Rick posted yesterday on an Arkansas registrar’s decision to cancel the registration of Leslie Rutledge, a candidate for state AG, after it was brought to the registrar’s attention that Ms. Rutledge had been registered in another state. One of the issues, apparently, is whether the cancellation came too late in the cycle.
There are two different provisions of the NVRA at issue.
One is section 8(c)(2) (which is apparently now 52 USC 20507, in the new and still unfamiliar Title 52). That’s the 90-day “pencils down” provision. It says that “a State shall complete, not later than 90 days prior to the date of a primary or general election for Federal office, any program the purpose of which is to systematically remove the names of ineligible voters from the official lists of eligible voters.” Unless the AG candidate (Leslie Rutledge) was caught up in a sweep under a “program” with the purpose to systematically remove similar people, the 90-day provision doesn’t appear to apply. This is the part that Rick was mentioning.
BUT. The other applicable provision is section 8(d), which says that a State may not cancel a voter’s registration on the grounds that the person has changed residence — any cancelled registration, not just a purge as part of a “program” — unless the registrant 1) confirms in writing that the registrant has changed residence or 2) has been sent a notice, hasn’t responded, and hasn’t voted in the jurisdiction for two cycles after the notice was sent.
I gather that the clerk’s official reason for the purge is that Ms. Rutledge has changed her residence (apparently, because she registered in DC in 2008 and Virginia in 2010), and that her registration in another state served as “written confirmation” of a change of residence. That’s an interpretation of the Act that I think is seriously suboptimal. But it’s one that I understand a fair number of registrars are using when they cancel registrations after receiving results of an interstate matching protocol, and finding what is ostensibly the same person listed as more recently registered in another state. More careful registrars (and there are many) do a lot more investigation when they get such results, just in case it’s the other registration that’s invalid.
Or maybe they’re both valid. Perhaps Ms. Rutledge was registered in Arkansas, moved to DC and became a voter there, and then re-established residence in Arkansas. (There seems to be a 2008 absentee ballot cast from DC, in Arkansas, that raises other serious questions.) The clerk apparently issued a registration card to Ms. Rutledge in 2013. Which means that whatever the validity of Ms. Rutledge’s status as an eligible Arkansas voter in the interim, it’s not clear how the (old) DC registration could serve as valid written confirmation that Ms. Rutledge has (now) changed residence out of Arkansas, particularly when she’s got more recent written confirmation that she lives in Arkansas.
More generally, I think it’s worth noting that an outside group apparently checking for duplicate registrations — the “Arkansas Libertarian Coalition,” if indeed they exist — seems to have raised the issue. Whether they had the most civic motives in mind or not, there’s also a substantial risk that groups that are data-mining for challenges to registration status or residency can misinterpret the information they get — and when the challenges come at the last minute, there can be insufficient time to clear things up before the election itself. That’s part of why the 90-day provision exists. But registrars ought to have a healthy skepticism about this sort of information — or, at least, double-check the ramifications — before acting on it, even when they’re not acting as part of a “systematic program.”