The GAB’s Duty of “Careful Examination”: Why Judge Davis Got It Right

I have written a few things on my personal blog about the GAB’s authority and duty to conduct a more thorough review of recall petitions than it apparently intends to conduct. Last Thursday, Judge Mac Davis ordered a more extensive review. Ed Fallone thinks that the judge got it wrong. I disagree. Here’s why.

Ed argues that “there are no explicit provisions in the statutes that direct the GAB to look for and eliminate duplicate, fictitious or unrecognizable signatures. Just a direction not to count signatures that are insufficient under Section 9.10(2)(e).”

I see two problems with this statement. First, the GAB’s obligation upon the filing of a petition is not limited to the elimination of signatures for the reasons set forth in § 9.10(2)(e). To the contrary, the obligation imposed on GAB is to “determine by careful examination whether the petition on its face is sufficient.” Wis. Stat. § 9.10(3)(b). Whatever that duty is, it is nowhere limited by § 9.10(2)(e). Second, as we will see, even if it is so limited, § 9.10(2)(e) does not relieve GAB of the obligation to do what Judge Davis ordered it to do.

So what does this duty of “careful examination” entail?

I don’t think it helps us much to note that the statute provides for a challenge procedure in which the burden of proof is on the challenger. GAB’s duty of careful examination is independent of the filing of any challenge. It has to do something. As Judge Davis noted, while the GAB does perform a quasi-judicial function, it also has a separate duty to administer and enforce the election laws.

We know two things. First, the examination requirement is limited to the face of the petition. This might be read to mean that the GAB need not resort to extrinsic sources or examine the circumstances under which any signature was obtained. My understanding (there is no written order yet) is that Judge Davis held that the GAB should consult extrinsic sources such as web searches, DMV lists, and voter rolls in the context of fictitious names. I’m not sure I’d go that far, although one of my colleagues at WILL makes a pretty good argument that testing the facial validity of the petition does require reasonable efforts to verify that, for example, an address actually exists.

Second, the statute specifies that the purpose of this examination is to determine whether the petition is, on its face, “sufficient.” A petition that is sufficient to initiate a recall is one which has been signed by the requisite number of qualified electors within the prescribed time frame. The GAB’s review must determine whether, on the face of the petition, that burden has been met

Why doesn’t this require the GAB to eliminate duplicate signatures? If careful examination requires determining whether the petitioners have met their number, one can’t count the same elector more than once. A petition that meets its number by counting the same elector multiple times is not facially adequate.

It also seems obvious that one must—at least—eliminate signatures where the address is illegible. Such signatures cannot be determined to be those of an elector residing in Wisconsin. They are facially invalid.

I would argue that the GAB must also eliminate names where the signature is illegible. There is no way to conclude that a qualified elector has signed if you can’t discern his or her name. It is not clear to me, however, whether Judge Davis so ordered and it seems a closer question. The law does not generally require signatures to be legible (or even to be signatures), but perhaps where, as here, the signature must serve as evidence that an eligible elector has signed, that interpretive gloss is inapplicable.

Fictitious names are a more difficult matter. The fictitious nature of the name must be apparent on the face of the petition, and it is, of course, possible that there are people in Wisconsin with odd names or even the same name as cartoon characters or genocidal dictators. It’s possible.

But when, in the course of Thursday’s ruling, Judge Davis said that “[c]ounting the signature of Bugs Bunny is something only lawyers could try to make seem OK,” I think he was calling for a common sense approach. There may well be signatures that are fictitious on their face (and, of course, their exclusion — as well as the inclusion of borderline questionable names — may be challenged.)

As I said, nothing in the statute limits this review to the criteria set forth in § 9.10(2)(e) nor does such a limitation follow from the structure of the statute. Given that § 9.10(2)(e)) includes things that cannot be determined by a facial review, see, e.g., §§ 9.10(2)(e)6-8, it seems a stretch to read them as constituting a definition of or limitation upon the “careful examination” requirement.

More fundamentally, limiting the GAB’s “careful examination” requirement to the § 9.10(2)(e) criteria doesn’t relieve the GAB from doing any of the things that Judge Davis has ordered it to do. For example, § 9.10(2)(e)4 provides that a signature may not be counted if “[t]he residency of the signer of the petition sheet cannot be determined by the address given.” If one cannot read the address, then one cannot determine residency.

