[EM] danger of coercion (Re: First U.S. Scientific Election Audit...)

[Abd ul-Rahman Lomax]

At 09:07 PM 5/16/2007, Juho wrote:
>On May 16, 2007, at 18:26 , Abd ul-Rahman Lomax wrote:
>Yes, there are ways to *reduce* the possibility of a coercer
>>verifying that the victim complied. None are guaranteed to work.
>
>What would be the most likely scenario where the coerced person could
>not avoid being coerced? I'm assuming that a typical coercer would
>not be a member in the team that counts the votes and he/she would
>not have open access to the ballots.

That's not reliable.

>  I also assume that vote counters
>do not necessarily count all the ballots but maybe just a relatively
>small part of them (and if really needed, we could use counters
>coming from another voting district), and that they do not have any
>special equipment nor much time to study the votes one by one. And I
>expect the votes to be packed and sealed after they have been counted.

Ballot counting procedures vary. Sometimes a vote counter counts all 
the ballots, in the presence of observers. Special equipment could 
include a concealed camera, and with such, not much time is required. 
The counting process takes time and ballots can be exposed for 
substantial periods of time. Seals are also possible to break and 
replace, with determination. None of this is very likely in any given 
circumstance, but when the government is in the hands of those who 
would steal elections, it does happen.

Coercion is, quite simply, not a common problem here. And coercion 
remains difficult to enforce without detection. Ballot imaging does, 
I've acknowledge, make isolated coercion easier, but isolated 
coercion is not likely to have any significant effect on election 
results. As to the harm to those coerced, I'd suggest that they are 
already in harm's way, and with coerced voting, they could, if they 
think best, simply comply. They could also comply and report it and 
cooperate with authorities. If they trust the authorities.

>>If the voter can handle the ballot at all, then it is possible that
>>it could be marked in a way likely to escape notice. Small
>>pinpricks have been used, apparently.
>>
>>Fingerprints can be used! And restricting access to ballots may
>>seem to work, but, as I pointed out, who watches the watchers? The
>>ballots are in the custody of someone, typically.... the
>>government. If you don't trust the government, if the coercion is
>>coming from an incumbent, what are you going to do?
>>
>>You are going to have to rely upon the fact that keeping that
>>incumbent in power through coercion depends on the fact that to be
>>effective, the coercion would *usually* have to be widespread. And
>>people tend to dislike being coerced.... you will be alienating
>>increasing numbers of people, and, if the government is at all
>>functional, and coercion is illegal, the arrest and prosecution of
>>those who attempt to coerce is the best remedy.
>>
>>Once again, coercion does not seem to be a problem in the U.S. I've
>>never heard of it in recent times! But voter fraud is not uncommon
>>-- the registration and voting of people not legally qualified, or
>>the mysterious voting of people who have died -- and, perhaps even
>>more common and more serious, election fraud, where ballots are
>>altered, or, more frequently, casting ballots by lawful voters is
>>impeded selectively, and properly cast ballots are not counted
>>correctly. (Or voting machines mysteriously change votes.
>>Selectively. It is bad enough when it happens at all, but when it
>>somehow seems to preferentially affect voters of one political
>>stripe, we certainly have grounds to suspect criminal activity.
>>It's fairly simple: just cause malfunctions to machines in
>>precincts loyal to the party whose votes you wish to damage.
>>Difficult to prove.
>>
>>It is this counting fraud that I am seeking to interdict.
>
>The methods I recommended for coercion prevention would be quite good
>for this purpose too.

I don't think so. However, a major problem is being overlooked, which 
is public confidence in elections and election results. That 
confidence is at a low point recently, due to what really looks like 
fairly massive manipulation of elections. Some of that is in the counting.

>I think most stable democracies do not have any meaningful problems
>with fraud in vote counting. This should not be impossible to achieve
>if one just wants it.

Be my guest.

We have a meaningful problem here. And it may well be fraud.


>>My suggestion is to allow public scanning or photography of the
>>ballots.
>
>Is this really needed?

I don't think you've though it through. This was discussed at length 
on the Range Voting list....

>  What is the main use of those records? Are
>voters supposed to be able to recognize their own ballots?

No. Of course not. Most voters won't look at them. But any voter can 
verify some portion of the count, and can look at databases compiled 
with the counts of others, plus the official count. This is why the 
ballots need to be serialized, so that anyone can verify any 
particular ballot count. With large numbers of people, each one 
counting a few ballots, you will have multiply redundant counting. 
Discrepancies will be flagged quickly.

