The “establishment” response to critics of the election
results (a.k.a. “conspiracy nuts”) has been, for the most part – no
response. The issue is virtually absent from the commercial media,
despite persistent investigation and passionate debate in the internet. When
the concerns of the critics do provoke replies from defenders of election
outcome, these replies usually take the form of ridicule and insult, or a
plea that the critics “get over it” and that “we all move on.” Nonetheless,
a few defenders of the fairness and accuracy of the election, respond
responsibly to the critics. Even so, these rebuttals fail, as I will attempt
to demonstrate below. Good reasons remain to suspect that the Presidential
election of 2004 was in fact stolen.
The Florida registration discrepancy. Soon after the election, some
sharp-eyed statisticians discovered in the Florida returns extraordinarily
large discrepancies between party registrations and the recorded
presidential votes. In particular, of the counties using optical scan
ballots most had large majorities of Democratic registrations, and yet the
voted went overwhelmingly to Bush, which means that “virtually every
unaffiliated voter appear[ed] to have gone for Bush.” Practically speaking,
this is impossible. In contrast, in the counties using the much-suspected
touch-screen machines, the unaffiliated voters split evenly, as expected.
As "The Squanderer" reports, “These two Florida populations who seem to have voted so
differently are roughly equivalent in size and relative party strength. Yet
the counties with the "optical scan" machines went disproportionately,
overwhelmingly, and seemingly illogically, for Bush. What could account for
the difference in results? Did the machine itself make the difference, or is
there another explanation?”
That explanation was soon provided by Jasjeet Sekhon of Harvard University,
and Jonathan Ward and Walter Mebane of Cornell University, who pointed out
that the discrepancy was due to what has come to be called “the Dixiecrat
effect” – the inclination of traditional rural Democrats to vote Republican. They write:
The pattern in which counties that have high Democratic
registration had high percentage increases in the vote for Bush reflects
the fact that all those counties have trended strongly Republican over the
past twelve years. The counties are mostly in the Florida Panhandle.
Given the voting history and registration trends, these counties seem to
have many old-style southern Democrats who have not bothered to change
their registration. [EP italics]. (Link lost)
“The Dixiecrat Effect” theory is confirmed by statistics from past
elections. However, unfortunately for this argument, the claim that “the
counties are mostly in the Florida Panhandle” is simply false, as indicated
by the map to the right. Moreover the panhandle counties in question are
rural with small populations, and thus do not weigh heavily in the totals.
To further test “the Dixiecrat effect theory,”
Liddle excluded the border-panhandle counties by comparing medium-sized
counties (populations between 80,000 and 500,000), and found that the
registration/voting discrepancy was still evident, though less so than when
all counties were included in the calculations. Liddle reports that “I have
just repeated my analysis of mid-size counties, this time first of all
omitting NW panhandle counties, then omitting al north Florida counties.
The machine effect remains robust.”
She concludes, “the apparent machine effect is not confounded by
something special happening in north Florida.” Thus it appears that
the attempt to dismiss the registration/voting discrepancy by citing “the
Dixiecrat effect,” simply does not “pan out.”
But even if, despite all this rebuttal, the full force of
the Sekhon-Ward-Mebane argument is conceded, this by no means proves that
the 2004 election was fair. The “Dixiecrat effect” is confined to Florida,
and moreover, virtually all the many remaining irregularities (most
significantly, the 130,00 to 260,000 advantage to Bush, discovered by the
UC Berkeley Quantitative Methods Research Team), worked in Bush’s favor,
and might together have sufficed to have “stolen” Florida, and thus the
Presidency, from John Kerry.
The CalTech/MIT Study. Nine days after the election, the CalTech/MIT
“Voting Technology Project” released a report:
"Voting Machines and the Underestimate of the Bush Vote." The report
concludes that “there is no evidence, based on exit polls, that electronic
voting machines were used to steal the 2004 election for President Bush.”