Sec. 9.10(2)(e)5 provides that a signature may be rejected if it is “of an individual who is not a resident of the jurisdiction or district from which the elective official being recalled is elected.” If a name appears to be obviously fictitious, then this requirement has not been met.

Once again, I think a reasonable argument can be made that § 9.10(2)(e) requires illegible names to be stricken as well. There is no way to determine that the criteria set forth in § 9.10(2)(e) have been met — on challenge or otherwise — if the name of a signatory cannot be read. (Again, it is not clear to me that Judge Davis went that far.)

Any reading of the statute that places a burden on the challenger to prove that someone whose name or address can’t be read is not anyone who lives anywhere in Wisconsin is, to put it mildly, highly problematic and fundamentally at odds with the statutory requirement that a requisite number of qualified electors sign within a prescribed time frame.

While the § 9.10(2)(e) criteria do not include duplicative signatures, no one disputes that there is an obligation to determine whether the petitioners have hit their number, and no one believes that this number must not consist of distinct individual electors, i.e., in meeting the number you can’t count the same elector more than once.

The best argument that I can see against not counting duplicates is that the statute authorizes a challenger to request the elimination of duplicative signatures . § 9.10(2)(i). But the grant of such authority to the challenger does not mean that the GAB may ignore facially duplicative signatures. In fact, given the clear obligation to determine whether the requisite number of signatures has been obtained at the “careful examination” stage, it makes little sense to think so.

I am not sure, in any event, that there is a clear divide between the bases for “not counting” a signature under § 9.10(2)(e) and for striking a signature under other parts of § 9.10(2). For example, both permit “not counting” or “striking” ineligible electors. Compare, e.g., § 9.10(2)(e)5-8 with § 9.10(2)(L) and (q). This suggests that they are not mutually exclusive.

But if one must have such a division, a more supportable distinction is that the former addresses facial review and the latter are directed toward challenges based on extrinsic evidence. See § 9.10(2)(h)(“ Any challenge to the validity of signatures on the petition shall be presented by affidavit or other supporting evidence demonstrating a failure to comply with statutory requirements.” (emphasis supplied)).

In past recalls, these questions haven’t been as salient because the number of required signatures has not been so large as to potentially overwhelm the GAB and recall opponents. This time may be different. The GAB’s reluctance to do more seems not so much to be based on an absence of statutory authority (in fact, I don’t think they argue that they can’t do what Judge Davis has ordered), but a belief that the agency lacks the resources to do it. But that doesn’t justify ignoring what seems to me a relatively clear statutory command to do more.

Even if one could establish a history of passive review by the GAB, past practice should not override statutory obligation. This does not, as Ed suggests, change the rules in the middle of the game or disturb the settled expectations of recall proponents. Certainly, they did not believe that they could recall the Governor on the basis of duplicate, illegible, and fictitious signatures that recall opponents would not have the time or capacity to detect.

This Post Has 11 Comments

  1. Ed Fallone


    This is a very clever little argument that you set forth. As you might expect, I remain unconvinced. First of all, none of what you say in any way undermines the straightforward interpretation of the statutory language contained in my original post. You simply posit a rather strained alternative.

    Further, I am heartened that you do not rely on the same argument that Judge Davis relied upon in order to impose these affirmative duties on the GAB. The transcript of the oral ruling by Judge Davis is now publically available as an Appendix here. It appears that Judge Davis ruled that Section 9.10(2)(e), which lists the grounds for refusing to count a signature, should be read as if that section also incorporated the remainder of the statute – starting with Section 9.10(2)(f) and following all the way through Section 9.10(2)(r). By this reading, the later provisions creating a right to challenge a signature also (silently) deny the GAB the authority to count any signature that might be subject to an unexercised challenge. If this is in fact the basis for Judge Davis’ interpretation – and it is possible that his written order may clarify his reasoning – then his interpretation of Section 9.10(2)(e) seems to be in direct conflict with the canon of expressio unius est exclusio alterius (items not on a list are assumed not to be covered by a statute).