>  This would
>be the same as having marked ballots! (Write-ins would be quite
>recognizable but in most elections they are maybe not really relevant
>and need not be supported. Write-in votes are also revealing in the
>sense that the write-in candidates probably get relatively few votes
>each.)

Write-ins are considered essential to democracy here, bypassing the 
nomination process. Sometimes write-ins win. in some elections no 
candidate qualified for the ballot so all there are is write-ins. 
That was the situation with the office my wife ran for. ("running" 
consisted of agreeing to accept the office when some neighbors 
suggested it; one of them was the outgoing officer. We think she 
received up to four or five votes. Making her the winner, unless 
someone else's votes were likewise not counted....)

Write-ins occasionally win elections. Don't discount them.

However, as I've written, if there is a concern that write-ins would 
be used to identify ballots, which is harmless in itself as far as 
I'm concerned, then there would be ways to implement a simple method 
of identifying write-in candidates. It requires registration, up to 
the last minute, of *all* candidates. It is not exactly write-in, but 
close enough. A booklet is produced with the names and a code for 
each name. What the voter does on the ballot is enter or mark the code.

The point of ballot imaging is that it becomes possible for the 
public to count the votes. If the public can count the votes, it can 
verify the official counts or detect corruption in them. It then 
becomes far easier to initiate legal action to overturn the official 
results. The needed images are already in hand, and the court can, 
where necessary, then compare the images with the actual ballots. But 
it won't even get to that point, I'd predict. I think that all 
attempts to alter counts would cease. They would become futile. And 
they are illegal. Personally, if I'm going to do something illegal, 
I'd prefer to at least get some profit out of it!

I have mentioned schemes that could be used to prevent a coercer from 
knowing for sure that the victim did not comply. I won't repeat them 
here. It's possible, if there is real fear of vote coercion where the 
voter cooperates by making the ballot identifiable, to interrupt 
that. I just don't think it is necessary, and it is not cost-free. 
Ballot imaging is cost-free.

Further, the official counting should be done with images as well. 
Thus the actual handling of the ballots would be reduced. It will, of 
course, be asserted that images can be altered more easily than 
ballots. That's true. Which is why we would have multiple image 
sources, including the public and media, and why the ballots 
themselves remain under tighter security than is possible if they are 
being hand-counted. And machine counting simply alters the 
oppportunities. Optical scan ballots essentially are imaged, whether 
or not the image is saved. It *could* be saved.... by open intention 
or otherwise.


>>There is no serious risk to privacy introduced by public ballot
>>imaging.
>
>The first one in my mind is write-ins where the handwriting is
>recognizable.

Yes. But you can easily avoid write-ins, or use block letters.

One needs to understand *who* needs privacy.... and what options they 
have. If the ballots are machine printed, the handwriting issue 
disappears, and that is likewise a low-cost option.


>>There is a small increase, perhaps, of risk of occasional -- not
>>organized -- vote coercion, due to the increased possibility of
>>validation by the coercer through subtle differences in marks or
>>other techniques. But any substantial increase in this would be
>>detectable. The coercion risk is not a privacy risk. You cannot
>>maintain privacy where the voter cooperates with the coercer.
>
>In good voting methods/processes it should be and is quite hard to
>prove to others how you voted yourself.

It has always been relatively easy. To prevent it entirely, you have 
to outlaw write-ins, you have to prevent all physical contact between 
the voter and the ballot, and you still can't avoid the possibility 
of special voting patterns, which will get easier to pull off with 
Range. Even IRV makes it easier.

>Of course cameras and video equipment should be banned in the voting
>location. It is not possible to guarantee 100% that such recording
>will not take place but one should try.

I disagree. And nobody searched me for my cell phone when I last voted.
>Combining multiple elections in one ballot is a risk.

Sure. But it is absolutely the norm. In fact, I've never voted in a 
public election where there were not many races on the single ballot.



>The risk of allowing access to the ballots to everyone is much
>riskier than having multiple vote counters (maybe not local people)
>each counting a small portion of the votes.

You say so. What is the evidence?

There is a lost performative in the last comment from Juho. "In 
having multiple vote counters"....

*Who* is going to "have" multiple vote counters?

The fact is that this is what I'm suggesting: multiple vote counters!

Ballot imaging would put the public in the position of being an 
election observer, to a degree.

By the way, ever try to get correlated election data? Did people who 
voted for Bush vote against the school bonds? That kind of data has 
plenty of legitimate use. And you can't get it now, unless you are 
willing to spend prohibitive amounts of money to get it....