The close election result, and the lawsuits that followed, have deepened our partisan split into a chasm. Usually sensible people have said amazing things. Columnist George Will, shortly after the election, described what Gore was doing as an attempted coup d'etat. Humorist Garrison Keillor, of Lake Woebegone fame, gave an astonishing diatribe on his PBS program (12/3), accusing George W. Bush of fraud. Newspapers report that some families, to prevent fights, banned the subject from Thanksgiving dinners. Gore's concession, after the second Supreme Court decision, has halted some of the public acrimony, but the bitterness remains underneath, like an onion with marmalade spread over it. Here I will try to present the principal facts on the election so you can decide what part of that bitterness is justified, and what part is not.
When did George Bush win the election? When the Florida Secretary of State, Katherine Harris, certified the election on November 27th. After that, he had 271 electoral votes, one more than the necessary 270 vote majority. Journalists continued to describe the election as an open contest, but in fact, from that time, Gore was contesting the Bush victory. If Gore had been able to overturn the results in Florida and get his own slate of electors certified, replacing Bush's slate, then he would have been the victor. If there had been two competing slates, the election contest would have ended up in Congress, where there were a number of possibilities, most of them favoring Bush. (Among the less likely but more entertaining possibilities: Bush could have been elected president by the House, and Lieberman elected vice-president by the Senate. Also, Jeb Bush might have gone to jail to preserve a victory for his brother, by refusing to sign a new certification.)
Who won the popular vote? Probably Al Gore. Why probably?
Because our voting rolls have so many people on them who should not be
there, and because some people not on the rolls voted. These illegal
voters are mostly immigrants, some legal, some not, but include a significant
number of felons, who are not permitted to vote in most states. The
rolls also have a large number of citizens who, though not necessarily
criminals, gained their citizenship improperly when the Clinton administration
opened the floodgates in 1996.
Besides those on the rolls who should not be, there were a significant number of voters who were not even on the rolls. The Miami Herald did a quick study of part of the precincts in Miami-Dade county and found that about 1 in 1000 voters were not on the rolls where they voted, but were allowed to vote anyway. Traditionalists will not be surprised that one of them used the name of a dead person to vote. (There's an ancient joke told about many places, including Louisiana. "When I die, I want to be buried in Louisiana so I can stay active in politics.") This 1 in a 1000 would be a minimum, since many illegal voters would not be found by the kind of search the Herald used.
These groups are likely to vote Democratic by at least 3 to 1. If more than 700,000 people in these categories voted last November, then Gore did not have a legitimate popular vote victory, since his final popular vote margin was just over a half million.
I have been unable to find any reasonable national estimates of the size of these groups. (Nothing surprising about that. Crime is always hard to measure.) There are some suggestive examples. After the disputed 1996 Congressional election in southern California, where Bob Dornan lost to Loretta Sanchez, investigators found 2538 non-citizens registered in that one House district. (Unfortunately for Dornan, they could find only 624 who had voted. Even with 124 illegal absentee ballots, that was less than his margin of defeat, 979 votes, so Sanchez kept the seat.) Given the very large increase in immigration, the problem has almost certainly grown since 1996. The Miami Herald, in a quick study of counties with just 8 per cent of Florida's population, found 445 felons who had voted in this last election. Again, this is a minimum, since some felons would have adopted new identities or in other ways evaded the Herald's checks.
If these numbers are at all typical, then I would guesstimate that the number of illegal voters in the election was somewhere between 200,000 and 1,000,000. And, I should add that I would not be terribly surprised if the real number was even lower or higher. Given the closeness of the election, illegal voters may have tipped the balance to Gore in several states. This is almost certain for New Mexico (366 vote margin), and possible for Iowa (4,144 vote margin), Oregon (6,765 vote margin), and Wisconsin (5,708 vote margin).
Ignoring the illegal voters, did fraud or irregularities change the
presidential results in any state? Probably not. There
were reports of fraud in the close states of Oregon, Pennsylvania, and
Wisconsin, but in none of them does there seem to have been enough fraud
to change the results. In Oregon, there were several reports of people
intercepting mail ballots. In Pennsylvania, some of the inner-city
Philadelphia precincts reported 100 per cent turnouts. In Wisconsin,
there were hundreds of reports of irregularities in Milwaukee. A
Democratic activist was caught offering homeless people packs of cigarettes
in return for votes. Many students at Marquette University claimed
to have voted multiple times. (The Catholic fathers who run Marquette
clearly have some work to do.)