    Second of all, if you are correct that we should read the staute so that there is no distinction between a signature that is not “sufficient” under Section 9.10(3)(b) and a signature that is subject to a challenge by the officer subject to recall, then it would follow that under Section 9.10(2)(r) those persons circulating the recall petitions have the right to file affidavits correcting any signatures that the GAB purges as a result of Judge Davis’ ruling. In other words, if the GAB determines that a signature is duplicative, fictitious or unrecognizable (not illegible) at any time before the challenge stage, then the GAB must have in place procedures that allow the petitioners to exercise their statutory rights under Section 9.10(2)(r) to correct these “insufficiencies.”

    If the GAB adopts such procedures (as it clearly would have to under your reading of the statute), then it becomes obvious that your interpretation actually inverts the statute’s very clear command that the burden of challenging a signature shall be on the challenger — not placed on the petitioner to demonstrate the signature’s validity.

    Perhaps this is why neither the nonpartisan agency charged with interpreting these provisions of law, nor any experienced practitioner in this area, have ever read the statute in the manner that you suggest.

    Moreover, by contravening the principle of judicial nonintervention in ongoing elections (a principle that Professor Edward Foley argues is the key to understanding Supreme Court precedent in the Equal Protection area), your interpretation invites future judicial challenges to GAB recall procedures using a vague “reasonableness” standard. One judge’s opinion as to what is or isn’t a “careful examination” is a very slim reed on which to delay or frustrate a recall election (the petitioning of which is guaranteed as a right by the Wisconsin Constitution).

    I might add that your offhand suggestion that “illegible” signatures not be counted would be a particularly pernicious restriction on the recall rights of electors (and one that has no basis in law). Do you really mean to suggest that a legally binding signature for purposes of access to a bank account, filing taxes or signing an affidavit might be purged from a recall petition because you can’t make out the spelling of my name?

    Your alternative interpretation of the statutory language opens the door for far too much partisan mischief. For this reason, if for no other, the legislature’s adoption of a “challenge system” for resolving these particular signature issues was wise when the statute was originally passed and it remains a wise choice today.

  2. Phil Scarr

    Shorter Ed: Who’s supporting judicial activism now, Rick?

  3. Rick Esenberg

    I’m not sure how clever it is, but its fairly straight forward. The GAB must carefully ensure that the petition is sufficient on its face, i.e., based on the four corners of the petitions, it must be able to determine that a sufficient number of persons who appear to be eligible electors have signed during the relevant period.

    I don’t think you succeed in undercutting that by arguing that the GAB’s review must be limited to a set of criteria specified in sec. 9.10(2)(e). Putting aside whether the statute can be so read or can be structured in a way that makes expressio unio work, the very criteria that you say are exlusive support much, if not all, of what Judge Davis ordered for the reasons that I explained.

    I do agree that the legibility of signatures is a thornier issue and I said so. I’m not sure how I’d come down but there are, as I pointed out, reasons not to follow the rule with respect to signatures in other contexts. In this case, the signature has to serve not only as evidence of assent but as evidence of assent by an individual who is eligible to do so.

    I did not argue that there is no distinction between a signature that is facially insufficient and one that is subject to challenge. My interpretation would not require the GAB not to count any signature subject to an unexercised challenge – only those who are facially invalid. Again, the very criteria that you rely on support what Judge Davis ordered. The agency’s duty not to count duplicates is implicit in the requirement that it verify that the petioners have submitted signatures from the required numbers of electors.

    As I argued in my original post, past practice – to the extent that it actually is what you assert it to be – doesn’t help us because 1)we have very little experience with recalls before the GAB and 2) we have no experience with a recall in which these allocations of responsibility was salient.

    That being the case, I am not sure I understand how requiring the GAB to determine that all signatures are at least facially compliant with the requirements of, at least, 9.10(2)(e)and that the required numbers of electors have signed “frustrates” or unduly “delays” a recall.

    With respect to the former, there is no right to a recall based on signatures which cannot be verified or that do not add up to the required amount. With respect to the latter, deciding, as you do, that more of the work must be done by the challengers would seem to support extending the challenge period in order to permit that work to be done.

    So would I be correct in assuming that you would support an extension of the ten day challenge period to whatever period of time that is required for the challengers to do what GAB will not?

