Last week, Arkansas Secretary of State John Thurston made the shocking decision to unilaterally disqualify a ballot initiative measure that would have reversed the state’s near-total ban on abortion and prevented the Legislature from enacting any further laws denying abortion access. The group had collected more than 100,000 signatures by a July 5 deadline, more than meeting the necessary threshold, but Thurston informed the group on July 10 that he was preemptively rejecting the proposal before the signature count even began, due to what turned out to be one piece of missing paperwork at the time of submission. (See updates below!)
Earlier this week, Arkansans for Limited Government, the group behind the Arkansas Abortion Amendment, sued Thurston, arguing that the group had in fact complied with all requirements. Even if the paperwork technicality caused some signatures to be disqualified, they argued, Thurston was required by law to continue counting and verifying signatures, potentially leading to an opportunity for the group to collect more signatures to make up any shortfall. On Wednesday, the Arkansas Supreme Court ordered the state to file its response to the lawsuit by today (Friday) at 4:00 p.m.
Attorney General Tim Griffin, representing Thurston’s office in the lawsuit, submitted that response a short time ago. You can read his response below. It will now be up to the court to decide whether the count must continue — giving the petition new new life, at least for now — or whether this is the end of the line for the petitioners.
Griffin argues that Thurston correctly carried out his duties when he unilaterally rejected the measure. His arguments are quite different from what Thurston described in his original letter to the abortion petitioners — and arguably at least imply that the reason Thurston gave then would violate the law. Griffin says that the missing paperwork was unusually crucial to turning in a complete submission in the first place, so the group simply did not successfully turn in their petition on time at all. He also says that the missing paperwork was incomplete and turned in at the wrong time, and could no longer be turned in after the deadline — and in event, he argues, it was signed by the wrong person.
However, Griffin’s claims do not appear to hold up to scrutiny or comply with the statute. In an update coming soon, we’ll argue that the attorney general engages in highly twisted logic, stringing together snippets of text across multiple statutes to arrive at his desired conclusion in a way that seems to undercut the purpose, structure and procedures of the plain language in the law. This acrobatic reading leaves multiple sections of the law clearly gratuitous, and possibly even contradictory — sections passed under the very same law Griffin references. It seemingly cannot be reconciled with the chronology of the petition review process described in the law. The Court may prefer to side with Griffin and stop the count for ideological reasons, but it’s difficult to see how they can justify his claims under the law.
For more background on what’s transpired over the last couple whirlwind weeks to lead us here, see previous posts here and here.
This post has been updated below with opinion and analysis on Griffin’s filing: What his claims are and how they fall short. If you just want the gist and the upshot, the paragraphs above should suffice. If you’d like to go a little deeper into both the legal analysis and the democratic principles at stake in this dispute, come take a dive with us. You can think of this as a throwback to Matt’s old blog, The Blue Hog Report — we’re going deep.
What Griffin is arguing
Under Arkansas Code Annotated 7-9-111(f), sponsors of a ballot initiative are required to submit a statement “signed by the sponsor” saying that their paid canvassers had received a copy of the most recent initiative handbook published by the secretary of state’s office and had been trained on the rules and regulations of signature gathering prior to going out in the field. On July 5, Arkansans for Limited Government submitted more than 100,000 signatures, but they did not submit the signed statement required under 111(f). They did include signed affidavits from all paid canvassers, affirming that they’d received the handbooks and the required training. There is no dispute that the group actually followed the handbook and training requirements. The group had also previously submitted the relevant signed statement on June 27 (and at least 16 times previously).
In his response to the group’s lawsuit against Thurston, Griffin makes a number of related arguments about the requirements of the initiative process. He contends that the failure to submit the signed statement required under Ark. Code Ann. 7-9-111(f) at the time they submitted the entire ballot initiative petition on the July 5 deadline means the group essentially failed to turn in a petition at all. Thurston “did not determine AFLG’s petition failed for a lack of sufficiency, but that it failed for want of initiation,’” Griffin argues.
According to Griffin, to submit a “completed petition,” a sponsor must submit at least the number of signatures required, an affidavit stating the number of petitions and total number of signatures being filed, and the signed statement required under 111(f) — and all of this must be submitted simultaneously as part of the final submission package by the July 5 deadline. Griffin contends the signed statement under 111(f) is “mandatory” and that Thurston did not have the duty to count and review the signatures submitted if a group does not submit all mandatory documentation along with the petition. In other words, the submission was not complete, he says — they didn’t make the deadline. Under Griffin’s argument, not turning in the 111(f) statement was equivalent to not turning in the required number of signatures at all.
