Can Dead People Vote in Hawaii? You Betcha!

Photo courtesy of
article top
Photo courtesy of

BY KENNETH R. CONKLIN, PH.D. – A commonly heard joke about political corruption in Chicago is that politicians tell their supporters “Vote early — and often!” Another joke about Chicago is that lots of dead people vote. But that’s no joke.

Can dead people vote in Hawaii? The answer is: Yes, definitely, some of them can. The only questions not yet answered are: How many dead people could have voted in Hawaii’s primary and general elections in 2010? How many dead people actually did vote in those elections? To answer those questions would require in-depth work by an investigative reporter with lots of time and great persistence to break through bureaucratic stonewalling and buck-passing at several state and county government agencies.


Following is a brief report of the limited investigation I conducted as a low-level election official at just one precinct in Kaneohe in the primary and general elections of 2010 (He’eia Elementary School, Precinct 48-01).


My most shocking discovery was that at least one dead person remained listed in the pollbook on the primary election date of September 18, 2010 and therefore anyone claiming to be that person could have received a ballot and voted. The requirement to show identification when receiving a ballot is so vague and loosely enforced that any family member, neighbor, co-worker, friend, or even stranger could successfully get a ballot in the name of a registered voter merely by bringing a current utility bill with the voter’s name and address on it, or by reciting the voter’s name, address, and birthdate (month and day but not year). In this particular case, the dead person’s name was published in a newspaper obituary, and the name and home address were listed in the phone book; so it would be easy for anyone to obtain enough information to get this person’s ballot.

The dead person listed in the pollbook for the primary election was removed and no longer present for the general election. But it’s not clear whether the name was removed through routine bureaucratic procedures (which, of course, would be a good thing) or whether it was removed because a fellow precinct official might have informed the Office of Elections about my discovery (see below for the story about attempts by the bureaucracy to stifle my little investigation). I also don’t know whether any imposter actually voted on behalf of the dead person, because halfway through the day the bureaucracy deprived me of access to the pollbook (where the fact is recorded that someone has voted).


In the general election, there was another dead person listed in the pollbook (who was probably also in the pollbook for the primary election although at that time the person was still alive and therefore did not attract my attention). This name had an “AB” preprinted next to it, which meant that an absentee ballot for the general election (and probably also the primary) had been mailed to the voter. Another person of the opposite gender and same last name, at the same address (probably the spouse) was also listed with a preprinted AB. The AB is in the pollbook to warn precinct officials not to give the person a ballot on election day, because the person has already been sent a ballot in the mail.

In this case the date of death of the 80-year-old was October 3 as shown in an obituary published later. October 3 was long after applications for absentee ballots could have been sent by voters to the Office of Elections, and was a week before the date when OE mailed out the ballots for the general election. So there’s no indication of fraud (or absence of fraud!) in this case. But was OE notified of the death in time to stop OE from mailing the ballot, even after preprinting the AB in the pollbook? Or was the ballot mailed out? And if it was mailed, was it actually voted fraudulently after death (perhaps by a family member) and then mailed back to OE?

The dead person’s name and address are listed in the phone book, so I could have made a phone call to the surviving spouse to ask those questions. But I chose not to do so out of respect for the grieving family. Those questions could be answered by an investigative reporter, perhaps with a subpoena to force OE to disclose the information. There is a space at the bottom of the application for absentee ballot where OE writes down the ballot stub number, date, and name of the OE clerk both when OE mails out a ballot and when the voted ballot is received back by OE. Instructions and application for absentee ballot can be downloaded from OE here:


I positively identified only one dead person in one pollbook, who could have had an imposter walk in to the precinct, gotten a ballot, and voted. I also identified one dead person who could have gotten an absentee ballot after dying which could have been voted by an imposter. Perhaps it seems unworthy to give attention to one very bad case, and another case that might or might not be bad.