    Your suggestion that my interpretation increases the potential for “partisan” mischief assumes that partisan mischief only is – or is more likely to be – practiced by recall opponents.

    And while I agree that judges should tread lightly, this is not a case in which a court is micromanaging the way in which the GAB chooses to exercise its responsibilities. This is a case where the GAB announced that it would not exercise those responsibilities. I do think there ought to be a judicial remedy for that.

    This is one of the reasons why I think Professor Foley’s meta-narrative (which I don’t think adequately explains Bush v. Gore) is not helpful.

  4. John Foust

    In this morning’s Janesville Gazette, a writer asks “Would Walker’s signature be judged illegible by the GAB?”

  5. Christine Wilczynski-Vogel

    If you are interested in this topic, you may want to attend …

    January 19—On the Issues with Mike Gousha. Government Accountability Board Director and General Counsel Kevin Kennedy—Kennedy will be front and center during what promises to be an historic election year in Wisconsin. Will there be a recall election for Governor? What about recalls for several state Senators? How will the state implement its new voter ID law? Those are just some of the important questions facing Kennedy and the G.A.B. staff, who oversee Wisconsin’s campaign finance, elections, ethics, and lobbying laws. 12:15 to 1:15 p.m., Marquette Law School, Eckstein Hall. Reserve your spot.

  6. Ed Fallone

    Professor Edward Foley, a nationally recognized expert on election law, writing about the Wisconsin GAB:

    “In contrast to other states, Wisconsin has a state institutional structure that seems well- designed to promote evenhanded decision making. Rather than vesting power in a partisan elected official or a board controlled by one of the parties, Wisconsin has just created a Government Accountability Board (“GAB”) consisting of retired judges who must be confirmed by a supermajority of the state senate . . . This structure provides a basis for optimism that the new GAB will operate by consensus, rather than serving the narrow interest of one party or the other.”

    See here at page 193.

    My definition of “partisan mischief” would include litigation that is designed to upset a nonpartisan consensus regarding the administration of the elecion laws– litigation that at the same time is highly unlikely to result in the elimination of a sufficient number of signatures to avoid a recall.

    If a recall election is going to happen anyway, then what is the objective of the lawsuit?

  7. Ed Fallone

    The definition of “partisan mischief” might also include drafting a proposed order for Judge Davis to sign that includes a judicial recognition of “the rights of those qualified electors who choose to exercise the elective franchise by not signing a recall petition,” language that Judge Davis expressly declined to endorse in his oral ruling for the very good reason that it is unprecedented in the law. Read the proposed order of the plaintiffs and Attorney Beilin’s response here.

    In this game, what matters is inserting language into a judicial opinion that can be used to play to the broader public audience, even if that language actually has no legal significance.

    Remember the mantra that Judge Sumi had “usurped the judicial authority?” It didn’t matter that the basis of the Wisconsin Supreme Court’s ruling in Ozanne was that the legislature violated the separation of powers when drafting the open meeting law. Somehow the majority still found a way to work that phrase into its opinion.

  8. Bill Dunn

    Republicans are once again “working the refs” in order to cast doubt where little need exist and to complicate something that’s not that complicated. More than a million voters cast ballots for someone other than Walker for governor. Why is it so hard to believe that 540,000 of them, plus disaffected former Walker supporters, would not sign a recall?
    It’s all about delaying a recall.

  9. Tom Kamenick

    If, as Ed believes, the GAB review is supposed to be cursory and the officeholder’s detailed, why does the statute provide such a lengthy time for the GAB to review petitions and such a short time for the officeholder to challenge them?

  10. Phil Scarr

    Tom: That’s a question for the legislature to address, not for a judge. They can amend the statute if they deem it necessary.

  11. Rick Esenberg


    Whatever the merits of the GAB as opposed to the old Election Board might be, the question of statutory interpretation remains. It doesn’t get us very far to say that we like the GAB and so we ought to do whatever it wants. There is some degree of deference due an administering agency, but it’s not dispositive.

    Nor does it help to say that the recall election is “going to happen any way.” I suspect that the petitioners can get enough valid signatures. The public employee unions are highly motivated and are very good at the type of organization required for this type of thing. But we still have to follow the requisite procedures.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.