Griffin further contends the group’s June 27 statement was not sufficient under 111(f) for several reasons. Because they added additional paid canvassers after that date, he says, the group failed to comply with the statute’s requirement that the statement cover “each paid canvasser.” Griffin also contends that the wrong person signed the June 27 statement. Allison Clark, the signatory, works for a company that contracted with the group to hire paid canvassers. Clark led the abortion petitioners’ effort to organize these paid canvassers, including compliance with all requirements for those canvassers. Because Clark was herself a paid canvasser, Griffin contends, she cannot also be a “sponsor” within the meaning of the statute. Nor could the group fix this lack of a signed statement with a document they submitted on July 11, Griffin says, because the deadline — which he says is a “constitutional deadline” — had passed at that point. Moreover, the needed document had to be filed “contemporaneously with the filing of the petition,” — in addition to missing the deadline, he says, the group failed to turn in the document as part of the overall final package it turned in.
Based on all of this, Griffin concludes, Arkansas for Limited Government’s violation of 111(f) “required the Secretary to reject its submission entirely.”
Unlike Thurston’s letters to the group, Griffin’s motion to dismiss also addresses the requirements of Ark. Code Ann. 7-9-126 and when the statute mandates that signatures not be counted “for any purpose.” Unfortunately for Griffin, that statute undermines the bigger point he is trying to make in the motion.
Why he appears to be wrong
We’re not going to spend much time on Griffin’s underlying claim that the group violated the 111(f) statute, which we’ve covered in detail before (see here and here, for example). The abortion petitioners claim there was no violation and all of their paid signatures are valid — that’s possible, but a challenging claim. They have a stronger argument that some of the paid signatures — those collected on June 27 or earlier — should stay even if the later ones are tossed. But there’s at least plausible cases to make on either side of that question. Finally, Griffin’s claim that the particular individual who signed the June 27 document doesn’t count as a “sponsor” isn’t crystal clear from the statute but we’ll see what the Court says.
But the problem for the secretary of state’s office is that even if the Court concludes there was a 111(f) violation, the law still seems to insist that the count continue, potentially granting the group a chance to secure a “cure” period: 30 days that allow the group to correct errors and collect more signatures to make up the shortfall. To justify Thurston’s decision to preemptively reject the entire petition without beginning the count, Griffin’s arguments rely on a reading of the relevant statutes that appears to be logically strained at best and flat wrong at worst.
Ark. Code Ann. 7-9-126 was created in 2013 by Act 1413, which overhauled several aspects of the initiative process. That section, entitled “Count of signatures,” lists eight specific reasons under 126(b) that entire pages of signatures may be rejected and lists another seven reasons in 126(c) that individual signatures can be culled. Under 126(b)(8), one basis for rejecting entire petition parts is where the part “has a material defect that, on its face, renders the petition part invalid.”
According to Griffin, that “material defect” includes a failure to submit the signed statement required under Ark. Code Ann. 7-9-111(f)(2). Thus, Griffin says, a violation of 111(f)(2) requires rejection under 126(b)(8). But this reading confuses facial validity, which is addressed first under 126, with sufficiency, which comes second under 111.
According to the process established by Act 1413 in 2013, there is a two-step analysis when a petition is submitted. The first deals only with facial validity and is governed entirely by section 126. That initial analysis works like this:
- Under 126(a), when the secretary of state gets a petition, he begins an initial “count” and determines “whether the petition contains, on its face and before verification of the signatures of registered voters” at least the total number of signatures needed to qualify for the ballot. That gross total (canvassing professionals sometimes call this the “raw” count) is the starting point. In practice, the secretary of state doesn’t necessarily have to start counting at this stage — at this point, the figure can be determined simply by the total number of signatures attested to by the sponsor when submitting. For example, the petition to eliminate the sales tax on feminine hygiene products and diapers only submitted 43,831 signatures, so their petition was rejected immediately without any further review.
- Next, 126(b) gives eight specific scenarios that would mandate that the secretary of state disqualify the entire “petition part” (a page with up to ten signatures). In such cases, the secretary doesn’t count the signatures on that page “for any purpose.”
- Third, 126(c) directs the secretary to look at individual signatures and see if any of them are disqualified for any of the specific seven listed reasons. These scenarios involve errors such as missing birthdates or addresses.
- Finally, there is 126(d), which says, essentially, after culling all the signatures disqualified under 126(b) and (c), count how many are left —- that’s the net total for analysis under this section (Or you can think of this as basically subtracting the (b) and (c) infractions from the total we started with —- (a).
- Per 126(d), if that net total is less than the minimum number of signatures required to place the initiative on the ballot, that is the end. There is no cure period, and the petition is rejected.
The process isn’t over yet — there will be another count, with different types of infractions that lead to culling signatures — to see if the final remainder meets a 75% threshold to earn the crucial cure period (again, that’s 30 days that allow the group to correct errors and collect more signatures to make up the shortfall).