Besides the fraud, there were many irregularities, perhaps more than usual. In Missouri, a Democratic judge got the poll hours extended briefly for some St. Louis precincts. All across the country, predominately black churches broke the law by electioneering. In part of Arkansas, for example, whole churches closed services and bused their congregations to vote on the Sunday before the election. (This may have cost the Republicans a House seat.) In Miami's little Haiti, Fox News reports that poll workers went into the voting booths to assist the voters, who often could not even read the ballots. (This raises an interesting question. How do people who are illiterate in English pass the citizenship test, which requires literacy in English?) Despite protests, they continued to do this until the police were called.
Why was the election so close? Polls in the week before the election showed a consistent Bush lead of 2 to 6 points, but the election was a virtual tie. Why? Because there was a surge to Gore in the last few days before the election, large enough so that one pollster, Zogby, actually had Gore in the lead in his last poll. Like most political analysts, I completely missed this because I was relying mainly on the polls that averaged results from several days. In most cases these polls are more reliable, but they do miss the last minute changes.
What caused late surge to Gore? Several things. Probably
the most important was the report on Bush's DUI that came just before the
final weekend. No fewer than 28 per cent of the voters said it was
a factor in their vote. There is also some evidence that the
Social Security ads that Gore ran in the last week had some effect.
Some who had planned to vote for Nader switched to Gore in the last few
days. Then, too, most of the undecided voters in the last week were
people who were Democratic in their leanings. The Bush campaign slacked
off a bit in the last few days.
Although not part of the surge, the Democrats again seem to have done better in the "ground war" than the Republicans. (Politicians call radio and TV advertisements the "air war" and the the nitty-gritty efforts to get people to the polls the "ground war".) Turnout by union members was considerably higher than average, and high turnout by blacks was critical in several states. The Republicans may have erred here by using money for things like automated phone calls, instead of trying to get people to call their friends and family.
Why did the networks get the projections wrong twice (!) in Florida?
For a whole set of reasons, some due to the Voter News Service, which collects
data for the networks, and some due to the usual mistakes in reporting
on election night. For the gory details, see this article
by Howard Kurtz.
Far worse than these clerical errors were the early calls by the networks, particularly in Florida. The Florida panhandle stretches so far to the west that it is in the Central Time zone, something the network prognosticators apparently did not know when the called the state before all the polls were closed. There are estimates that the Republicans lost thousands of votes in Florida because of the early call, since the panhandle voted heavily Republican (This seems high to me, but I would believe hundreds.) The networks were also much quicker to call states for Gore than for Bush. Benjamin Wattenberg suggests we take away spectrum rights the networks were given and are sitting on. This strikes me as both fair punishment and good policy.
Did Nader Make the Difference? Perhaps, but so did Pat
Buchanan and the Libertarian candidate, Harry Browne, as an analysis by
John Cavanagh in Nation magazine shows. There were just 8 states
in which neither Gore nor Bush received a majority of the popular vote.
In two of these, Maine and Minnesota, Gore's total was so close to 50 per
cent that he probably would have won them regardless of which third party
candidates ran. Of course, Florida was so close that Nader made the
difference there. New Hampshire was close enough so that Nader may
have made the difference there, as well.
However, the same argument applies to Buchanan and Browne. If Buchanan had not run, Bush probably would have won the states of Iowa, Oregon, New Mexico, and Wisconsin, which have a total of 30 electoral votes. If Buchanan had not run, Bush's margin would have been large enough in Florida so that we would not have had the disputed recounts. Even fifth place Browne's votes were enough to cost Bush New Mexico, and possibly Oregon and Wisconsin. So, there was more than one spoiler in this election.
Why was there so much confusion in the voting and the counting? Because election day is amateur day. Nearly all the voters and most of the poll workers are novices. Few voters go to the polls more than once every two years. The poll workers are a little more experienced, but not much. Most, like the voters, are people who do this on one day every two years. Places that had the most brand new voters, like Florida, had the most problems. And, it gets worse when the form of the ballot or the method of voting changes. In Chicago, there was a very high rate of errors, partly because the punch for a straight party vote had been eliminated. American voters have to vote on more offices and more issues than voters in most countries; this adds to the confusion.
Why was Florida so close? Florida generally leans Republican,
so most expected George W. Bush to win the state easily. He did not
because of something his brother, Governor Jeb Bush, had done. Florida,
like many other states, had a system of preferences that favored blacks
in admission to the state's colleges and universities. A court challenge
to the preferences was likely, and Ward Connerly, who helped pass initiatives
abolishing racial preferences in California and Washington, had tried to
do the same in Florida. Jeb Bush proposed eliminating the preferences
and replacing them with a system that gave guaranteed admissions to those
in the top 20 per cent of their graduating classes. This passed,
but there was an enormous outcry in the black community in Florida.