But there were probably quite a few more cases I did not discover at my precinct. There are many reasons why my small investigation was incomplete. Due to bureaucratic stonewalling (see below), I was not able to get a list of the dead people who had allegedly been removed from the voter list; so I was forced to use newspaper obituaries as my sole source of names. But there are dead people for whom no obituary is published. Also, I began keeping track of obituaries only about 6 weeks before the primary election; but there could have been people who died months or even years before that who were still on the list of voters. Also, obituaries are sometimes not published until several weeks after death, so there could be several people who died shortly before election day about whom I had no information — indeed, those are the dead people most likely to still be in the pollbooks because of bureaucratic slowness.

It’s quite possible that there could be dozens of dead voters whose names remain in the pollbooks on election day in every precinct. There are 242 precincts in the State of Hawaii for 2010; so if my two or ten bad cases at my single precinct are typical, that’s about 500 or 2500 potential fraudulent votes. Of course that’s not likely, but the possibility is alarming. Even if no dead people actually vote, the fact that even one dead person remained in the pollbook on election day displays the proverbial “tip of the iceberg” — there are serious glitches in the bureaucratic process of the Office of Elections which might allow actual voter fraud to happen. Mathematicians and scientists know that even a single piece of evidence that contradicts a formula or hypothesis can disclose a serious flaw.


Absentee voting by mail obviously has far more opportunity for fraud than in-person voting — ballots can be stolen, sold, or given away; election officials working in back rooms unseen by the public might “lose” votes they don’t like; etc.

Elderly people in nursing homes, or members of labor unions, might feel compelled to fill in their absentee ballots the way their caregivers or union stewards tell them. Just imagine a union boss, flanked by a couple of legbreakers, telling a hundred union members “I want all of you to fill out these applications for absentee ballot. We’ll mail them for you. And then we’re gonna bring all those ballots to this meeting hall three weeks from today to fill them out together.”

During the past few months there have been aggressive efforts to get voters to sign up for permanent absentee voting, which will make it even more likely for bad things to happen and harder to purge the pollbooks of voters who have died. With a single application a voter will automatically receive absentee ballots for the rest of his life — and perhaps beyond. So long as someone keeps voting those ballots and they are not returned by the Post Office as “address unknown,” the Office of Elections will send out ballots for the next election year.

My little research project proves that there is no system now in place which works correctly all the time to eliminate dead people from the pollbooks or absentee voting list. Tens of thousands of absentee ballots are already mailed for a single year to people who applied during that year for either or both the primary and general election, and some of them will die while the unmarked ballots are still in their possession. Tens of thousands of additional voters are signing up to permanently receive absentee ballots without applying again for each new election cycle, making it increasingly likely that dead people will receive ballots.


Voters are usually asked to show photo ID such as driver license or military or state ID card. That would make it difficult for anyone other than the actual voter to get a ballot in the voter’s name. But Hawaii does not really require photo ID.

The 2010 Precinct Official’s Manual, page 20, says that the following would be acceptable identification: “Copy of a current utility bill, … paycheck, or other government document that shows the name and address of the voter …”

The pollbook includes the month and day of birth for each voter. Pollbook officials are instructed that if a voter has no photo ID, no utility bill, etc., it’s OK to issue a ballot based on the voter telling his name, address, and birthdate. The year of birth is not needed, only the month and date. I have been told several times over the years by trainers and precinct chairpersons that the year of birth is not requested for reasons of chivalry, so a lady does not need to tell her age! They’ve got their story straight! But this “chivalry” makes it easy for anyone in the world to get a ballot in the name of anyone else, even if the imposter’s age is decades older or younger. And of course if the real voter is dead then the imposter knows the real voter will not be coming by later to discover the falsification. A conspiracy theorist might say that the rule allowing voters to prove their identity merely by showing a utility bill; and also the other rule requesting only month and day of birth but not year of birth; were intentionally written by legislators or Office of Elections bureaucrats at the behest of political hacks for the purpose of making it easy to commit voter fraud.

There have been news reports that the U.S. Department of Justice is vigorously forcing states to comply with federal laws requiring outreach to encourage voting, such as the law requiring voter registration to be offered at the same time as driver license application or car registration and the law requiring special facilities for voters with handicaps. But the news reports also say there is no DOJ effort to enforce laws against voting by illegal immigrants or fraudulent voting.