Where Griffin’s argument falls apart is in his attempt to shoehorn requirements from 111(f) into the analysis under 126.
Ark. Code Ann. 7-9-111, entitled “Determination of sufficiency of petition,” deals entirely with determining whether the remaining signatures — the ones that were not culled in the 126 analysis — are sufficient to warrant a cure period.
In other words, Thurston could only get to 111(f) after he went through the steps in 126. Griffin seems to be inventing a new chronology not supported by the statute, and oddly imagines a scenario in which a 111(f) issue stops the process from starting even as it seems to also be a violation that would impact the end of the process. This can be demonstrated by the history of 126 as well.
According to the uncodified language in the statute, Act 1413 was necessary in 2013 because, a year earlier, sponsors of four separate initiative petitions submitted petitions containing over 298,000 signatures, of which one petition had an initial signature validity rate of 56%, while the other three were each below 31%. The three with the lowest validity rates faced several allegations of fraud and forgery in the signature gathering process. While none of those measures ultimately made it on the November 2012 ballot, all four were given a cure period to try to collect additional signatures under 111 because the law at the time did not allow the secretary of state to disqualify signatures on the front end and not count them when determining whether a cure period was allowed. (It was not until the adoption of Amendment 93 in 2014 that initiative petitions had to have at least 75% of the required total of valid signatures to get a cure period.)
Act 1413 sought to change this. Importantly, however, it did not add language to 111 about culling signatures before determining whether a cure period would be granted. Rather, it created 126 specifically for that purpose and clearly laid out both the process for determining which signatures could be rejected out of hand and what criteria would allow such a disqualification. In 126(b)(2), (b)(4)(A), and (b)(7), the Legislature explicitly referenced violations of other statutes that would require disqualification of certain petition parts. In 126(c)(6) they specified a statute whose violation required disqualification of specific signatures.
Nowhere in 126(b) or (c) did Act 1413 reference section 111(f) as a statute whose violation required — or even allowed — for rejection of all or part of the petition. There is also no mention of 126(a) — the preliminary phase under which Griffin chronologically implies the 111(f) review takes place.
There is a long-standing rule of statutory construction that says the express mention of one thing in a statute means other similar things that weren’t mentioned are excluded. For instance, if a statute said it was illegal to tint the windows of vehicles over 25,000 lbs., that would mean it was explicitly not illegal under that statute to tint the windows of a vehicle that weighed less than 25,000 lbs. Make sense?
Here, because the Legislature specified Ark. Code Ann. 7-9-104, 7-9-107, and 7-9-601 as statutes whose violation would result in rejection of petition parts or signatures under 7-9-126, then by definition a violation of 7-9-111 cannot be used under 126 to reject the entire petition at that stage.
Griffin tries to get around this by pointing to the language in 126(b)(8) about “any material defect.” If that clause is given a reading as broad as Griffin suggests, however, it would render multiple other subparts of 126 redundant or superfluous. Indeed, it seems you wouldn’t need to list any of the other types of infractions that count as facial violations (at least 17 specific violations are listed, apparently gratuitously). Griffin seems to be implying that the category of “facial” violations is so wide open that despite all the specifications in 126, anything that doesn’t require looking something up to verify it would count. That would certainly be a convenient tool for state officials hoping to disqualify more petitions prior to the phase of review that allows for a cure.
The Supreme Court has repeatedly held that statutes are construed “so that no word is left void, superfluous or insignificant.” Yet, if “material defect” means “violated some other statute,” which is what Griffin is urging, there would have been no reason for the Legislature to specify certain statutes in other parts of 126. All statutory violations would fall under 126(b)(8).
Griffin’s argument also fails to consider section 111 as a whole. Under that section, once a petition has reached the next stage after the facial review described above, signatures are checked for sufficiency (such as checking to verify that each signatory is a registered voter), sponsors are entitled to a cure period as long as they have at least 75% of the total valid signatures required. There is literally nothing in 111 that says a failure to comply with 111(f) precludes a cure period or allows disqualification of signatures. There is nothing that says 111(f) must be satisfied before the secretary has to review the sufficiency of the signatures under that section. And, perhaps most damning to Griffin’s argument, there is nothing that says 111(f) acts as a threshold that must be met before the secretary will even look at the petition under 126. It looks like just another violation that would be dealt with during the sufficiency review that allows for a cure if the count reaches 75%.
In short, the statutory scheme post-2013 is: (1) review the facial validity of petition under 126; (2) review the sufficiency of the petition, including the signatures, under 111. Griffin’s argument would blow up that two-step process by combining 111(f) into the section 126 analysis despite nothing in the statue to support such a tortured reading. Why is 111(f) given special qualities that the rest of 111 doesn’t have? And why would it supercede 126, written specifically to note special cases of facial violations? It is hard to parse.
