If allowed to stand, the reasoning behind U.S. District Court Judge Mark A. Goldsmith’s December 7, 2016 decision [PDF] in Stein v. Thomas to halt the Michigan presidential “recount” is flawed, at best. Issued, ironically enough, on the day we commemorate what President Franklin D. Roosevelt described as “a date which will live in infamy”, it is by no means an exaggeration to suggest that Judge Goldsmith’s reasoning could inflict greater harm on the very foundations of our constitutional form of democracy than that inflicted by the Dec. 7, 1941 attack on Pearl Harbor.
The halt to the “recount” came just two days after Judge Goldsmith issued a temporary restraining order (“TRO”) directing the MI Canvassing Board to immediately commence the “recount” and one day after a U.S. Sixth Circuit Court of Appeal decision, upholding that TRO.
Under that 6th Circuit appeals ruling, Judge Goldsmith was obligated to revisit the issue if “the Michigan courts determine that Plaintiffs’ recount is improper for any reason.” Separately, on Dec. 6, the Michigan state appellate court ruled that, under MI law, only a candidate who has a reasonable chance of winning has a right to initiate a post-election count. But that state court ruling, by three Republican judges, did not justify Judge Goldsmith’s decision to halt a “recount” that had been predicated on Dr. Jill Stein’s rights under the U.S. Constitution.
As he acknowledged in his original decision, the Green Party Presidential candidate did not base her federal claim on state law. To the contrary, in his initial finding, Goldsmith held that the Plaintiffs had shown a likelihood of success on the merits of their claim that the two-business day waiting period mandated by state law “would likely violate their right to vote under the First and Fourteenth Amendments.” Judge Goldsmith, in that first decision, added, “the [federal] right to vote, and to have that vote conducted fairly and counted accurately” [emphasis added] is not merely “fundamental” but serves as “the bedrock of our Nation.”
State law, whether directed at the timing of the recount or to the aggrieved status (standing) of the candidate seeking the count, should not be allowed to infringe upon a fundamental right that every citizen has to a verifiably accurate count of their votes.
The truly damaging aspect of the decision to dissolve the TRO lies not in the question of standing but in Judge Goldsmith’s upside-down reasoning as to who should bare the burden of establishing the integrity of the vote. That reasoning is directly at odds with the rulings made in two landmark cases in Germany and Austria, to the effect that the need for election integrity and transparency are paramount in any nation that values democracy…
Backwards burden of proof
As Brad Friedman reported when Stein first announced her intent to seek a “recount” in three states (WI, MI and PA), the Green Party Presidential candidate cited “the many concerns brought to her by computer scientists and voting systems and election integrity experts, all questioning whether paper ballots were counted accurately by error-prone and easily-hacked computer tabulators in WI and MI, and whether touch-screen systems were manipulated in some fashion in PA.”
Stein was seeking a first-time hand-count of the optical-scan paper ballots, where they exist, along with an independent forensic audit of the touch-screen voting systems by computer experts. Her goal, she said, was to “verify the vote and be confident our votes were counted.”
Judge Goldsmith did not dispute the validity of Stein’s concerns. To the contrary, he acknowledged them, even while going on to rescind the TRO which would have allowed for either verification or refutation of the accuracy of the unverified machine tallies.
In his Dec. 7 decision, he wrote:
Judge Goldsmith has utilized a backward application of the burden of proof that can only serve to undermine a fundamental right that the court, itself, acknowledged “is the bedrock of our nation”: the right of the people to know that their votes were “accurately counted.” Short of an impossible result, like the negative 16,022 votes recorded for Al Gore on Volusia County, FL’s optical-scan system during the 2000 election, the only reliable means for verifying or refuting whether the official optical-scan count is the product of “significant fraud or mistake” is to have human beings publicly hand-count the paper ballots. To require proof of fraud or machine error before ordering a hand-count, or a forensic audit, is to put the cart before the horse.