Sounds pretty cut and dried, doesn’t it?
But as was the case with Mebane,
et al, even if that conclusion is
sustained, it does not follow that the election was not “stolen;” only that
the theft was not accomplished via individual e-voting machines. And there
are reasons to doubt that conclusion, notably the employment by the Project
of circular reasoning (discussed below).
Before the election, “black-box-voting” skeptics focused their attention on
the threat of “retail hacking” – deliberate software manipulation of each
individual touch-screen voting machine (which recorded about 30% of the
votes). To be sure, unlike stuffing paper ballot boxes, this software fraud
would not require individual attention to each machine. The standard (but
secret) software issued by Diebold, ES&S and Sequoia could do the trick, as
could a two-way networking of the machines. And the poll workers at the
precinct level would be none the wiser.
It turns out that the threat of “black-box-voting” is far more widespread
than the critics had earlier feared. The same private Republican-owned
corporations that build and code the touch screen machines, also compile the
total votes incoming from the local precincts. (See our discussion above of
the “Florida registration discrepancy”). One might suppose that the optical
screen method would be tamper-proof, since this system utilizes paper
ballots. However, independent auditing is possible only if the ballots are
made available by the state government officials (in most cases, the
Secretaries of State). Otherwise, the system might just as well be
“paperless.” In Florida, the Republican Secretary of State, Glenda Hood
(handpicked by George Bush’s brother, Governor Jeb Bush), refuses to release
the optical scan ballots for inspection.
Far more significant than the individual touch-screen machines are the
privately owned, operated and coded computing systems that compile regional
and statewide totals. These systems, provided once again by the Diebold and
ES&S corporations, compile 80% of the national votes. These systems are
vulnerable to “wholesale hacking” in “real time.” Their vulnerability has
been publicly demonstrated numerous times, by hackers with wide-ranging
technical abilities, from Ph.Ds in Computer Science, down to teen-age
computer geeks. The systems can be invaded, the totals altered, and the
intruder can exit without leaving a trace of the invasion and tampering –
all within a few seconds.
one famous instance, Bev Harris did just that on a live TV program,
hosted by Howard Dean.
Obviously, we have gone far beyond the issue of whether individual
touch-screen machines can be tampered with. And the CalTech/MIT study was
confined to the question of “retail hacking.” The larger, and far more
likely problem of “wholesale hacking” was not dealt with in this study.
However, the CalTech/MIT study has a far more serious flaw: it treats the
“later exit polls” as predictors of the election, as it then points to the
close correlation between the late polls and the “official” final results as
validation of the election.
In fact: the early and the late exit polls are different in kind, and not
degree! The early poll predicts the election, while the later poll
is “adjusted” by incorporating the actual election returns. As
Farhad Manjoo correctly points out, “The [later] exit polls that are
currently on new sites like CNN have been re-weighted to match the final
results – a standard practice.” Thus,
Baiman, "The ‘final’ numbers are not mean to be independent predictors
of the outcome but rather a data source on who voted and why.”
Freeman correctly points out, “The MIT CalTech Voting Project...
[concludes] that [the later] exit poll data were consistent with state
tallies and that there were no discrepancies based on voting method,
including electronic voting systems. But they used these adjusted data to
validate the process. In other words, they used data in which the count is
assumed correct to prove that the count is correct.” (EP’s Italics)
This is what logicians call a “circular argument,” whereby one assumes what
one proposes to prove. (For example, a prosecutor’s opening statement: “we
will prove that this murderer is guilty as charged.”). In the CalTech/MIT
case, the late polls, which were "adjusted" to conform to the official
returns, are presented as "proof" of the accuracy of the returns. And
so, as one reads the CalTech/MIT study with this fallacy in mind, the
critical reader will find that the arguments and the conclusions of that
It is impossible to overstate the significance of the error of treating the
later poll as a “refinement”of the earlier and as a predictor of the
election. This error is employed repeatedly, albeit fallaciously, by those
who cite the late poll as “proof” that the election was fair and accurate.
