory”), but if they can’t have that, they would just want to get out. The remaining 15% has bought into the anti-terrorist shtick, and can’t conceive withdrawal, although they don’t want to drop nukes. In the first round, the count would be:

A: 45%
B: 15%
C: 40%

Since no choice has a majority, you eliminate B and redistribute the votes. So the instant runoff recount goes:

A: 45%
C: 55%

Therefore, this IRV ballot would lead to the nuking of Iraq, even though the most popular first choice (winner in a plurality system) was the opposite one and the majority of the voters would prefer option B instead, however reluctantly. (Note that this does not violate the Pareto Criterion, because C was not ranked lower than B on all ballots, just on a majority.)

If that seems a bit loaded, because it’s such a naked matter of life and death, you can rewrite it to use other polarized issues, like nationalized healthcare or immigration. (And you can devise “better” failures; this was just the first simple one I came up with tonight.) The point is that IRV has nearly as much opportunity for seemingly inexplicable results as plurality voting. Yes, it would have kept Bush out of office, but that doesn’t make it the be-all end-all of voting.

Note, by the way, that Approval Voting might or might not come up with a better solution in this hypothetical. It depends on how many of the voters mark their second choice as “acceptable”. But under Approval Voting, if everyone in that scenario marked their second choice as acceptable, choice A would win by a vast majority…

@markfromireland, #43:

I don’t claim to know enough about Ireland’s specific system in action to give a serious critique, but it sounds better than straight IRV from your description. The problem, however, is that often “proportional representation” isn’t possible. Some contests really do need to have a single winner (referenda, for example, which can still have a purpose even with valid representation). In those cases, the proportional representation which is the feature which makes Ireland’s system shine fails.

As I said before, though, there is no known system which passes all tests.

Posted by: The Truth Gets Vicious When You Corner It | Nov 12 2006 4:21 utc | 48

@mark – thanks for the lecture and @John and truth, thanks for the discussion …

Posted by: b | Nov 12 2006 7:09 utc | 50

Vicious Truth :
I don’t see how you get from 40% for nukes to 55% for nukes without having had everyone who voted to muddle through as their first option also choosing nukes as their second option. In which case you got your “desired” result, right?
If more than a third of the armadillos chose to pull out in the event that they can’t muddle through then sanity would prevail.
If none of them expressed any secondary preference at all their votes would not be redistributed and we’d have a “hung” election, to be resolved, presumably, as elections are now, by the simple plurality.
And so on, in between.

Posted by: John Francis Lee | Nov 12 2006 9:36 utc | 52

@John Francis Lee, #52:

Not all the B voters have to give C as a second choice for this to work out with the same outcome under IRV. Just (roughly) two thirds or more, since two thirds makes up 10% of all voters, and C already has 40%.

Here, take a slightly more complex version. In an election with three choices, there are 9 possible ways to mark the ballot:

A – B
A – C
B – A
B – C
C – A
C – B
A
B
C

(Those last three options represent “if I can’t have my first choice, throw away my vote entirely”. In a mathematically defensible IRV setup, this is not allowed, because if you don’t require people to express preferences, you can no longer prove that IRV has any benefit over plurality voting at all, just that it may have some in some cases, making it difficult to justify any change, just as plurality voting may have a benefit over IRV in some cases, such as the original version I gave of this hypothetical. But we’ll allow it, since you’ve asked.) Now suppose that the voters voted as follows:

29%: A – B
2%: A – C
1%: B – A
9%: B – C
24%: C – A
13%: C – B
14%: A
5%: B
3%: C

All choices now have a value, so it’s more “real”. The first round count would be:

A: 45%
B: 15%
C: 40%

As before, there is no majority and B has the fewest votes, so remove/redistribute B and you get:

A: 46%
C: 49%

Deadlock. (The “B or I don’t care” votes are just plain discarded, leaving that 5% gap.) We have a number of choices:

1. Go with A, because it had the most first-choice votes.
2. Go with C, because it has the most aggregate votes.
3. Hold a fresh two-way election between A and C, since all that this election resulted in was the elmination of B.

Almost nobody will choose option 1, because the whole point of IRV is that first-choice-only counting is misleading. Most go with 2, because places that have IRV at all usually assume that it leads to a valid representation. A few go with 3, and if the voters were honest the first time, C will win — the voters who said “B only” will not vote in an election between A and C, which means that “46% to 49%” becomes “a bit over 51% to a bit under 49%”. But more people would support B than C in a one-on-one election (Of the original voters, it would be 44% to 42%.) And the number of voters who said A was their first or second choice was 70%. So even with a less stilted version, IRV still arguably fails to pick the best fit.

Now, suppose we assumed that in an Approval Voting contest, giving a second preference was the same as voting for it. That would give the following count:

A: 70%
B: 59%
C: 51%

So A would win, with B as a second. Anyone arguing for choice C after the fact would have to explain why that option came in last.

The moral of the story? Don’t trust a straight form of IRV to always produce a better result than plurality.