Real issue is transparency
When it comes to elections, the word “fraud” is frequently invoked but rarely substantiated. Oh, there certainly have been instances where insider election fraud has been proven. See, for example, this site’s coverage of the guilty verdicts handed down against high-ranking Clay County, KY election officials for multiple counts of felony election fraud. Those officials had illegally manipulated elections, including voting systems, for decades. But proving actual fraud is far too time consuming to be of any use within the context of the severely limited time constraints of a recount in a Presidential election. (In the Kentucky case, the guilty verdicts were handed down more than one year after the initial arrests).
When it comes to election integrity, the key issue is not to establish actual fraud, but to operate a transparent system that significantly reduces the opportunity that is afforded by e-voting systems for either hacking by outsiders or the rigging of the ultimate count by an insider. “I follow the vote,” CIA cybersecurity expert Steven Stigall informed the U.S. Election Assistance Commission (EAC) during a 2009 presentation, “and wherever the vote touches electron and touches a computer, that’s an opportunity for a malicious actor potentially…to make bad things happen.”
When it comes to what Judge Goldsmith acknowledges is the “fundamental right” to an “accurate count,” U.S. courts would do well to follow the lead of Germany’s highest court, which, in 2009, banned all forms of e-voting. There the issue was not actual fraud but transparency. The court ruled that there is a “constitutional right to a public observed count” — a right which would be denied by all forms of unverified e-vote counting, which, of course would include, optical scan tabulations as well as 100% unverifiable touch-screen (Direct Recording Electronic, or DRE) systems.
Earlier this year, Austria’s Constitutional Court extended this core transparency principle to the counting of absentee ballots. The court went so far as to vacate the results of a Presidential election and to order a re-do of the election. Although, in that instance, the absentee ballots in question had been hand-counted, Austrian election officials had failed to comply with an election law requirement mandating two officials be present to personally observe the unsealing, opening and reading of the absentee ballots so as to insure a verified count.
The Austrian court ruled that the question as to whether someone illicitly manipulated the count “is of no relevance. All that a challenger need show is that the number of votes in which election officials failed to adhere to that nation’s strict procedures is high enough that it may have altered the outcome.”
If the U.S. truly aspires to genuine democracy, our courts should pay heed to the German and Austrian decisions. Whether or not there’s sufficient time to overturn Judge Goldsmith’s latest decision and to carry out a Michigan hand-count, his reasoning must be rejected as unsound. Anything less will serve to destroy a core pillar of our constitutional democracy.
Ernest A. Canning is a retired attorney, author, Vietnam Veteran (4th Infantry, Central Highlands 1968) and a Senior Advisor to Veterans For Bernie. He has been a member of the California state bar since 1977. In addition to a juris doctor, he has received both undergraduate and graduate degrees in political science. Follow him on twitter: @cann4ing
Well written. But I cannot help but wonder what youand others would say if HRC had won. I believe you would have criticized Trump for wanting a recount.
Joe D @ 1:
I can’t speak for Ernie, but as far as me (owner of this site), you’d be wrong. You can see my coverage of Republican NC Gov. Pat McCrory’s recent call for a “recount” in his failed re-election bid as evidence. I supported that, just as I have supported many other such calls, even by Republicans. (And, as I also argued before the election, Trump would be correct in demanding one after it.)
There are several issues of fact that can be refuted in this article:
1) The burden of proof is virtually always on the plaintiff. If Stein wanted a recount she needs to establish a prima facie case that there was fraud or mistake. If she cannot do that, which she can’t, then the court is well outside it’s proper role in ordering a recount. She can petition the government to validate the claims later.
2) Humans are not reliable. I can’t think of a single task in which humans are completely reliable. This holds true for election recounts. Having said that, Wisconsin chugs along showing very very few ballots were incorrectly counted (most of those miscounts would be on the part of the human recounters).
3) The voters fundamental right was observed: the votes were counted. Again, Wisconsin establishes that there will be virtually 0.0% change by a hand recount. Before the court could weigh in and challenge the counting method it would need to be established that there was sufficient fraud or mistakes to warrant a judicially ordered change in the counting process.
