Although election officials and other advocates of forced mail voting promise us that "signature verification" is a sound security measure, those who read this blog have seen plenty of examples of ballots that were both wrongly accepted and wrongly rejected. But in fact, the whole arrangement of signed ballots has a gaping loophole that bypasses the bogus signature verification process altogether and offers one of the most promising opportunities for mail-vote fraud -- the so-called "witnessed" ballots.
WAC 434-250-120 states that a mail ballot shall be counted if:
(3) The signature has been verified ... or if the voter is unable to sign his or her name, two other persons have witnessed the voter's mark;Surely it's fair to offer some provision for those who are eligible to vote but permanently or temporarily incapable of signing a proper signature. But the way this is implemented, it's an open call for vote fraud. All that is checked in practice is the appearance that the voter made a "mark" of some kind on the ballot envelope and the appearance that two different people signed the envelope. There is no provision that the witnesses identify themselves, only that there are two signatures (which effectively means any two squiggles that look like they might be signatures) --
I've found examples where even looser standards have held sway -- a case where a ballot signed by only one witness, was cast in the name of a legally incompetent nursing home resident -- and accepted. I found a ballot that was accepted in King County even though the two "witness" signatures were identical. Indeed, the requirement for two different "signatures" is merely theoretical -- without any requirement for the "witnesses" to identify themselves, any two different squiggles that resemble signatures have to be accepted. Envision the fascinating possibilities for fraud --
1) If one finds oneself in possession of someone else's unvoted mail ballot (perhaps the mailman delivered a neighbor's ballot to one's home by mistake, a former resident's or a deceased relative's ballot is still coming to one's home, or maybe one just lifted somebody else's ballot from their mailbox when nobody else was watching, etc.), one can vote the ballot and fill out the outer envelope using the "witness" method. All one has to do is make an "X" on the voter signature line, and put two different plausible signatures on the witness line. (For added realism to avoid suspicion, one might make the X with his opposite hand (i.e. the left hand, if right-handed) and use two different pens for the "witnesses".
[This raises an interesting question -- what would the elections office do if a voter calls before an election to complain that he never received his ballot, and the elections office discovers that a "witnessed" ballot had been submitted on the voters behalf? Would they issue a new ballot and count it if returned? Or would they maintain over the voter's protestations that he already voted?]
2) One might also pull the same stunt as in (1) with his own ballot upon receipt. One would wait until a few days before the election, having given the submitted ballot enough time to have been irretrievably processed through the system. One would then call up the elections office and report that he never received his ballot and ask for a new one. Regardless of the official procedure for dealing with such situations, there is an excellent chance that the elections office would end up counting the second ballot anyway, and it would be effectively impossible for anybody to prove that it was the nominal voter who pretended to forge his first ballot in order to vote a second one.
If Deanron and the Democrats on the King County Council succeed with their scheme for forced mail voting, a ballot will be mailed to every registered voter, including the very large percentage who choose not to vote and/or don't bother to update their addresses. That means that hundreds of thousands of unwanted and unsecured ballots will be floating around every election. The temptation and lack of disincentives should make "witnessed" ballots increasingly common.
Posted by Stefan Sharkansky at May 25, 2006 08:32 PM | Email ThisWhile granted in 04 we had a once in a lifetime margin of victory, but really how many people are going to go through all this nonsense just to vote twice. Hell most people don't even vote once.
While, yes we should do what we can to prevent fraudulent voting, I really don't see this as a major problem effecting elections. Especially given that voting is almost identical across the state meaning that the fraudulent votes would likely track the actual vote. Paranoid fantasy about King County aside.
Pinhead is also the name of Scottish steel cut oats, which is a delicious breakfast food. It gave me and my men the strength to defeat the British army.
Samuel Johnson referred, disparagingly, to this in his dictionary definition for oats:
A grain, which in England is generally given to horses, but in Scotland supports the people.
To which his biographer, James Boswell, is said to have retorted:
Which is why England is known for its horses and Scotland for its men.
One thing you forget is that there are thousands of people on the far left that have shown a proclivity to commit crimes (vandalism & arson) to advance the "cause". They probably have been doing it with votes too and now they can really do some work to make sure that a socialist eutopia will emerge and our City will become a pile of dung.
Posted by: Dengle on May 25, 2006 10:37 PMSignatures simply don't pass the test for being unique and personal enough to be used for authentication. In the computer world, we often speak of two-factor authentication. One factor is what you have, say an ATM card, and the other factor is what you know, say a PIN number. The two must be combined to successully authenticate oneself for retrieving money from an ATM.
Signatures have caused all kinds of problems in the credit card industry because they began use for authentication back when everything was done by hand and with paper and carbon copies. In a pre-computer pre-magnetic swiping world, there was no real need to use any sort of computer readable authentication, and although credit card fraud has always been a problem, it was much less possible when there was not the ambiguity of the Internet and the ease of duplicating magnetic stripe cards. Today, signature verification is a huge problem for the credit card industry. Most people are now familiar with the extra 3 digit code called the CVV2 code on the back of their card on the signature line. This number is serially generated by the issuing bank when the card is created and only exists on your card, for the given 16 digit number on the front of the card. Thus, this number is sort of like a PIN, as long as your card is never lost or stolen.