I suspect that no voting system which can be stated simply (as Plurality, IRV, or Approval can) can pass all the possible tests, although it is likely that there is some method that passes the five tests I gave above. (I haven’t gone to check; there are about twenty named methods of running an election/counting votes.)

Posted by: The Truth Gets Vicious When You Corner It | Nov 12 2006 19:07 utc | 54

@John Francis Lee:

Are you even reading what I write? The point of the example is that C is the worst fit to the actual preferences of the population in one way (it comes in last if the voters pick two choices), a second-rate choice in two other ways (it does not have the most first-choice votes, and it would lose to another option in a one-on-one election), and never unequivocally a first choice, and yet IRV ends up choosing it. When dealing with more than three choices, even worse setups are possible; in a five-way vote, for instance, the winner can be a choice which would lose to three other choices in a one-on-one election, and which has the second-lowest number of first-choice votes.

When you throw in extra limitations, such as truncation, IRV becomes absolutely demonic. Say, for example, that you have a four-way contest and truncate the ballot at two choices; unless you hold another election between the last remaining choices (to pull back in the people whose votes were truncated; if you do this, you might as well have just not truncated in the first place, so most people wouldn’t), it is possible for the winner to be a choice which would lose to all others in a one-on-one contest, and which has the fewest first-rank votes. And if you hold that extra election, all it does is make sure the Condorcet loser (the one which loses against all others) doesn’t win. You can still have that second-worst choice…

Note, by the way, that in one way we’re talking about apples and oranges: IRV and PR-STV (the system in Ireland) are both transferable vote systems (that’s the TV part). That is, the total value of your ballot is constant, but it is possible as the election is processed for one choice to gain some or all of that value at the expense of another choice. In Approval Voting (AV from now on), your ballot’s value is never transferred, but different ballots may have different total value. (Voter A only picks one choice, voter B picks seven.) The easy-to-illustrate flaws with IRV derive from the order in which value is transferred. Since AV does not transfer value, that particular class of flaw just doesn’t apply.

Instead, the main flaw with AV is that it has such severely limited ranking, which makes it hard for people to make an honest choice unless all the choices are very polarized. (There are other flaws as well, but this one is easiest to discuss and most likely in practice.) In game theory terms, which is the branch of mathematics where this sort of thing is usually studied, every possible utility function available to a voter under AV has a range of the two discrete values {0,1}, regardless of domain.

As an example: suppose we had an election in which there are four candidates, where the choices are evenly spaced from A to D across some spectrum of opinion. On an IRV ballot, a hypothetical voter whose preference lean toward one end of the spectrum might vote “A – B – C – D”. With AV, though, the same voter would have to make a choice about where to draw the line, which might, in the case of a close contest between their lowest “yes” preference and their highest “no” preference, be agonizing. Would they want to say “I refuse to accept anything but the most extreme form of my choice, so my ballot says A”, “I would prefer an extreme position, but I am willing to compromise for my side to win, so my ballot says A B”, or even “I can’t take the chance that D is chosen, and I could probably live with my conscience if C wins, so I am willing to hold my nose and vote A B C”?

In practice, this works out to voting in reaction to polls. If you think most other people will vote for only their first choice, and your first choice is not the likely winner, then you are likely to be more inclusive in order to avoid the selection of the choices you really can’t stand. Suppose the U.S. adopted approval voting, and in 2008 the candidates were clones of Karl Marx, FDR, Barry Goldwater, and Joseph Goebbels. (For the sake of argument, assume that those are a smooth continuum from far left to far right.) If you’re a regular poster here, you’d obviously start with one or two from the first half of the list, whether you start with Marx or Roosevelt, but suppose it turned out that Marx and Roosevelt were polling too low to possibly win, and the contest was ultimately going to be between Goldwater and Goebbels. There would be no need to drop Marx or Roosevelt off your ballot, because an extra “yes” vote to a losing candidate doesn’t alter the case in AV, but would you add Goldwater to prevent Goebbels from winning, or not? (And what happens if you add him, then in a subsequent poll he’s ahead, and all the people who added him drop him again and Goebbels wins?)

You can start to modify the system to take this into account — say, for example, that you rank each choice as “yes”, “no” or “don’t care”, and count those votes respectively as 1, -1, and 0. But then you can run into weird possibilities: you could have two choices which lead to the same numeric result, say, but which give different amounts of polarization. (Say, for example, that one of them has 10% “yes” and “no”, and 80% “don’t care”, while another has 45% “yes” and “no”, and only 10% “don’t care”. Does the first win, to avoid making more people upset, or the second to make more people actively satisfied with the outcome? Or you could have one candidate who has 55% “yes”, 35% “no” and 10% “don’t care”, and one with 20% “yes” and 80% “don’t care”… Do you alienate a third of the population, or make the vast majority disaffected?) (Presumably, by the way, if you chose to use this system, you would make a choice about this issue in advance. That doesn’t mean it’s not an issue, it just means that you end up having to live with the result if the situation ever arises.)

Posted by: The Truth Gets Vicious When You Corner It | Nov 13 2006 6:24 utc | 56