4) This is all caused by the hurt feelings of sore losers. Stein doesn’t think anything will change because of her recount, she just wants to hold up the electoral process and make a name for herself.
For fun here are the mistakes found in Wisconsin so far (all human caused):
Dane County: 11 Absentee ballots were left in their certificate envelopes on Election Night for the City of Madison Ward 2.
City of Verona Wards 2-4 results have been removed from the Day 10 update pending further clarification from Dane County Board of Canvass.
Oconto County: Increase of 10 votes in City of Gillett Wards 1-3, increase of 19 votes in Town of Abrams Wards 1-3, Increase of 33 votes in Town of Little Suamico Wards 1-8, Increase of 24 votes in Town of Pensaukee Wards 1-2, and an increase of 12 votes in Town of Spruce Wards 1-2. All increases in Oconto County are attributed to the use of non-standard pens or ballots being marked incorrectly. Voter intent was determined during recount canvass.
Outagamie County: Net decrease of 14 votes in City of Kaukauna Wards 6-7 and net decrease of 12 votes in City of Kaukauna Wards 8-10 due to ballots being run through the tabulator twice on Election Day in error.
Addition of 2 votes to Day 10’s City of Appleton Ward 33 totals. The polling location includes Winnebago County reporting units and 2 ballots Ward 33 ballots were found during Winnebago County’s recount. These ballots were delivered to the Outagamie County recount canvass.
Ironically, the results and processes in Michigan, Wisconsin and Pennsylvania were all sufficient in 2008 and 2012. There was no “fundamental right to an accurate count” talk then.
Joe D @1–
Ernie will probably speak for himself, but I’m gonna bet you are dead wrong in that belief.
There are some of us out here who actually just want to know who really wins and who really loses our elections. Our hope is that at some point enough Americans will realize that until we have the kind of justifiable certainty in election outcomes that they do in Germany and Austria, we can claim to be living in a democracy. Then, perhaps, we will find the wherewithal to demand the very simple voting system reforms that would provide us with essential election transparency and integrity.
I believe, Joe D, that you’ve missed the thrust of my argument. As Brad is fond of saying, election integrity isn’t about Left or Right. It’s about right and wrong.
That was the core point I made when I covered the decision by Austria’s Constitutional Court to overturn that nation’s presidential election. Even though that decision gave Norbert Hofer, a right-wing extremist with neo-Nazi connections a second bite at the apple, I applauded the decision because it supported the core principle of transparency.
DGAF at #3 and #4–
1. What part of error-prone, easily hacked, programmed in secret and therefore not to be trusted(except on blind faith) voting machines do you not understand?
2. Please do your homework and look up what any number of computer and cyber security experts have to say about the unreliability of our voting machines.
3. The way our system is at present the STARTING point is that there is no reason, other than blind faith, to believe our computer tallied election results are accurate.
4. What part of until you check to see if the error-prone, easily hacked machine tallies are correct or not can you possibly prove fraud or error do you not understand?
5. Humans are fallible, yes. Humans also built and programmed the machines, read the machine tallies, create the hackable centralized computers that tallies are sent to, and the hackable internet system that’s used to convey so many vote tallies. Do you have a reason to not distrust human fallibility in the enormity of all that fallible context?
6. The point is not to rely on fallible humans OR machines. That’s why we want publicly witnessed hand counted paper ballots, hand counted by teams from both sides and immediately displayed for all to see at each precinct. Humans may still screw up but it can be enormously minimized.
7. Whenever there is an agreed upon urgency to determine as accurately as possible an election’s outcome, as in Minnesota and the Franken/Coleman contest of a few years ago, what is the universally acknowledged most reliable method that is used? Why?
Dear Ernie,
So do we have nothing in our Constitution, as it sounds like Germany does, requiring election transparency?
Do we have anything in legal terms that we could use to argue for the replacing of the machines with a system we could trust?
