In response to this post, my Loyola colleague and head of the Western Law Center for Disability Rights Eve Hill sends along the following response:

I saw your blog entry re. touch screen voting and the Secretary of State’s paper trail requirement. You asked about the theory behind any potential lawsuit on behalf of potential voters with disabilities. There are actually several theories, but they are not based on the fact that most counties will likely be priced out of buying touch screen technology and forced to use optical scan voting. Although optical scan voting machines are inaccessible to blind voters (and language minorities), they are not prohibited by either the ADA or HAVA. HAVA requires that there be at least one DRE at each polling place in order
to ensure that there is a means of accessible voting.
Of course, we don’t want counties to use optical scan. We prefer touch screens across the board – for a variety of reasons, including that if there’s only one accessible machine, it will undoubtedly be broken or lost or out of order or no one will know how to use it (this is what always happens with the “special” equipment for people with disabilities). I don’t know yet whether counties using optical scan with 1 touch screen at each polling place will be violating the law.
But the main theory at this point is that requiring that one touch screen machine per voting place to have a paper trail will violate the law (or, alternatively, where all the machines are touch-screens with paper trails). Right now, the touch screen machines give the voter a screen showing how she voted and, for blind people, an audio report (via headphones) of how she voted. Because the information is available in an accessible format and an inaccessible format, it doesn’t violate the ADA or HAVA. However, the Secretary of State’s requirement would require an additional paper printout that the voter must verify is correct and deposit with the poll worker. This second verification is purportedly necessary because, in case a recount is needed, the voter-verified printout will be a check on the electronic count (for a variety of reasons – some of which were discussed in my Daily Journal op ed on November 17 – I don’t believe that this is necessary or that it will actually help much). However, this second verification is not accessible to people with vision impairments. It is not printed in Braille or verified audibly (unless a poll worker reads it to the voter,
which obviously destroys the privacy of the vote).
The inequality is in having 2 points of voter verification, only one of which is accessible. Shelley’s position indicates that the paper printout presumptively trumps the electronic vote in any recount, but blind voters do not have the opportunity to check that paper printout. This inequality may raise an ADA violation (Title II). In addition, if the single DRE at each polling place has a paper printout, it may violate HAVA (because the voting machine that is required to be accessible to blind voters is actually not accessible to blind voters – ironically!). It may also violate state law. In fact, the California Attorney General believes requiring a paper trail will violate the ADA, HAVA, and state law.
So that’s the legal problem. The practical problem is that paper trails increase the cost of touch screens by $500 per machine, plus add paper, printing, and paper storage expenses, which make touch screens more expensive than optical scan machines. So we’ll end up with the least accessible technology (for people with disabilities, language minorities, and people who are illiterate) being the norm, rather than the better technology.

The Daily Journal article that Eve refers to is called “MOVING BALLOTS INTO AGE OF COMPUTERS; People’s Worries About Security Of Touch-Screen Voting Machines Are Exaggerated.”
This is the first time I can remember where there has been such a split between the voting rights community and the disability rights community.