(This is ironic at the very least, since the result of the Bush plan was
to increase the number of blacks admitted to Florida colleges and universities.)
Their anger carried over to the election. There was an astonishing
65 per cent increase in black participation, and a remarkable 93 to 94
per cent of blacks in Florida voted for Gore. This surge in black
votes enable Gore to almost tie Bush, instead of losing by 3 or 4 points.
This shows, by the way, that George W. may be smarter than his brother Jeb, at least about politics. When faced with the same problem in Texas, George W. stayed away from the issue until after the court decision. Then he came up with a guaranteed admission plan for the top 10 per cent in each high school, which increased the number of black students without explicitly using race. If anything, he may have gained politically with blacks in Texas.
Jesse Jackson, among others, has claimed that the black vote was
deliberately suppressed in Florida. Is there any evidence for this?
No. The black rate of participation was actually higher than the
white rate. This, in itself, caused problems as some areas were swamped
by new voters. The newspaper USA Today, the Florida attorney general,
a Democrat, and the black columnist, Clarence Page, have all looked into
the specific charges and found no evidence for them. In a state with
6 million voters, there were certainly some incidents in which voters were
treated badly, just as there were some incidents in which people were allowed
to vote when they should not have, but there is no evidence of systematic
One systematic thing was done that could have reduced the black vote, but this was simply a matter of enforcing the law. Katherine Harris used a firm to identify felons and sent the list to the counties so that they could purge their rolls. The first list that the firm supplied included many who were guilty only of misdemeanors, not felonies. They corrected this in a second try and made it clear that the counties should verify the felonies, but after the furor from their first try, many counties simply ignored both their list and a more accurate one from the Department of Corrections. The counties that ignored it were mostly controlled by Democrats, which brings me to one last point. Most black voters, in Florida and elsewhere, vote in jurisdictions where the Democrats control the voting. The poll workers there would have every incentive to try to increase, not decrease, the black vote. (Sometimes even up to an improbable 100 per cent of registered voters as happened in a number of inner city precincts.)
Did the Gore campaign try to reduce votes from the military?
Maybe. Democrats certainly did, but it is not clear how much the
Gore campaign knew about the effort when it started. A Democratic
lawyer sent out a memo to Democratic officials in Florida explaining how
to disqualify the overseas military ballots, and they were quite successful
following his advice, where Democrats controlled the counting. In
counties carried by Bush, only 30 per cent of the overseas ballots were
disqualified; in counties carried by Gore, an astonishing 60 per cent of
the overseas ballots were disqualified. Note that even this may underestimate
how successful the Democratic operatives were, since not all of the overseas
ballots are from the military. (If even 30 per cent seems high, it
is in line with rates from previous elections in Florida. The state
has very complex rules for absentee ballots, and it is easy to make a mistake.)
To be fair, I should add that some Bush counties may have used a more lenient
standard than they had in the past. Gore probably gained a net of
about 200 votes from this effort.
The Democrat effort to reduce military votes continues even now. There is an old law that forbids polling places on military bases. In some places, the law has been ignored for decades and National Guard armories, and even a few remote bases, had polling places. Before the election, the Clinton administration sent out a reminder about the old law, which caused some polling places to be moved. After the election, the Republicans introduced a bill to allow voting at military facilities, and it has been blocked by the Senate Democrats. There have also been many reports of delays and mistakes in getting absentee ballots to the military. So far, I have seen no evidence that this was anything other than the confusion that will always happen in an election of this size.
What was wrong with the "butterfly" ballot in Palm Beach County?
It was confusing for people who do not regularly fill out forms.
Those who do, like most school children, had little trouble with it.
The Democratic official, Teresa LePore, who designed it, was trying to
make it more readable for people with poor eyesight. To make the
type larger, she had to put candidates on both sides of the ballot or use
two pages. She chose the first alternative. (The second alternative
doesn't work very well, either. Duvall county used two pages for
the presidential candidates and got a very high number of invalid votes,
since many voters punched a second hole on the second page. Another
county, which uses optical scanners, put the choices on the same page in
two columns. This also led to a high rate of errors.)
The underlying problem was that there were simply too many candidates for some voters. Florida had ten candidates for president on the ballot, which tied several other states for the highest number. Election experts have found that more than 6 or 7 candidates for an office results in a higher rate of errors. (Florida had such a high number because, in a reform passed by the voters a few years ago, the state made it easier to get minor candidates on the ballot.)