Finally, and quite amazingly, the study makes an “apples and oranges”
comparison of pre-election opinion polls with exit polls, thus presuming to
diminish the significance of the troublesome gap between the early exit poll
predictions and the final reported outcomes in the “battleground states.”
(pp 2-3) But of course, opinion polls and exit polls are radically
different. Opinion polls rely on the subjects’ report of how they intend to
vote. Exit polls report how the subjects have in fact voted. Thus, while
opinion polls typically have margins of error of several percentage points,
the margins of error of exit polls rarely exceed one point. Because they
have proven to be extraordinarily accurate and dependable indicators of
actual votes, exit polls are often employed as “checks” of the legitimacy of
elections. And when there is a marked discrepancy, they can lead to a
rejection of the election results, as was the case recently in the Republic
of Georgia and Ukraine. But not, unfortunately, in the United States of
America. The proven accuracy and reliability of exit polls, combined with
the extraordinary gap between these polls and the official totals in the
“battleground states” of Florida, Ohio and Pennsylvania have, in the hands
of some capable statisticians, yielded compelling evidence of fraud in the
election. (See “Freeman and Baiman Revisited,” below).
Farhad Manjoo at Salon.com
recapitulates many of the
defenses of the conventional view that I discuss elsewhere in this article,
so there is no need to repeat them here. However, one of his defenses of the
legitimacy of the election deserves special attention.
Manjoo cites the Ohio statistics that John Kerry gave as the reason for
On Nov. 4 ... the Cleveland Plain Dealer reported that
there are only about 93,000 spoiled ballots in Ohio. There were also about
155,000 provisional ballots cast in the state -- votes cast as a last
resort by people whose names could not be found on registration rolls when
they went to the polls. Bush is currently leading Kerry by about 136,000
votes in Ohio. For Kerry to win, then, Ohio would have to have a way to
count all 248,000 outstanding discarded and provisional votes -- which
isn't going to happen -- and then 77 percent of those ballots would have
to go to Kerry. Such an outcome is all but impossible.
But if the number of unexamined spoiled and provisional
ballots was insufficient to change the outcome, does this prove that Kerry
did not lose due to an alleged “rigging” of the Ohio election?
Of course not. If Kenneth Blackwell and his accomplices stole the Ohio
election, the dastardly deed was accomplished through a multi-front
campaign, including vote suppression, “lost” ballots, and perhaps hacked
e-voting and compiling machines.
In short, Manjoo, like many defenders of the conventional view, seems to
assume that if none of the irregularities, separately, will overturn an
election, then the election result is as it should be. And so we hear, over
and over, “this was not enough to affect the outcome.” It does not seem to
occur to these advocates that a multitude of electoral shenanigans can
together add up to a stolen election. The same fallacy was prominent in the
apologetics for Bush’s alleged “win” in Florida in 2000, and again in 2004.
Russ Baker displays a paradigm example of the smarmy contempt heaped
upon the skeptics by the defenders of “the official view.” Baker thus begins
his TomPaine.com article:
Many of us fear that the Ohio election was stolen because
people—like talk show sleuths, blogger number-crunchers, forensic
attorneys, crusading professors and partisan activists—keep telling us so.
We don't even know most of these people, yet we gladly forward their
e-mails and Web links, their pronouncements, analyses, essays and
statistical exercises. While their credentials may not be that impressive,
we listen to their conspiracy theories because—frightened by the direction
our country has taken—we want to believe them.
We “fear” a stolen election “because people ... keep
telling us so?” No!, we suspect fraud on the basis
of compelling evidence. It is the “establishment” that is attempting to
convince the public that the election was copasetic, on little more than
We “don’t know who these people are?” “Their credentials
may not be that impressive”? Pay attention, Mr. Baker, while we
introduce just a few of “these people” (i.e., “crusading professors”) and
cite their “credentials.”