And it gets weirder still. Griffin wants to treat a 111(f) violation — omitting one document — as functionally equivalent to canvassers turning in too few signatures, or none at all (in other words, a 126(a) violation). That’s what he means when he says Thurston didn’t disqualify the petition “for a lack of sufficiency, but that it failed for want of initiation.” This directly contradicts Thurston’s letter, by the way, but Griffin is saying that Thurston never disqualified the paid canvasser ballots at all (that was not a part of his official process whatsoever, he simply offered that information as part of the most confusing “courtesy” one can imagine to the abortion petitioners, according to Griffin). Instead, Griffin says, he rejected the entire petition outright before any specific disqualifications of signatures or signature sheets was even pondered. But wait. If Thurston tossed the whole thing during the 126(a) phase, how could it be that the problem was a 126(b) violation, which doesn’t come until next in the chronological process, and is clearly meant to be a validity problem different in kind than 126(a)?
Griffin seems to want 111(f) to be in a kind of superposition — enforced in the phase described in 126(a) when it blows up the process before it starts, under a category relevant to the next step in 126(b), all while it definitionally remains a sufficiency violation that comes at the very end of the process (the cite here, perhaps tellingly, is…111). We are left to wonder if Griffin’s desire to keep the abortion amendment off the ballot pushed him to such gymnastics.
Finally, the cases Griffin cites in his motion are…interesting.
To support his argument that Thurston could reject the petition “for want of initiation,” he cites to a 1946 case that dealt only with petitions that lacked the bare minimum number of signatures required and says nothing that agrees with Griffin’s contention that 111(f) is a threshold issue and failure to comply with it means the group failed to turn in a complete petition.
He also cites Benca v. Martin as part of his argument about 111(f), but that dealt specifically with paid canvasser hiring and training under 601 and had absolutely nothing to do with 111(f) beyond a passing mention. Nothing in Benca says or suggests that a violation of 111(f) requires rejection of the entire petition. Indeed, the court in the Benca case went through and analyzed the various categories of signatures being challenged under 601 and then performed a count, rather than dismissing on the front end.
Indeed, it is unclear why Griffin thinks Benca helps his position. In that case, the secretary had already counted the signatures under section 111. Logically, this means it had already completed the analysis under 126 as well. Once the 126 analysis was complete, the sufficiency review under 111 was triggered. The plaintiff in Benca then challenged the sufficiency of some of those signatures under Ark. Code Ann. 7-9-601, which is not at issue in this case.
Okay, let’s wander back in from the weeds and think about the basic principles at play here. It’s unclear whether Griffin has thought through the due process issues with the protocols he claims to be conjuring from the statute. According to Griffing, all of the following is kosher and above board: A group turns in more than 100,000 signatures collected in a massive undertaking, along with extensive paperwork. The secretary of state thanks the group and tells them that their submission has been received. Then a week later, it sends a letter and says: Never mind! And the letter describes reasons that they failed to make the deadline because certain signatures were disqualified because of a missing document in the group’s final submission. Only those aren’t the real reasons at all. That’s just, according to Griffin, an unofficial act in which the secretary of state was trying to make the petitioners feel better by telling them that they would have failed later in the process anyway. Then, over the course of the next few days, the state officials refine their reasons a couple more times. At every step of the way, the group is told it’s too late to fix the paperwork problem it wasn’t informed of until it was, well, too late.
This seems obviously offensive to basic fairness. While this issue has not been raised by the abortion petitioners explicitly, in some instances, courts find that if bureaucrats are using flimsy rules infractions to interfere with a good faith effort to comply with the law, that represents “substantial compliance” and can’t be undercut by meaningless nit-picking. In this case, the purpose of the statue is to prevent fraud — indeed, that is one of only two allowable purposes for legislation impacting ballot initiatives under the Arkansas Constitution. Griffin has no allegation that the group did correctly implement the mandates to offer training and handbooks to all paid canvassers before they began collecting signatures. He’s just saying they were late on their paperwork.
Imagine you turn in a massive term paper to a professor. You are missing the cover page. Should the professor give you a zero? That is the scheme envisioned by Griffin that appears at least in his telling to be an invention with a strained relationship to the statutes. It would be one thing if the professor told all students that a missing cover page meant an automatic zero. But that is not what happened in this case. The 111(f) statute instead describes a violation with a clear penalty under the very same statute — removal from the count during the sufficiency review, with the opportunity to cure. To return to our analogy, you might say the professor told everyone he would dock 5 points for a missing cover page. Fair enough. But that doesn’t mean the student didn’t turn the paper at all. It doesn’t mean a zero. It means it’s time to start fulfilling the professor’s obligation to grade the paper.
If a plain reading of the law is any guide, Thurston must start the count.