And, if I may be permitted a politically incorrect thought, perhaps people who do not know how to vote and do not know who the candidates are, should abstain from voting, for the same reasons that people who do not know how to drive and what the road signs mean, should abstain from driving.
Did Buchanan get more votes as a result of the butterfly ballot? Probably, although far fewer than William Daley and others have claimed. He did better in Palm Beach than he did in most Florida counties, receiving 3,407 votes, or .79 per cent of the total. (He received .46 per cent of the vote in all of Florida.) But he did far worse in Palm Beach than he had in the 1996 Republican primary, when he received 8,788 votes. This was after the race had essentially been won by Dole, so it shows the hard core support for Buchanan in the county. And, other Reform party candidates have also done comparatively well in Palm Beach. Three of the five Reform candidates in Florida, besides Buchanan, had parts of the county in their districts. The two candidates for state senate, whose districts covered most of Palm Beach, received nearly 15,000 votes between them, without much campaigning. The Reform candidate for Congress, whose district included about a third of the county, received 2,651 votes there.
Was there anything suspicious about the original election results
in Florida? Yes, a number of things, particularly in Palm Beach.
The county reported very late, after most of the state. This is a
standard tactic used in stealing elections; you wait to report until you
know how many votes you need. The county had a very high rate of
error in the ballots, 6.4 per cent, which is more than twice the national
rate. Bush's vote was suspiciously low; he ran ahead of the Republican
Senate candidate, Bill McCullom, by more than 200,000 votes in the entire
state, and ran ahead of him in nearly all the counties. In Palm Beach,
in contrast, he ran 1,682 votes behind McCullom. Bush even ran behind
Republican registration in the county; he received just 35 percent of the
vote, even though 36 per cent of registered voters there are Republicans.
(If Bush had received the same share of independents and Republicans that
he did nationwide, he would have received about 44 per cent of the total
vote in Palm Beach.) Patrick Cadell, President Carter's pollster,
and a long time student of Florida elections, reports that some of the
precincts in Palm Beach had very strange results; for example, in one heavily
African-American precinct, McCullom ran 2 to 1 ahead of Bush.
Two explanations come to mind for these anomalies in Palm Beach, the Jewish vote and fraud. Normally Republican Jewish voters may have crossed over to support Lieberman. Palm Beach is among the most heavily Jewish counties in the country, so this would help explain why Bush ran behind McCullom in Palm Beach. However, Broward county, just to the south of Palm Beach is also heavily Jewish, and there Bush ran 2,299 votes ahead of McCollum.
Fraud can not be ruled out entirely. Robert Cook, an engineer with substantial statistical expertise, suggests that votes may have been stolen from Bush as follows: The vote thief would take a stack of punch cards and press a rod of some kind through the Gore slot. This would change no Gore votes but would change votes for other candidates to errors (and a few blank ballots to Gore). It would increase the error total and decrease the votes for all Gore opponents. There was opportunity to do this since the Republicans had very few poll watchers in Palm Beach, 50 to 60 for the entire county, from the news reports I have seen. (If you are interested in Cook's analysis, I can email you a copy.)
Was there anything suspicious about the first recount in Florida?
Yes. It produced an improbably large gain for Gore of 1,484 votes.
To see how improbable this is, just think about the recount process.
In all except some of the smallest counties, where they use paper ballots,
the election workers simply ran the same boxes of ballots through the same
counting machines. If there were no clerical errors the first time,
one would expect almost the same results on the second run. (According
Tell at the Weekly Standard, there was one clerical error that cost
Gore 343 votes in Pinellas county on the first count. Most of the
Gore gains were far more dubious.) You would expect the counts to be almost
the same, rather than identical, because people do not always mark their
ballots clearly enough to be read the same way each time. With punch
card ballots one would expect small gains for each candidate, because running
them through the machines tends to clean out loose chad. In most
of the counties, this is exactly what happened, and there were trivial
changes between the first count and the first recount. (In Washington
state, which used a similar process to recount the votes in the Gorton-Cantwell
Senate race, there was an insignificant difference between the first and
second counts. Washington, like Florida, uses a mix of punch card,
optical and paper ballots.)
In several Gore counties, notably Palm Beach, Gadsden, and Volusia, the results of the recount were suspicious. Consider the Palm Beach recount. It produced gains of 787 votes for Gore and 105 votes for Bush, for a net gain of 682 votes for Gore. Remember this result came, supposedly, from running the same punch cards through the same counting machines. By way of comparison, Broward county, which is larger than Palm Beach, found 43 additional votes for Gore and 44 for Bush. Both the size of the changes, and the bias in the Palm Beach recount, are completely implausible, without some human interference. Several researchers have made statistical estimates that these Palm Beach recount results happened by chance; all found more that the odds against it were more than 1,000,000 to 1. (Although the Palm Beach recount was obviously unethical, it may not have been illegal. One would have to know what was done to the ballots and what is allowed by Florida law to decide.)