Freeman (University of Pennsylvania),
Caithamer (University of Southern Indiana),
Dr. Ron Baiman,
(University of Chicago), Dr.
David Dill (Stanford University), Dr. Avi Rubin (Johns
Hopkins University), and
many, many more. But you get the point. (For long lists of articles
and links dealing with
Electoral Integrity and
2004 Fraud see The Crisis Papers).
Baker then cites his investigations of several “charges” made by the
skeptics. Consider this one: “Charge: Miscounting of absentee votes.
Finding: False.” His evidence? The testimony of one – just one
– witness to one – just one – event.
There’s more to this article, but having seen this much, why bother to
Mitofsky’s Cop-Out. The gap between the early exit
polls in the battleground states, and the official announced results
presents Warren Mitofsky, the chief guru of the exit polling organization (NEP),
with an excruciating dilemma: either the polls were accurate, in which case
there is something criminally rotten in the state of Ohio (and Florida, and
Pennsylvania, etc.), or, on the other hand, the administering of the polling
and/or in the compiling of the data was horribly botched.
Mitovsky, who is no fool, has chosen the second alternative: he is willing
to take his lumps and affirm the validity of the official election tallies.
Crossing the GOP and the White House is just not a good business decision.
How then does he explain the gap? According to USA Today, “Kerry’s supporters were more willing to
participate than Bush’s. Also the people they hired to quiz voters were on
average to young and too inexperienced and needed more training.” In
addition, “early results were skewed by a ‘programming error’ that led to
including too many female voters.” (Link lost). And “Some local officials prevented
interviewers from getting close to voters.” There’s more, but that’s the
gist of it.
These explanations betray a hint of desperation – as if they were concocted
to serve a pre-defined purpose: explaining (away) those discrepancies
between the early polls and the reported votes. These are, strictly
speaking, not “explanations at all,” they are hunches. And unless and until
independent evidence is produced to support these suppositions, there is
simply no reason to believe them. And no independent evidence has been
, pp. 13-16, addresses and debunks most of these alleged “explanations”
of the discrepancies. And this just in! Mark
Blumenthal reports a study which appears to debunk the "reluctant Bush
It gets worse, as the “explanations” generate further troublesome questions:
(1) Why are the greatest discrepancies in the crucial “battleground states,”
while the results in the “safe states” are within the margin of error. Why
does it just happen that the interviewers in Ohio, Florida and Pennsylvania
were incompetent, while the interviewers in (for example) non-contested
states such as Maine, Maryland and Montana were so brilliantly competent
that they came up with statistical bulls-eyes? (Follow
this link, then search "truthisall")
(2) Similarly, what is it about the Bush supporters in Ohio (etc.) that
makes them embarrassed to admit their preference, while the voters in
Maryland (etc.) have no such reticence?
(3) Why was the “shift” from the exit polls to the final results in 42
states toward Bush, while the in the exceptional nine states, the shifts
were all within the margin of error. (See Figure 2, below).
Despite these nagging questions, the “conventional wisdom” must be upheld at
all costs: Bush won fair and square. And if, despite the best efforts
of the GOP and the media, public discussion of “the exit poll puzzle” cannot
be avoided, that discussion must be contained within the “official” frame.
Thus, for example, the USA Today article begins: “The exit polls of voters
on Election Day so overstated Sen. John Kerry’s support....” The very notion
that the polls were accurate and that they disclosed a massive fraud (as in
the Ukrainian elections) is, well, just not to be discussed in polite
Freeman and Baiman Revisited.
Dr. Steven F. Freeman, (Ph.D, MIT) of the University of Pennsylvania’s
Center for Organizational Dynamics (described by Farhad Manjoo as “an
amateur”), has created a sensation in the internet with his research report,
Unexplained Exit Poll Discrepancy". In his latest version of the report,
dated December 29, 2004, Freeman calculates the odds against the
discrepancies in the Ohio, Florida and Pennsylvania polls occurring together
and at random to be 662,000 to one. In effect, statistically impossible.