I thought/hoped Goldsmith was initially making a standard, easily recognized as valid point about our right to have our vote counted properly and to know that it was until he swallowed that Catch-22 pill. But what exactly do we have written in law about this? Anything? Or was Goldsmith just making a reasonable, common sense interpretation of what should be our right but actually isn’t written down anywhere(and then changed his mind after the Catch-22 pill)?
All those people who donated over 7 million dollars for this recount are aggrieved too, and certainly not due to Jill Stein.
What about us?
DGAF at #3 writes:
Wrong! The burden of proof always lies with the party who takes the affirmative in their pleading.
In this case, Jill Stein did not affirmatively allege fraud. Instead she alleged that e-voting is vulnerable to fraud, manipulation and mistake. She presented affidavits from experts that substantiated that allegation, thereby justifying the need for a hand-count and a forensic examination to verify or refute the validity of the official count.
Re: David Lasagna @8. There is no express provision in the U.S. Constitution that refers to “transparency.” However, if Judge Goldsmith’s conclusion in Stein I that citizens have a constitutional right to have their votes fairly and accurately was correct, that reasoning would justify a decision by the U.S. Supreme Court to arrive at the same conclusion reached by Germany’s highest court in its 2009 decision to ban all forms of e-voting.
The fact that the U.S. Constitution, which was drafted in the 18th century, does not expressly refer to “transparency” is not a cause for concern. The framers could not possibly have envisioned the advent of e-voting.
You guys don’t like it when someone disagrees, huh?
@ Ernest
The plaintiff can also be referred to as the complainant. In this case its Stein. She needs to establish that fraud, manipulation and mistake actually happened. Not that a group of sore losers want to make political hay. There have been plenty of recounts, not one single recount has found the kind of egregious errors you’re praying for. Not one. Voting machines are better than humans hand counting by an order of magnitude.
Also, Wisconsin disproves her (and you).
@ David
Everything you wrote is a conspiracy theory. Why didn’t the hand count in Wisconsin show massive fraud, manipulation and mistake? You can’t, and won’t, answer that. I posted all the mistakes found so far. Not a single one was due to your list of conspiracies; all were human error.
No matter how you slice this, try to reframe, pull in extraneous fact or otherwise bluster the vote was accurate the first time; and Trump won fair and square.
DGAF @ 3 said:
Well, there have been mistakes in the vote counting shown in the initial days of the MI (and WI) count. So, if, according to the same MI state election law, if a precinct pollbook differs by even a single vote from the number of ballots in the box, the entire precinct may not be recounted, why shouldn’t a single mistake in the computer reported vote count lead to a human examination of the ballots? In other words, by your interpretation of MI standards, it sounds like the case has already been made that there was errors in the initially reported computer counts.
Setting aside the consensus from the world’s top computer scientists and voting systems experts that disagrees with you, are you suggestion that if an election result shows a one vote margin based on the initial computer tally, we should just go with that without having humans check it? Or you are suggesting a second computer tally would be appropriate? If so, what happens if that second tally shows a one-vote margin for the other candidate? What then?
Actually, most of the ballots are simply being re-tallied by computers in WI. But, among the ones being hand-tallied, there are hundreds (perhaps thousands, we don’t yet know) of votes mistallied originally for both candidates. You don’t seem concerned about that. I am. I think votes should be counted accurately. Either way, even if every vote was found to have been tallied correctly the first time by computers, how that demonstrate that votes should not be examined by humans? (Or, are you saying we should have gone with the computer results when they reported only 16 votes were cast in a municipal election in the state in 2014? And that the losers named as “winners” by the computers in Palm Beach County should have been allowed to serve?)
Okay. Well, this site has supported the calls for hand-counts by many candidates that we abhorr (most recently NC Republican Gov. Pat McCrory). But if it’s more convenient to pretend the above — presumably when your candidate “won” — I’ll leave that to you.
DGAF then said @12:
It seems the folks here like ya fine. That’s why they took the time to politely explain how you are wrong. But…you don’t like it when someone disagrees, huh?
Turns out the bank near where I live did not get robbed last night. Therefore, banks cannot be robbed. No need to waste money by protecting or deterring bank robberies! Hooray!