In Gadsden, the election board went into a room and "reconstructed" some 2000 ballots, secretly, and came out with 170 new votes for Gore and only 17 for Bush. This secret count is illegal under Florida laws. (The board says that people could watch through the windows, which seems obviously insufficient to me.) In Volusia, the seals on the ballot containers were open when they were brought out for the recount, suggesting that some one had tampered with the ballots. A ballot bag was found in the trunk of an election workers car. And, 264 absentee ballots, counted in the first count, disappeared in the second. All three of these counties have histories of electoral fraud, which will not surprise you at this point.
There were 2 Bush counties, Martin and Seminole, that also produced strange results, but Bush gained only 203 votes from those recounts. Net, Gore gained more than dubious 1000 votes from this first recount. This partisan recount explains much of the Republican objections to further recounts.
Finally, there was one genuine mystery in the recount. Nassau county ended up with 218 fewer votes in the recount than in the first count. The Republican election board decided that the recount had missed a box of ballots; some Democrats were sure that the first count had counted a box twice. Using the first count added 51 votes to Bush's margin.
Which is the better method for counting punch card ballots, machine
counts or manual counts? Theoretically, the best method is a
combination, counting as many ballots as possible with machines and then
doing the rest manually. The machines don't belong to any political
party, but they can miss some clearly legal votes. Some of the smaller
counties in Florida used this method, but larger counties tend to avoid
it because of the cost.
The difficulty with manual counts, even partial ones, is that they provide opportunity for what James Baker called "mischief", that is, manipulation of the ballots, and different standards for the two parties. A manual count requires clear standards and honorable counters. Neither were available in Palm Beach and Broward. And, even with those, the count requires careful checking since studies have shown that counters tend to make mistakes that favor their own sides. (This is a general problem. I have been amused to find that, when I balance my check book, I am about twice as likely to make mistakes in my favor, as in the bank's favor.) It is worth mentioning that the Democratic election official, Teresa LePore, now favors machine counts after watching the process in Palm Beach, because she thinks they are fairer. She also reports receiving many emails, some from friends, urging her to do a partisan recount, that is, urging her to commit vote fraud.
Was the second, manual recount, in Broward, Palm Beach and part of
Miami-Dade counties, suspicious? Yes. The boards kept changing
the standards during the count to admit more and more ballots. They
changed them for an obvious reason: they were not gaining enough votes
for Gore with the earlier standards. Finally they even started counting
"dimpled" ballots in Broward, that is, ballots in which the chad had, or
so the counters said, a dent in it. One can get the idea of
just how far the counters went from one incident. Democratic Senator
Kerrey, who had come to observe the count, was shown a ballot and told
that it was a vote for Gore. He studied the ballot and was unable
to see a vote. At that point he decided he should leave.
There were also many reports from Republican observers of dubious actions by the Democratic counters. Some of this can be ascribed to the paranoia that was infecting many, but I am not sure all of it can be. Unfortunately, none of the major news organizations have done the extensive study of the video tapes of the counts that would help clarify this. Fox News did find several dubious actions by Democratic counters.
Have "dimpled" or "pregnant" chad been counted before in Florida?
No, as a Gore lawyer admitted in the first Supreme Court case. The
reason for this is simple. We simply can not know with certainty
what a voter intended when they just dent the ballot. For that matter,
we can not even know whether the voter put the dimple there. It could
have been introduced by the counting machine or a counter, accidentally
or on purpose. (There is one exception. If all the votes on
a card are dimples, then I would concede that we probably have an inept
voter, but one whose intentions are clear.) Broward county added
nearly 1500 votes to Gore's total and 500 votes to Bush's total by counting
votes that had never been counted before in Florida.
Nor have dimpled chad been counted anywhere else in the United States, with two notable exceptions, a Massachusetts special election for Congress and, ironically, some Texas elections. (In his argument in the first case before the Florida Supreme Court, David Boies claimed they had been counted in an election in Illinois, but he was in error. He has yet to go back and correct the error, as required by legal standards.) The Massachusetts case caused so much furor that, after a judge had allowed dimpled votes, the Secretary of State there banned punch cards from future elections. The Texas laws allow them to be counted if they meet certain standards.