The Office of Elections in every state has a basic conflict between encouraging all citizens to register and vote, vs. policing the voting process to ensure that the only people who vote are registered, living voters. Hawaii’s extremely low voter turnout is an embarrassment. Thus the Office of Elections feels pressure to err on the side of letting people vote rather than stopping them from voting. The Office of Elections is likely to use its limited resources to encourage voting rather than to guard against voting by non-citizens or dead people. This policy could be dangerous if political parties, labor unions, or other interest groups decide to make Hawaii resemble Chicago as a place where voter fraud runs rampant.

Some Hawaii voters are already doubtful about the integrity of our voting system. While monitoring the voting machine in my precinct and helping people whose ballots were rejected for stray marks, more than one voter commented “We know already who is going to win” and, pointing at the big black machine, “So I guess it doesn’t matter whether I’m able to get my ballot into that shredder.”


I choose to protect the privacy of the two dead people I discovered in the pollbooks, and their families, by not publishing their names, addresses, or genders. But I will gladly give all my information to officials of the Office of Elections or to any bona fide investigative reporter. I am hopeful (but doubtful) that someone will follow up my small, localized investigation with a big statewide one.


The flagrantly bad case in the primary election had a date of death of August 24 as published in a Honolulu Star-Advertiser obituary later (but still before the election). The person’s name was in the pollbook used in the primary election on September 18. Therefore the Office of Elections had three weeks to delete the name from the pollbook after the date of death, but failed to do so. And there was no inclusion of the name in the last-minute list of corrections. The Bureau of Conveyances keeps track of real estate title transfers and liens right up to any current date so that sales can be closed without doubt about ownership or encumbrance. The Office of Elections should be able to handle the deletion of dead voters with comparable efficiency.

“Corrections Order #1” is usually the only list of corrections, and I don’t know how close to election day it gets compiled and printed. In my precinct for both the primary and general elections, the corrections list contained several hundred names in alphabetical order. Pollbook officials are ordered to write the corrections by hand into the pollbook by each person’s name. That process usually requires one or two precinct officials for each pollbook to spend 45-60 minutes before the polls open to get all the corrections into the pollbooks. In my precinct there were six pollbooks. Each election there are larger numbers of early or absentee voters; thus, the time spent on this chore is growing.

The corrections are usually things like “AB” (voted absentee or early in-person and therefore not eligible to be given a ballot) or “QA” (questionable address therefore send the person to the voter assistance desk) or “Add voter” (someone registered to vote after the pollbooks were printed?) or “Delete voter.” Presumably one reason for “delete voter” would be that the Office of Elections was notified that a voter had died; but in my precinct none of the names to be deleted had been published in newspaper obituaries. Indeed, the two names of dead people I discovered in the pollbooks were not included in any corrections order. In the first example the corrections order should have said “delete voter”, and in the second case the corrections order should have said “cancel AB and delete voter” (although the AB alone would ensure that no ballot would be given to any walk-in imposter).

The fact that I found the name of one dead person still in the pollbook and not listed as “delete voter” or “AB” in the Corrections Order is alarming evidence there could be more. For example, I compiled a list of recently dead people only by looking in the newspaper obituaries for just a few weeks before the election for people described as “of Kaneohe.” But there are many dead people for whom no obituary is published, and many more people who died before I started keeping track My method was also inefficient because there are multiple precincts serving Kaneohe, so that most of the names I searched for in the pollbooks were probably not residents of my precinct.


Here’s how I tried to get a complete list of residents in my precinct who had died within a few weeks before the election — how the bureaucracy stonewalled me, and how an investigative reporter might be able to get the job done.

I hoped to get a list of names and addresses of recently deceased residents of my zipcode, and then compare their street names against the map of my precinct which is available on the office of Elections website. Then I would know which names and addresses to look for in the pollbooks in my precinct polling place.

