[EM] HR811 and Federal paper trail legislation

[Abd ul-Rahman Lomax]

At 07:50 PM 5/24/2007, Kathy Dopp wrote:
>On 5/24/07, James Gilmour <jgilmour at globalnet.co.uk> wrote:
> >
> > I am surprised you say "millions".  Does this reflect the 
> illiteracy level?  Or some other factors > affecting ability to 
> complete ballots?   If so, what factors?  What type of voting 
> method could be > used effectively by those unable to use paper ballots?
> >
>
>So if you are claiming that there are not millions voters with
>disabilities who cannot use paper ballots, then let us know.
>
>How many are there?  If not millions? Then how many?

This is distressing. Mr. Gilmour did not claim there were not 
millions. He stated he was "surprised." He did not, in fact, express 
any doubt about the number, and he asked for details. the response he got?

No details, just a question in return. If not millions, then how 
many? The number was asserted, and the one who asserted it would 
properly be expected to justify it. However, that was not the point at all.

The point was that there are many who cannot use paper ballots, it is 
asserted, and if this is true the exact number does not matter, as 
long as it is large enough to justify the investment in accomodating 
them. If it is one person, we would not invest millions of dollars to 
do it. If it is one percent, which it well could be, I'm not 
suggesting that as any kind of limit, quite a lot could be spent to 
accomodate them, though it should still be a balanced number.

The real point, and the reason why I actually assert that the comment 
about millions of people who cannot use paper ballots is a red 
herring, is that *any* specific method will not be usable by some 
people. Alternate methods of voting can and should be provided, 
within reason, to everyone. "Within reason?" Well, there is a limit, 
no specific number, to what can properly be spent to accommodate an 
isolated voter. In the real world, there are funding limits. If we 
devote a million dollars to enable a handful of voters to vote 
easily, we don't have a million dollars that could save thousands of 
lives in another context. We must make choices. Absolutes generate 
bad decisions, when they aren't flexible.

Now, the standard alternative, the default, which is actually in use, 
is to allow those unable to vote directly using a paper ballot, to 
vote an absentee ballot with assistance. Generally, if I am correct, 
it is unlawful to assist another to vote. But an exception is made 
where a person is unable to use the ballot directly. I don't know the 
details, but the point is that an absentee ballot can be filled out 
by anyone designated by the voter, where this is necessary.

This is, indeed, one reason why I find the decision of a Florida 
court to discard all absentee votes on the grounds that a few hundred 
of them, at most, may have been corruptly influenced. In a word, 
bought, though the circumstances could just as easily have been 
described as the offender helping some voters to vote, perhaps some 
homeless people or loiterers. The court, in its reasoning, claimed 
that absentee voting was a privilege, not a right, and that therefore 
it could not stand in the way of the right of the plaintiff, the 
losing candidate, to a victory; this was a truly bizarre case, since 
what the absentee vote did was to turn a victory for that loser into 
a majority failure; there was then a runoff, and the Republican, the 
one whose unofficial aide was accused -- not him -- won. Without any 
absentee vote. The court decided that, because of the alleged vote 
buying, the absentee vote was suspect, and, by its privilege 
argument, could be discarded entirely, and then the original election 
was won by the Democrat. So the runoff, where the election in a 
pairwise contest actually chose the Republican, became moot.

Absentee voting should not be a privilege. For some, it is an 
essential element in the right to vote.

And once you have that, you have a means of dealing with any 
disability at all. With a paper ballot. The objection was spurious 
and irrelevant.

Now, I should research HAVA. But I must say that I have, just from 
the name of the Act, a serious suspicion about it. Have you heard the 
standard Libertarian joke? I read a variation on it in the 
Smithsonian Magazine a while ago. A scientist was capturing sturgeon 
and determining which ones to keep for various purposes aimed at 
restoring sturgeon population in a U.S. river, seriously threatened. 
He caught this sturgeon in a net, and said to it, "I'm from the 
government. I'm here to help."

Help America Vote Act? Did we need help that we did not already have? 
Indeed, much of the problems that had arisen recently were a result 
of various means adopted ostensibly to help us vote, or, alternately, 
to make it easier or supposedly more accurate to count the vote. Hand 
counting of pieces of paper can be done with high accuracy and speed. 
My own experience as a printer required me to do this on occasion. 
Banks do it with paper money all the time. There is even standard 
equipment that will do it. Hand sorting of paper, likewise, can be 
done very quickly. What is tedious and inaccurate is what seems to be 
a standard procedure for hand counting ballots: a person holds up the 
ballot and reads off of it the vote on each race, and a clerk tallies 
the vote. And that's it. Slow, and inaccurate. Multiple places to 
make mistakes, and no redundancy, no verification, beyond the 
observers. The observers cannot watch reliably every aspect of this.

I've never been in a counting room, so I don't know exactly what they 
do, I have only a few reports of this or that. But the procedure 
above was, in fact, described to me, and it is inherently unreliable 
and slow. I'm sure there are reasons it is done that way. But they 
may not be valid ones, on balance, and if we were considering 
switching to machine counting, we should also consider improving hand 
counting procedures.