What kind of court is the Florida Supreme Court? It is
Democratic, activist, and, above all, incompetent. All seven of the
members were appointed by Democratic governors, though Jeb Bush concurred
in the last appointment. Even in an era where activist courts are
common, it is astonishingly so. They ruled off the ballot Ward Connerly's
anti-discrimination initiatives as confusing, even though they were similar
to those passed in California and Washington and drew on the language of
the 1964 Civil Rights act. For more see John Fund's Wall Street Journal
Their incompetence has drawn less attention than it deserves. Writers like Morton Kondracke and Andrew Sullivan, neither part of any right wing conspiracy, have concluded that the court's decisions were intended to rig the election for Gore. In practice, the court decisions may have had the opposite effect, making it impossible for Gore to win a legitimate victory.
Was the first Florida Supreme Court decision, to extend the counting time in selected counties, correct. No. To make this decision, they had to say that "must" means "may" and "may" means "must". In other words, they had to ignore the plain meaning of the Florida law, as they more or less admitted in their decision. Although they did this to help Gore, in retrospect, it is clear that Gore would have been better off if the court had let Harris go ahead and certify the results on schedule since this would have given Gore more time for a recount in the protest phase.
Was the first US Supreme Court decision to reverse the Florida court correct? Yes, though they may have been too diffident in their reversal. What they did was much like a mother who comes into a messy house and asks her children what caused the mess. The mother knows what caused the mess and doesn't really want an explanation; instead she wants the mess cleaned up. Similarly, the US Supreme Court didn't want an explanation from the Florida Court; instead it wanted them to clean up the mess they had made. As any mother can tell you, sometimes this approach works, and sometimes it doesn't. It didn't this time. Having been clearly warned, the Florida Supreme Court paid no attention and made a still bigger mess.
Was the second Florida Supreme Court decision, to count just the undervotes in all the counties correct? No. The Florida election laws require that manual recounts include all the ballots, precisely to prevent this kind of cherry picking. Two details show just how weirdly partisan the decision was. As I mentioned above, Nassau county got different results in the first and second counts. The obvious solution to this problem, it seems to me, is to have them recount one more time, this time with a little more care. Instead the Florida Supreme Court simply told them to use the second count, which, surprise, surprise, was the one that favored Gore. Even stranger, as Mickey Kaus, a Gore supporter, noted, was their proposal for Miami-Dade. The county had counted part of the precincts with a manual count, using one set of standards. The court ordered, in its second decision, that just the undervotes for the rest of the precincts be counted using a different set of standards! No one could accept the results from Miami-Dade arrived with that strange mix of procedures.
Was the second US Supreme Court decision to reverse the Florida court
correct? Yes and no. It probably was the best solution
available given the time constraints, after the mess the Florida Supreme
Court had made of matters. The crucial vote was not 5-4, but 7-2,
in favor of overruling the Florida Supreme Court's second decision.
(For that matter, even Justice Ginsburg, who dissented on both votes, had
doubts about the fairness of the recounts.) Justices Souter and Breyer
thought it still might be possible to get a new count, and so did not agree
with the majority on the remedy. In practice it is hard to see how
a new fairer count might have been done, since the Florida court had shown
no interest in that outcome in their two previous decisions.
Although one can defend the decision on practical grounds, it is hard to defend its strange compromise reasoning, which rested vaguely on the equal protection clause. Three justices, Scalia, Rehnquist, and Thomas, had a coherent position intellectually; they simply wanted to go back to the certification and stop there. (To put it another way, the three were saying that the two Democratic judges, Terry Lewis and N. Sanders Saul, had been right in their original decisions, and that the Florida Supreme Court had been wrong when it overruled them.) The two justices who wanted to try again for a complete recount, Breyer and Souter, also had a coherent position, though the practical difficulties were great. The swing justices, O'Connor and Kennedy, not for the first time, had a position that makes little sense intellectually. (Although I must say I admire O'Connor's common sense on one point. She suggested that the standard for counting a ballot be the directions, an idea that can't be faulted when the directions are as clear as they were in this case.) And, I should add that the positions of the two dissenters, Stevens and Ginsburg, though they may be coherent, show no concern for justice in the case.
Overall, was the entire legal struggle just another a fight between Republicans and Democrats? No. The Gore campaign sued Democratic election boards, including Broward, Palm Beach and Miami-Dade. They lost significant decisions before Democratic judges in Florida. Their only significant legal wins were the two weird Florida Supreme Court decisions. The Bush campaign sued Republican boards, and was unable to get significant help from federal courts, even though the majority of judges on them had been appointed by Republicans.
