Sorting ballots in Alameda County, California, 2008. Photo credit: Joe Hall / Flickr (CC BY 2.0)


The existence of paper ballots is generally touted as the ultimate backstop guaranteeing the integrity of American elections, because “if there is a problem or any doubts, those ballots can always be recounted.”

They can be — but will they be?

Now we have seen three “recounts” up close and learned that, in practice, this amounts to a false and dangerous assurance. The effort to recount these ballots, where they do exist, has been blocked, subverted, and turned into a sham in each of the three states in which it has been attempted this month.

The sheer number (and variety) of obstacles that have been thrown in the path of the recount efforts in the battleground states of Wisconsin, Michigan, and Pennsylvania begs the question: What evidence are these blockades trying to hide?

In the same spirit that Rosemary Woods managed to erase just those 18 minutes of an hours-long Nixon tape that many believe to contain the “smoking gun” about the Watergate scandal, so we are led to suspect the Election 2016 smoking guns may be in places that refuse to recount by hand — counties that destroy or prevent the creation of ballot images by scanners; states that ruleagainst recounting in precincts where ballot bag seals are broken, or the number of voters does not match the number of ballots; and states whose courts, by partisan majority, simply rule that the recount cannot go forward at all.

A combination of administrative, financial, judicial, and operational tactics were used to hamper or stymie the recount effort in each state in which it was undertaken. A few examples of these tactics:

•  Refusal to hand count in Wisconsin in the very counties with the brightest forensic red flags — Outagamie, Brown, Rock, e.g., where Trump vote shares dramatically exceeded expectations.

•  Determination by a Republican Wisconsin judge that state law barred her from orderinga statewide hand count despite her own preference for one as the “gold standard.”

•  Fees inflated in Wisconsin from an estimated $600,000 for statewide hand recount in 2011, to $3.9 million for a less laborious mixed machine/hand count in 2016.  Racine County, e.g., charging more than 50 times to count by machine what it charged for its last county-wide recount, which was done by hand.

•  Suits filed to block recounts in Wisconsin, Pennsylvania, and Michigan by the Trump campaign or surrogates.

•  Interpretation of Pennsylvania law to mean that more than 27,000 individual petitioners had to file in more than 9,000 precincts in order to proceed with a recount at all.

•  Pennsylvania state court requiring the posting of a $1 million bond by Green Party candidate Jill Stein to even consider her petition for a statewide recount.

•  Michigan lower court halting the recount on the dubious legal basis that Stein can’t win (there is no such requirement in applicable state law) and that no “credible” evidence of fraud (which of course was what the recount was looking for) had been presented.

•  Michigan Supreme Court affirming a lower court ruling on straight 3-to-2 party-line vote, concluding that Jill Stein is not an “aggrieved candidate.”

•  Bush-appointed federal judge stating in his ruling that it “borders on the irrational” to suspect hacking occurred in Pennsylvania, where the vast majority of ballots were electronic and non-recountable, while ignoring tens of thousands of provisional (paper)ballots, which could have been recounted.

This is not the behavior of campaigns confident of the validity of their victories. Nor of administrators confident of the security and accuracy of their electoral processes.  The result was a partial recount in some places and none at all in others, leaving the whole enterprise riddled with gaping holes and dark, inaccessible crevices where evidence may lie concealed.

Media coverage has ignored or soft-pedaled most of these roadblocks, misreported the actions and motives of Jill Stein, and adhered tightly to the “nothing to look at here, folks” theme. Jeffrey Toobin, writing in The New Yorker, went so far as to attribute the entire motivation for the recount to Jill Stein’s “narcissism.”

While allegations swirl about possible Russian interference in our election, the very process by which light might have been shed on such, or any, interference has been stopped in its tracks withbarely a raised brow or a whimper of protest from our supposedly free press.

The recount — a major and massive effort to protect democracy by shining light on an electoral process designed for concealment — has thus far been of dubious, if not negative, value.

But the lessons learned can be invaluable:

1) The American vote-counting process has been revealed as a process designed to conceal;

2) Recounts and, by extension, paper ballots, unless hand counted in public on election night, are in no way sufficient to ensure the honesty, accuracy, and fidelity of that process; and

3) American vote counting must, by public demand and public action, be brought into the light and made openly observable so that faith in the most fundamental of democratic processes can be restored.

Jonathan Simon is Executive Director of Election Defense Alliance, a nonprofit organization dedicated to restoring observable vote counting in electoral integrity. He’s also the author of Code Red: Computerized Election Theft in the New American Century.

Related front page panorama photo credit: Adapted by WhoWhatWhy from counting ballots (PBS NewsHour / Flickr – CC BY-NC 2.0).

16 responses to “Recounts Are Only as Good as They Are Allowed to Be”

  1. Termie says:

    This is ALMOST correct.. except for the thought that a hand count of paper ballots gives an assurance. The fact is hand counts are easily manipulated and manicured to falsely reconcile — The key is to get the ” first count”.

    • Virginia Martin says:

      If hand counts are conducted publicly, transparently, and fairly, and if the chain of custody is provably secure, then “manipulation and manicuring” will not be a problem.