First I called the Office of Elections the day after my July 6 training program to ask them exactly what they do to make sure there are no dead people still listed in the pollbooks on election day. The OE said they get a list of dead people from the County Clerk’s Office. But OE refused to give me the list, saying it is confidential; and OE also did not know or would not tell how often they get such a list and how close to the election they get a final list. I called the County Clerk’s office and got the same reply; but the CC did tell me that they get the list of dead people from the Department of Health, because all people who die are required to have a death certificate filed with DH. I called DH and got the same reply: “The list is confidential.” Presumably the list is in the public domain and should be available for research; but the bureaucrats don’t want to cooperate and seemed not to like being scrutinized. It occurred to me that the U.S. Social Security and/or Medicare bureaucracy might have a frequently updated list of dead people that could be sorted by zipcode; but I decided their bureaucracy would be even more impenetrable and uncooperative than the state and county governments’.

On primary election day at least one of my co-workers at the precinct polling place did not like what I was doing, and apparently used a cell phone to complain to the Elections Office. OE responded by ordering my precinct chairperson to order me to stop looking through the pollbooks to look for dead voters. Of course elections officials, including myself, look in pollbooks all day long, beginning with Corrections Order #1 at about 5:45 AM and continuing until 6:00 PM closing time, whenever a voter comes to get a ballot. I had made a point of doing my research only when I was on a break and when there were no voters in line at whichever pollbook I wanted to examine. But the chairperson told me that although I was authorized to look in pollbooks when making corrections or handing out ballots, I was not authorized to look in them for purposes of research. Thus, having discovered a dead person’s name in one of the pollbooks around mid-morning, I was not able to look again at the end of the day to see whether that name had actually been issued a ballot.

On general election day my precinct chairperson made an announcement at about 6:30 AM, before the poll opened, regarding the fact that a pollwatcher had been authorized for our precinct and he had the right to look in the pollbooks and take notes; but that otherwise only those workers assigned to the pollbooks are allowed to look in them and only for the purpose of voter identification and ballot issuance. He then assigned me to sit at the voting machine to protect its security and to assist voters who had difficulty using the machine; and that was my only assignment for the entire day. Thus I was kept away from the pollbooks and prevented from doing any more (prohibited!) research. But the list of names and addresses in the pollbooks is also posted on a wall outside the school cafeteria polling place, for the public to see whether they are at the right place. Thus I was able to check my list of recently dead people in Kaneohe during a “bathroom break”, and discovered the one name. Someone who knew of my research was able to look in the pollbook at the one dead person’s name I had found, and informed me that name had a preprinted AB next to it.

Despite all attempts by the Office of Elections and some of my fellow precinct officials to stop me, I was nevertheless able to do the research at least partially (but only for my one precinct). Now we’ll see whether anyone from the Office of Elections, seeing this essay; or any investigative reporter; asks for the detailed information I am ready to provide about my two cases. In the flagrant primary election case there needs to be an explanation of what procedural errors allowed that particular dead person to be listed in the pollbook (enabling any imposter to get a ballot), and whether the pollbook shows that a ballot was actually issued (to an imposter), and what changes are being made to ensure such a thing cannot happen again. In the general election case there needs to be disclosure whether the dead person was actually mailed an absentee ballot, and whether that ballot was then marked with votes and returned to the Office of Elections; and if so, what actions are being taken to prosecute someone for voter fraud in this case and to prevent other similar cases in the future. And of course a thorough investigation is needed that covers all the precincts and pollbooks for both the primary and general elections. Is anyone interested?





  1. Congratulations to Ken Conklin for establishing that at least one dead person was still listed as eligible to vote in Ken’s precinct for the primary election this year; and also at least one in the general election. Now the election office should look into it and determine whether votes were cast in those two instances; and, if so, explain why that was not prevented. The election office should then investigate and determine in Ken’s precinct how many votes were cast in the name of dead registered voters in each the primary and general elections. The result of that investigation should be announced and the election office should also announce what steps, if any, it recommends to fix the problem.

Comments are closed.