Why was there never a complete recount? Because the Gore campaign never requested it. In fact, they explicitly rejected it. Although he, and his campaign, have said, again and again, that he wanted to count all the votes, his attorneys explicitly rejected a complete recount in their arguments before the Florida Supreme Court. (Gore did put out a vague proposal to Bush for a complete recount, but he did this after the deadline for requesting recounts had passed. In any case, Gore did not need Bush's agreement to ask for a complete recount.) They may have been motivated to do this by their automatic opposition to anything Katherine Harris wanted. In the first case, she had proposed that the Florida Supreme Court combine all the separate cases. If they had, the next obvious step would have been a complete recount of the entire state with a common standard. Since Harris wanted it, the Gore people were sure it was bad for them and opposed it. (The Bush team were privately furious at Harris for this proposal.) One Gore lawyer said later that they opposed a complete recount because they thought they would lose.
Is it now possible to do a fair recount of the ballots in Florida? No. Too many of the ballots have been handled too many times by too many partisans. Both punch cards and optical ballots are easily altered in handling, even without trying. It is certain that many of the ballots were altered, given the strange results of the first recount.
If it were possible to do a fair recount, who would win?
Probably Bush, if the usual standards for legal ballots were used.
Possibly Gore, if the new standards were used. To put it another
way, counting legal ballots would probably give Bush the win, but counting
"intentions" might give Gore the win. (Of course, if we are counting
intentions, we might want to include those discouraged Bush voters out
in the Florida panhandle, and the military voters who didn't follow the
directions, and who knows what other groups.) There have been several
attempts to estimate who would win, but all that I have seen suffer from
the same mistake; they begin with the wrong baseline, accepting the dubious
results of the first recount and the machinations in the manual counts
in the Broward and Palm Beach. Although this is an interesting practical
question, it evades the larger question of fairness. (For a example
of such estimates, see Slate for two
articles by Jacob Weisberg.) If these dubious results are excluded,
as they should be, Gore would have to overcome a Bush lead of more than
2000 votes, which is highly unlikely.
There is a larger point that I touched on above. Even if Gore did gain the lead in a recount, he would be unable to get a large enough margin so that Republicans would think his victory legitimate. Suppose, for example, that Gore was able to gain a lead of 500 votes in a recount. Then, he would owe his lead to the votes of the murders, rapists, child molesters, drug dealers, and other felons that Democratic election boards had refused to remove from the rolls. Suppose his lead was larger, say 2000 votes. Then, he would owe it to felons and people who received their citizenship illegitimately during 1996, because of his pressure on the INS. Even a lead of 20,000 votes would not be sufficient, because there are almost certainly more than 30,000 non-citizens on the rolls in Florida. As long as our election rolls are so open to fraud, any close victory for a Democrat will look dubious to Republicans, and rightly.
In summary, who has the most reason to be bitter, Democrats or Republicans?
Probably the Democrats, since they lost and never had a full recount of
the votes in Florida. However, their bitterness should be directed,
not against the Republicans, but against Al Gore, who never requested a
full recount, and against the Florida Supreme Court, who with all their
imaginative legal rulings, never ordered one. The Democrats, and
all the rest of us, also have reason to be bitter toward those county canvassing
boards who, in the first recount, produced such a biased result as to poison
any succeeding recounts.
However, had the result gone the other way, the Republicans would have far more reason to complain than the Democrats do now. It is not going too far to say that Bush won in spite of fraud, not by the Gore campaign, but by Democratic activists. Bush is like a basketball player, who, in his drive to the winning basket, is hacked over and over without a call from the referee. Winning may make up for that, but it doesn't make the bruises go away.
Finally, what do the parties ignore in their views of the dispute?
The Democrats ignore the considerable evidence of fraud, from the illegal
voters to the implausible recount results. I can not think
of a single instance in which a Democratic official has admitted that there
is a serious problem with our voting rolls, or that the results of the
first Florida recount are impossible to believe. The Republicans
ignore the honest desire to find legal votes that a machine could not see.
Both parties ignore the faults in the court decisions that favored them.
Curious facts about chad: Chad is one of those nouns, like
sheep, that have the same form for singular and plural. For example:
The chad from this ballot is now in the pile of chad on the floor.
Saying "chads" is as wrong as saying "sheeps". (Or perhaps I should
say was wrong. Since it has been used wrong so frequently during
the last months, it will probably have a regular plural in the next edition
of the dictionary.) Although it is a fairly recent word, its etymology
There was a Saint Chad in medieval England, who is most famous for giving up a bishopric after it was charged that he had gotten it improperly. This impressed many, and, a few years later, he was again made a bishop. For more on his story, check the Encyclopedia Brittanica.