  2. EmilyLevy says:

    Nevada’s presidential recount has barely been reported on at all. As with the others, news reports focus on the small difference in vote counts, not in the procedures not followed, machines that can’t be trusted, or laws broken. Like that Nevada’s Clark County (where Las Vegas is and most of the state’s votes are cast) conducted its recount in secret, then staged a public recount!

  3. David S says:

    In Detroit, 36% of the precincts (thanks to Jill Stein’s recount) showed that MORE PEOPLE VOTED THAN WERE REGISTERED. You know damn well those votes didn’t go to Trump. Time to stop with the “Trump stole the election” narrative.

    • Land Shark says:

      These Detroit precincts did not show more ballots than registered voters, they showed more ballots than voters RECORDED AS voting. There is a huge difference.

      You presume and assume these were “extra” votes when it could be that not everyone was crossed off the pollbooks as having voted, and all ballots cast were in fact legal. You also assume that the votes went to Clinton, when a fraud could have reduced Clinton votes by ten percent and still left more ballots than voters in the pollboooks, and cast the first suspicion elsewhere, which would be a smart thing to do.

      You further assume that the result of finding (or in your case, merely suspecting) that one side cheated or may have cheated should be that this particular side should shut up. But in fact, your assumption here reveals your own partisan attitude and motive. Any impartial person would continue investigation no matter who would be implicated. It is simply inexplicable to a fair minded person why finding evidence of something should END AN inquiry rather than redouble it.

      Simply put, if you were right, then evidence of potential cheating by any side only redoubles the necessity for investigation and transparency. But nobody is being allowed to look at these precincts to find anything out. And that’s a problem.

  4. Yong Liu says:

    Just investigate who is in charge in those counties.

  5. Wemble says:

    The simple reason that Clinton lost Wisconsin, no election fraud needed, is that too many voters on the Left either chose not to vote at all or chose not to vote for the one and only candidate that could have beaten Trump: Hillary Clinton. In my Wisconsin country the turnout was over 80% and Clinton won by 9.5%.
    But if hand recounts are so much more accurate, then surely there must be studies and evidence to back that up. Well, surprise, surprise, there is not.

    According to the Tyler Morning Telegraph (5 Dec 2016),
    “The same finding emerges in Wisconsin. When we analyzed the 2011 Wisconsin recount, we found that the average discrepancy for scanner-counted paper ballots was 0.17 percent, compared with 0.28 percent for hand-counted paper. In other words, both methods are highly accurate, but scanners are slightly more so.”

    • wifather2000 says:

      You are assuming that the machines were not rigged to miss or switch votes for one candidate. The only reason not to hand count would be because of rigged machines, not because hand counting would take too long or not be accurate. Running the suspected ballots through the same machines that may have inaccurately tallied the votes the first time proves nothing, except the fact that people who go along with this folly are incapable of self thought and could in fact be intellectually lazy.

    • Virginia Martin says:

      If the scanners are working properly, and of course if there’s been no tampering, then they may be a little more accurate than hand counters, although I would have to question how experienced the counters in that recount were. But scanners don’t always work properly, and the reason can be just because something goes wrong with them, inadvertently. Voting machines are aging, as we know, and computers, which they are, sometimes don’t work right. Sometimes brand-new scanners don’t work right. So, use the scanners and then do a hand count. Compare the two. Figure out what the difference is, and decide if the hand counters are right or wrong. Do this right after the election, and make it a very public, open exercise. Then no one will question the results. No more need for recounts. The hand count becomes part of the standard operating procedure.

    • Wemble says:

      In Wisconsin, the only state to do a statewide recount, 60% of the ballots were recounted by hand but it wouldn’t matter if the number had been 100% because a new conspiracy would arise to claim something was wrong such as “oh, they had too many people who are Republicans counting the ballots” or “well, the county clerk is a Republican so those results can never be acceptable”. It will ALWAYS be something.

  6. roberto di camerino says:

    Just investigate who is in charge in those counties.

  7. Randy Divinski says:

    Does anyone know whether POLLING LOGS in Wisconsin were examined? Reported voter turnout was suspiciously high in several of the counties that refused to hand-count. This is a red flag for ballot stuffing, but because Wisconsin is a same-day-registration state, it is also possible that there was a high “walk in” turn out. Examination of the “walk in” poll log would help clear this up.

  8. Virginia Martin says:

    If counties hand-counted to determine what the election results are (not to confirm, in a recount effort, what the machines said, but to determine the results in the first place), all this skepticism would disappear and these contested recounts, which have devolved into a circus show, would be unnecessary. We hand-count in Columbia County, NY to determine what the results are. It costs us about 1% of our annual budget, and in this election cost about $.13 per vote cast. Not very much for very, very high confidence in the results. Yes, we use optical scanners to get quick election-night results. But our electoral results are based on our hand count. It doesn’t take that long, isn’t that hard to do, and doesn’t cost that much.

Subscribe to the Daily WhoWhatWhy

Relevant, in-depth journalism delivered to you.
This field is for validation purposes and should be left unchanged.