Sigh…
Yup, the Trump team sure is acting like they’re confident they won fair and square. Yup, believe me. And I got a coupla towers I’d like to sell you in Cancun.
https://medium.com/@nick_sharp/what-i-saw-at-the-michigan-recount-7c46fdc87243#.4s86ss983
I observed part of the “recount” in Waukesha WI (where shenanigans and problems have occurred previously) one of the 4 most populous counties: The votes aren’t being hand-counted, and the optical scanners being used are not the ones used throughout the county, which are sitting in storage wherever they go between elections. Therefore, not only are we not going to get the true election results verified, but even the machines that are re-counting the ballots aren’t the same actual machines used in the first place – they are the same kind, though – you know, the type that are way past their useful life, error-prone and easily hack-able… Much more prone to irregularities than the many, many polls showing a comfortable margin of victory for the Democrat candidate in WI, MI & PA. Add to that the CIA assessment that Russia intended to help Trump win through hacking (like they did in the Ukraine), and I am extremely un-convinced that all is A-OK with the vote…
DGAF,
Ernie, in this great post of his, is pointing out that it was democracy that lost.
Brad, et alia, who commented are merely showing you the ins and outs of the danges of computer hacking, which has some extremely accomplished musicians playing for the anti-democracy league (Ye Olde Drone Hackers Ride Again).
Democracy is a process requiring supremely exquisite procedures that are well worth it.
When democracy loses, we will all become sore in many more ways than one.
Losing freedom hurts everyone.
DGAF (clearing up the typos),
Ernie, in this great post of his, is pointing out that it was democracy that lost.
Brad, et alia, who commented are merely showing you the ins and outs of the dangers of computer hacking, which has some extremely accomplished musicians playing for the anti-democracy league (Ye Olde Drone Hackers Ride Again). [even drones can be hacked]
Democracy is a process requiring supremely exquisite procedures (that are well worth it).
When democracy loses, we will all become sore in many more ways than one.
Losing freedom hurts everyone.
PA’s voting machines are that old that they have no connection to the internet which prevents “russian hacking”. The only voting discrepancy was that straight republican votes were flipping the presidential selection to Clinton early on election day.
Tim @18 wrote:
It would be helpful if people like Tim actually bothered to read or listen to what computer scientists have to say.
As Prof. J. Alex Halderman recently observed:
Thus, the fact that PA’s 100% unverifiable touch screens are not connected to the Internet in no way renders them invulnerable to hacking either by the Russians or anyone else. More importantly, the absence of an Internet connection does not prevent an election insider from rigging the results via computer programming. See, e.g. The Clint Curtis story — an instance in which, at the request of former GOP Congressman Tom Feeney, Curtis, a computer programmer, designed a software system that could (absent access to the source codes) undetectably rig an election.
Also, while there were some election day stories about Trump voters seeing their votes flipped to Clinton on screen, it really doesn’t matter what the screen displays. What matters is what takes place within the bowels of these electronic black holes. Absent a thorough audit by experts, there is no way to know whether the official electronic tally has anything to do with the actual votes cast on those touch screens.
Tim incorrectly stated @18:
That is simply untrue. Here’s how just one county found in 2011 that an “unknown computer” had “remotely accessed” the central tabulator in Venango County, PA for hours at a time “on multiple occassions”. The system is still being used today.
Moreover, if actual science is actually of interest to you, see Princeton computer science Professor Andrew Appell, who has hacked many of these systems, explaining how “all of them” are vulnerable to Internet hacks.
That’s not true either. But I suspect you don’t actually care.
Tim emerges from the principal’s office, red and sore.
The above exchange looks like a bit like a Thesaurus-aided troll-pack attack, with an orange-tint.
It’s the right thing to be polite to them. And honestly try to answer their “questions”.
But ofcourse they’re not really looking to communicate.
Only obfuscate.
Weaponized questioning—So different from actual discussion.
I wouldn’t call them “deplorables”.
They are, more accurately, “mechanicals”.
Just doin’ their job.
Dittoheads by another name.