And because one of the things that slows down and makes hand counting 
expensive is security, I've suggesting ballot imaging. If ballots are 
imaged, the ballot information becomes redundant. If one copy (the 
original or the image) is sequestered, alteration to the other 
becomes detectable or at least verifiable. If there is more than one 
copy made, alteration then requires altering *two* copies for the 
original vote to become unrecoverable.

I'm suggesting, bottom line, that the entire process be rethought.

I've heard only one substantial objection to ballot imaging, and I 
consider it spurious and misled or misleading. And that is the 
privacy of the voter. The privacy of the voter is not violated by 
ballot imaging. First of all, there is only even an issue if the 
images are public, which is merely an independent part of the 
proposal. Concern about privacy would only require, at most, that the 
images only be available to those with legal need to see them. 
Secondly, ballots do not identify voters. Even if they are not 
English speakers, an issue that was raised. Under circumstances where 
a special ballot would identify the voter, as being the only one or 
one of a few in the precinct, other arrangements can be made, such 
ballots could be transported to be counted with other similar 
ballots. There are risks involved in this, too, and so I would avoid 
special ballots entirely! And that can be done with no inconvenience 
to the non-English-reading voter at all.

I'm actually getting a little angry at the objections, because it 
seems that a good and simple idea is being rejected without cause.

The one difficult issue is the issue of vote buying or vote coercion, 
where a voter marks the ballot to make the vote identifiable. My 
contention is that this is already a problem, if there are those who 
wish to coerce votes or buy them, and public ballot imaging will 
actually make it more difficult to alter election results through 
such means, not worse. But this applies only to public imaging. 
Confidential imaging has no more risk than what we already have: the 
risk that those who have access to the ballots can verify how a voter 
voted. They can already do that with any paper ballot, just as they 
can do it with any voting machine imaginable, if they are determined. 
There are ways.

Paper ballots are probably one of the more secure ways to interrupt 
various forms of election fraud. The problem with high-tech solutions 
is that a lot is invested in one particular aspect of the problem, 
and then comes a hacker or spy or someone with time and resources, 
who figures out how to bypass the security, and it was all for 
nothing. Consider the Microsoft operating systems: a huge investment 
is made to make them as secure as possible, but the ingenuity of 
those determined to bypass security is endless. A way can be found.

What really must be done is not to single out one method and claim 
that it has this or that problem, but to balance all methods and the 
problems, to determine how such problems could be addressed with each 
method, and then determine the various costs of each solution. It's 
improper to reject one solution on its own, because in the real world 
what we have is always choices. It is not *this* solution or the 
ideal. It is a choice, almost always, between two non-ideal 
solutions. Each of them has flaws that could be used to argue against it.

Paper ballots have been around a long time, and the flaws with them 
are fairly well-known. The big problem with a new solution is that 
the bugs in it may take some time to discover.

Ballot imaging is really just an extension of the right of the 
interested parties to observe the election process. It's already 
possible for people, including people who might be in cahoots with a 
vote buyer, to see ballots. We cannot and should not eliminate this 
possibility by taking ballots out of public view, it is an open 
invitation to fraud, which thrives in secrecy and invisibility. We 
are far safer with the process taking place in public, with 
*everyone* having an opportunity to observe. Indeed, I would have 
webcams watching every aspect of the process, with the resulting 
video available in real time as well as being recorded. Vote counting 
should be a totally open and transparent public event. The actual 
voting should be as protected as possible.

I think that ballot imaging would make security higher, not lower. 
And I've seen nothing, so far, to make me question that. It's hard to 
coerce somebody if everyone is watching. Ballot marking is visible, 
by definition, if it is not visible, it would not be in the images 
and the images would not assist in vote confirmation.

Whereas it is quite practical to mark ballots invisibly, where 
ordinary examination would not see the marks but the conspirator 
could readily detect them. I can think of many ways, and some of them 
I understand have actually been used. For example, a small pinprick 
at a critical location. Then there are fingerprints, made even easier 
if the voter puts a UV flourescent substance on their thumb.

You can eliminate all these by using untouched forms of recording 
votes that cannot, theoretically, identify the voter. But the problem 
is that all these are circumventable, and a totally new problem is 
introduced: the possibility of altering votes with no discovery 
possible through audit. So what if the machine prints a record, which 
the voter deposits in a box, these records being available for audit. 
Fine. Does the voter touch this record? If so, the voter can be 
identified. Easily. It's getting easier all the time, as technology 
advances, but it's been relatively easy, particularly with a 
cooperative voter, for a long time, through fingerprints.

I think the best approach is to take a simple process, do it entirely 
in public view, and watch it closely. A pattern of vote buying and 
confirmation through marked ballots would show up like a sore thumb, 
with so many people watching. Thousands of eyes and intelligences 
involved. The criminals would be so outmatched that I think they 
would not even try. The system is vulnerable to corruption in other 
ways that remain accessible. Ballot imaging will have no effect on 
efforts to manipulate voter eligibility, for example, for partisan purpose.

If you want to spend thousands of dollars, instead of buying votes, 
spread lies about the opposing candidate. It is more efficient than 
trying to buy votes, which is generally too expensive to see wide 
use, and the latter is illegal and risky, the former less so. Less 
illegal and less risky. Ever seen an election overturned because the 
winner condoned spreading lies? Or a candidate convicted because of 
complicity in it?