Freeman’s paper is brief and accessible to the non-technical reader.
(Excluding tables and figures, it is about a dozen pages long). It should be
read by anyone with more than a casual interest in the issue of the
integrity of the 2004 election. These are his conclusions:
In this report, I have: (1) documented that, in general,
[early] exit poll data are sound, (2) demonstrated that it is exceedingly
unlikely that the deviations between exit poll predictions and vote
tallies in the three critical battleground states [Florida, Pennsylvania,
and Ohio] could have occurred strictly by chance or random error, and (3)
explained why explanations for the discrepancy thus far provided are
Given that neither the pollsters nor their media clients have provided
solid explanations to the public, suspicion of mistabulation or even fraud
is running rampant and unchecked. The fact that so many people suspect
misplay undermines not only the legitimacy of the presidency, but faith in
the foundations of the democracy.
Systematic fraud or mistabulation is as yet an unfounded conclusion, but
the election’s unexplained exit poll discrepancies make it an unavoidable
hypothesis, one that is the responsibility of the media, academia, polling
agencies, and the public to investigate.
concurring paper, Dr. Ron Baiman, a research statistician at the
University of Chicago, argues that “exit polls leave little doubt that in a
free and fair election John Kerry would have won both the electoral college
and the popular vote.” This too is a “must read.” Dr. Baiman concludes.
These unexplained statistical anomalies in the vote count
in critical states, such as Ohio, Florida, and Pennsylvania, and in the
national popular vote for the 2004 Presidential elections, indicate:
a) Implausibly erroneous exit sampling especially for the national sample
and for the most critical states where one would have expected pollsters
to be most careful, and/or
b) Election fraud and/or discriminatory voter suppression that resulted in
a in an election result in Ohio, Florida, and other states, and in the
national popular vote outcome, that is contrary to what would have
occurred in a free and fair election.
I conclude that, based on the best exit sample data currently available,
neither the national popular vote, or many of the certified state election
results, are credible and should not be regarded as a true reflection of
the intent of national electorate...
In Conclusion: Despite the firm empirical and
statistical foundations of the critics’ case, the conventional spin-meisters
have decisively won the early rounds of this contest, as even such
progressive stalwarts as Bernie Sanders, Al Franken, Paul Begala, and most
of all, John Kerry, “concede” that Bush won the election “fair and square.”
Once again, the Repubs, with the invaluable assistance of the corporate
media, have succeeded in “framing” the issue to their advantage, as the
critics have been effectively banished from polite political society and
discourse. And so, the Democratic National Committee and the Democratic
Congressional and Senate Campaign Committees proceed, firm on the assumption
that they’ve got a “good shot” at retaking the Congress in 2006 and the
White House in 2008. Implicit in these assumptions, is the belief that these
will be fair elections. But if they are not, then the Democrats will be
wasting their time and their contributors’ money. The results of these
“elections” will be pre-ordained.
But facts are persistent things, and “truth crushed to earth will rise
again,” unless, of course, the critics surrender and abandon the contest.
Crimes engender cover-ups and conspiracies have a tendency to unravel,
especially when they are doggedly investigated. Such investigations on the
Federal level are, of course, out of the question. However, federal
elections are administered by the states, thus opportunities are available
for investigation by state, county and municipal prosecutors.
Meanwhile, we are left with a residue of unanswered, and perhaps
unanswerable, questions about this election – questions which probe into the
heart of the issue. These are questions which, if persistently thrown
against the barricades of “conventional wisdom,” may erode the foundations
of the malignant Bush regime and eventually bring it down.
In my next essay, I will list some of these questions, examine their
implications, and suggest how they might be employed as weapons against the
emerging theocratic-corporate oligarchy that is the current United States
Copyright 2005, by Ernest Partridge