The reason the credit card industry has had to move away from signatures, and the whole reason for my comment is that signatures are very difficult to verify. Especially when they are written on paper and simply compared by a human or machine. While a trained eye or a machine can look for similar loops and other distinguishing features of a paper signature, ultimately, as any teenage kid will tell you, it's pretty easy to forge a credit card signature. Since I share the same name with my Dad, I routinely signed for his card all through high school. I had his signature down to the point where even he could not tell the difference. Of course I did this with his permission as a means of allowing me to pick up purchases for him without him having to be present. But the point is, signatures are not good enough to authenticate anything. The only time this would not be true is if both the original "on-file" signature, and each subsequent authentication signature were performed on an electronic touch pad. Then, a computer can actually record not only the swirls of the hadwriting and the outline of the signature, but also the speed, pressure and timing of the pen stroke. This would be enough to uniquely identify someone because it is biometric and much, much harder to duplicate.
But let's be clear, with King County's crappy, out -of-date paper registration file and human or machine simple visual comparison, there is simply no authentiation at all when using signatures, and less than zero authentication when using the "witness" method that Stefan describes in this post.
In short, the system is fundamentally flawed, easy to cheat and entirely dependent on the honor of citizens and elections personnel to conduct themselves with integrity.
Does it occur to anyone to wonder why banks don't do it this way? Or why almost any other type of transaction we deal with today is more secure than our elections?
The absolute minimum for voting should be picture ID verification from either an official state of WA license or ID card, or a US Passport. And combined with this, we should have poll voting only, except for the permanently infirmed or disabled who can't make it to the polls.
If we are going to allow any sort of general absentee vote by mail, it ought to be secured with a fingerprint or PIN number or some other legitimate electronic form of two-factor authentication. Signatures are simply not enough. Anything less than legitimate authentication will gaurantee the result of any close election will always be in doubt.
The simple and cost effective solution is simply to require poll only voting. And while we are at it, let's require that everyone dunk there finger in purple die after dropping off their ballot. At least then the Moonbats would have to chop off a finger to vote twice for Gregoire.
Posted by: Jeff B. on May 25, 2006 11:04 PMRE Dangle: I would also highly doubt that are are thousands of people in washington (or the whole country for that matter) commiting vandalism and arson for political reasons. We have our crazies just like you do, or lest we forget clinic bombers, crazy militia members, and gaybashers.
Posted by: Giffy on May 25, 2006 11:06 PMI have many Scottish friends....I would never intentionally insult any of them. Food is important. Those "steel cut oats" made the Scots a hearty breed....as well as mighty regular! If the Scottish Army all ate them for breakfast together, I'll bet 15 minutes after a steaming, hot bowl of those oats, you Scotsmen were literally mugging each other as you all raced to the sh*tter at the same time!
Posted by: Mr. Cynical on May 26, 2006 02:19 AMStefan is one guy, who can only look at a small fraction of all the ballots cast. That he's finding as many bogus ballots as he is suggests that there are a lot of bogus ballots he hasn't found yet.
Further, Stefan is only likely to find bogus ballots that are fairly "obvious". Is there any reason to believe that none of the people faking ballots are taking any measures to avoid detection?
Posted by: supercat on May 26, 2006 03:34 PM"entirely dependent on the honor of citizens and elections personnel to conduct themselves with integrity."
hmmmmm . . . geeee . . . lets get Teamster's Union people to count our votes!!
friggin' jack-a$$ shameless blind liberal apologist for petty criminals and butt pirate liberal Democrats like eric will love that idea.
Posted by: Amused by liberals on May 26, 2006 03:54 PMGiffy says . . . it seems unlikely that people do what they do . . . and there are millions of ballots . . . so . . .
. . . in 04 we had a very tight margin where victory hung on very few votes either way but . . .
” really how many people are going to go through all this nonsense
(all this nonsense of filling out a ballot and putting it in the mail) just to vote twice?” . . .
fraud might be bad . . . but . . . I don’t see it . . . so . . .
. . . it must be a fantasy.
Gee thanks Giffy, I see where you’re coming from dood.
If different names, birthdates, and addresses are used on each registration, how can they be matched up?
Another factor that camouflages the scope of double voting is the (so-called) mail ballot reconciliation process instituted by Logan after he took over at KC Elections.
The proper method of reconciling (accounting for) mail ballots is to start with the number of ballots returned, and subtract the number of ballots rejected. This number can then be compared with the number of ballots counted. If the two numbers don't match, then ballots were added or lost during the counting process.
Also, the number of ballots returned can be compared with the number of ballots issued. And the number of ballots issued can be compared with the number of registered voters. These checks are useful, in that they may indicate anomalies in the system.
Unfortunately, the process instituted by Logan simply takes the number of ballots counted and adds it to the number of ballots rejected. Unsurprisingly, when this result is compared with the number of ballots "processed", they match perfectly.
(For a better explanation see
http://www.soundpolitics.com/archives/004619.html
and also Susan B. Anthony's comments at
http://www.soundpolitics.com/archives/004615.html)
Logan's process also tries to avoid reporting the actual number of ballots returned, and the number of ballots issued.
If the mail ballots aren't properly handled, counted, and reconciled, double voting can be detected, but the extent of it can't be determined.
Which is why it is a bad idea to increase the amount of VBM at all, much less go to 100% VBM.
Posted by: ewaggin on May 26, 2006 10:23 PM