Curious fact: British researchers think they have found a woman who sees in four colors, technically known as a tetrachromat. (I guess that makes most of us trichromats and a few of us guys bichromats.) She sees colors more vividly than other people and can see differences that we trichromats can't.
Truly Silly: There are so many candidates this time, that
it is hard to pick just one. For instance, there was writer Ron Rosenbaum's
claim that columnist Lars-Erik Nelson died because of Republican chicanery
in Florida. (He had a heart attack. So far as I know, no Republicans
have claimed that Dick Cheney's heart attack was caused by Al Gore.)
But even in this large field, Michael Kinsley, editor of the online magazine,
stands out for a series of post election columns in which he insisted that
the Republicans had stolen the election and that it was "clear beyond a
reasonable challenge" that Gore was supported by a plurality of voters
in Florida. This is both disputable and irrelevant, since the question
is not who had the most support, but who received the most legal votes.
In this series of columns, Kinsley completely ignored or dismissed opposing arguments. He never discussed the amazing results in the first machine recount, described above. He sneers at the Republicans for attacking rule changes, but, in fact this is exactly what counters in Palm Beach, Broward, and other Democratic counties, did. Kinsley's performance is remarkable because these very points were discussed in his own publication, Slate, by more sober, though no less Democratic, writers like Jacob Weisberg. (One wonders. Kinsley edits Slate, but does he read it?) I should add that, to my knowledge, no one at Slate has done a serious examination of the far larger problem of illegal voters, which casts doubt on Gore's popular vote victory.
Quiz: Almost universally, journalists say that Bush was the first to win the presidency with fewer popular votes than his opponent since 1888. There is a more recent election in which this may have happened. Which one was it?
Worth a Look: This USA Today map
shows the counties carried by each candidate. Two points worth mentioning:
First, in recent elections, American journalists almost always reverse
the colors when they make these maps. Left wing parties all over
the world are identified as red or pink, so, logically, the Democratic
counties should be in red and the Republican counties in blue, rather than
the other way around. I have never seen an explanation from the journalists
for this reversal; I suspect the journalists use red for the Republicans
because, in the United States, the color red is commonly used to identify
Second, you can see much American history in the map, as two examples from the Civil War era show. If you look at parts of the South, like along the Mississippi river, you can find in blue what was long known as the "black belt", both for its soil and the black field hands who worked its plantations. Even now, these counties have majority black populations. These counties were strongly Democratic before the Civil War and have continued to be, though now for opposite reasons. In contrast, the mountainous areas of the South had poorer soil and fewer slaves. They opposed secession and have backed the Republican party since Lincoln. Look at eastern Tennessee for an example.
Worth Reading: An (11/21) New York Times article on a grenade
proof greenhouse in Switzerland. What requires this kind of protection?
A new variety of rice, containing beta carotene, that could prevent blindness
and save millions of lives in very poor countries. (Beta carotene,
as you probably know, is converted into vitamin A by the body.) Unfortunately,
environmental extremists in Europe are so superstitious about any product
of genetic engineering, like this rice, that they have blocked its export.
A column by Michelle Malkin on how UNICEF is helping to spread AIDS to babies. In the past, companies that make baby formula used unscrupulous practices to sell it in some poor countries. Some activists have never forgiven them for this, so they are now blocking even the free distribution of baby formula to mothers with AIDs.
Recommended Web Sites: Two lively personal sites, by writers Andrew Sullivan and Mickey Kaus. (Sullivan's site, unfortunately, has been "designed", which, as usual, makes it prettier but harder to use. The content is worth the trouble.) I have disagreements with both on a number of issues, but nearly always find them worth reading. For examples of their work, see Sullivan's defense of the drug companies, or Kaus's piece on the subtle shift by George W. Bush on poverty, when he dropped some of his condescension toward the poor. Kaus also has an amusing feature that I intend to steal, an "assignment desk", where he outlines articles he thinks should be written and assigns them to likely culprits.
Follow Up: In the column on the candidates' economic plans,
I gave the then current budget surplus estimate, 4.6 trillion over the
next 10 years. According to USA Today (12/22), that estimate is now
being revised upward again, adding 1.4 trillion to the estimated surplus.
The increase comes from more optimistic assumptions about productivity
growth. The same cautions I gave then about the estimate still apply,
but there may now be more leeway for both tax cuts and new spending - if
the estimates are right about productivity growth.
Copyright © 2001 by James R. Miller