The opinion
Pfeifer, J.,
dissenting.
{¶ 63} As I stated in my dissent in Ohio Manufacturers’ Assn. v. Ohioans for Drug Price Relief Act, 147 Ohio St.3d 42 , 2016-Ohio-3038 , 59 N.E.3d 1274 ("OMA I”), I would dismiss this case because this courts lacks jurisdiction. It lacks jurisdiction because the challenge procedure set forth in Article II, Section lg of the Ohio Constitution does not apply to proposal petitions filed pursuant to Article II, Section lb. This court has taken on the wrong job and has done the job wrong, to the long-term detriment of Ohio election law.
Jurisdiction
{¶ 64} Article II, Section lg was amended through an Article II, Section la initiative in 2008. 2008 Am.H.J.R. No. 3 (adopted by Ohio voters in the November 4, 2008 general election). The amendment created an important role for this court in resolving challenges to petitions for statewide ballot measures: “The Ohio supreme court shall have original, exclusive jurisdiction over all challenges made to petitions and signatures upon such petitions under this section.” Ohio Constitution, Article II, Section lg. But since the passage of the constitutional amendment, this court has never been called upon to resolve a challenge to an Article II, Section lb proposal petition. The hands-on practicalities of dealing with this case—including the absurd timelines discussed below in the “Remedy” section of this opinion—demonstrate that the challenge procedures outlined in Article II, Section lg are inapplicable to Article II, Section lb *264 proposal petitions. Instead, those procedures apply to petitions for initiatives for constitutional amendments (Article II, Section la), petitions for referendums (Article II, Section lc), and supplementary petitions for initiatives to enact laws (Article II, Section lb). All of those petitions share the same filing deadline of 125 days before the election. And all of those petitions, properly signed and filed, directly result in an initiative or referendum appearing on the ballot; in contrast, a properly signed and filed Article II, Section lb proposal petition results only in the proposed law receiving consideration by the General Assembly.
{¶ 65} This court’s jurisdiction under Article II, Section lg does not begin until the filing that determines the election at which the initiative or referendum will be voted upon by the electorate. Article I, Section lg states that “[a]ny challenge to a petition or signature on a petition shall be filed not later than ninety-five days before the day of the election.” (Emphasis added.) At this point in this case, there is no “day of the election” from which to measure this jurisdictional timeline. An Article II, Section lb proposal petition with sufficient signatures does not result in getting an initiative on the ballot; it merely allows citizens to force the General Assembly to consider the proposed law. It is the supplementary petition that results in the initiative getting on the ballot. Under Article II, Section lb, the election concerning the initiative is “the next regular or general election occurring subsequent to one hundred twenty-five days after the supplementary petition is filed.” (Emphasis added.) Thus, the election day from which deadlines for a challenge under Article II, Section lg are calculated is not even set until after the filing of the supplementary petition. There is no challenge process available before this court until the supplementary petition is filed-—and that has not yet happened. This case does not belong here.
Overcounts
{¶ 66} Twice, the part-petitions in this case have been sent out to county boards of elections for verification. The county boards have painstakingly reviewed each signature on each part-petition and have verified that they are the signatures of registered voters. The professionals have done their jobs, determining that registered Ohio voters have signed the part-petitions calling for the Ohio Drug Price Relief Act to be submitted to the General Assembly for its consideration. The secretary of state has certified that there were sufficient petition signatures to merit the submission of the proposed law to the General Assembly. It was submitted to the General Assembly. More than six months later, this court, somehow, determines that that never happened.
{¶ 67} How the court gets to the conclusion has profound implications for future election cases. Most notably, this court invalidates around 9,000 signatures—which nobody disputes were genuine—because the circulators overstated the number of signatures contained on individual part-petitions. The term of art *265 is “overcounting.” In a practical sense, overcounting is corrected at the county level—workers for the boards of elections review every signature and count only the ones belonging to and matching with registered voters in the particular county. The Ohio Secretary of State’s 2015 Election Official Manual, Chapter 11, Section 1.03(D), at 11-9, http://www.sos.state.oh.us/sos/upload/elections/EOResources/general/2015EOM.pdf (accessed Aug. 11, 2016), discusses how election officials should handle overcounts:
If the number of signatures reported in the statement is equal to or greater than the total number of signatures not crossed out on the part-petition, then the board does not reject the part-petition because of the inconsistent signature numbers. [State ex rel Citizens for Responsible Taxation v. Scioto Cty. Bd. of Elections, 65 Ohio St.3d 167 , 602 N.E.2d 615 (1992).] Instead, the board must review the validity of each signature as usual.
Example: The circulator’s statement indicates that the circulator witnessed 22 signatures, but there are only 20 signatures on the petition.
Note: In determining whether the number of signatures reported by a circulator of a non-statewide candidate’s petition matches the number of signatures on that part petition, particularly with regard to crossed out signatures, board of elections should take care so as to not make a determination that is “too technical, unreasonable, and arbitrary” given the unique fact set of that petition and information available to the board, if any. [State ex rel. Schwarz v. Hamilton Cty. Bd. of Elections, 173 Ohio St. 321 , 181 N.E.2d 888 (1962); State ex rel. Curtis v. Summit Cty. Bd. of Elections, 144 Ohio St.3d 405 , 2015-Ohio-3787 , 44 N.E.3d 261 .]
(Boldface sic.)
{¶ 68} On the other hand, the same source states that when the circulator attests to having witnessed a lesser number of signatures than those that appear on the part-petition, the secretary of state’s policy is to strike the entire part-petition. Id. This difference in treatment at least makes intuitive sense: if a circulator states that he or she witnessed 20 signatures and there are 22 signatures on the part-petition, that means that there are two that he or she did not witness. How can the reviewing board of elections be expected to tell which signatures the circulator didn’t witness? On the other hand, if there are 22 signatures and the circulator states that he or she witnessed 28, the circulator has attested to having witnessed at least 22.
{¶ 69} When the secretary of state returned the part-petitions for re-review, he instructed the county boards of elections to carefully review overcounts:
*266 The Ohio Supreme Court has accorded flexibility to circulators, providing that “* * * arithmetic errors will be tolerated, but only if the error does not promote fraud.” [State ex rel. Citizens for Responsible Taxation, 65 Ohio St.3d at 172 , 602 N.E.2d 615 .] The relevant example in the Election Official Manual recognizes that “arithmetic errors” may occur.
* * *
By their nature, however, “arithmetic errors” should be isolated, unintentional oversights.
The “over-reporting of signatures” (e.g., a circulator statement purporting to witness 28 signatures on a part-petition bearing only two signatures) is so strikingly prevalent in this submission that the suggestion that unintentional “arithmetic errors” are to blame strains credulity. This cannot be the result envisioned by case law; otherwise the exception would swallow the rule.
Secretary of State Directive 2016-01, at 2-3, http://www.sos.state.oh.us/SOS/Upload/elections/directives/2016/Dir2016-01.pdf (accessed Aug. 14, 2016).
{¶ 70} After that guidance, the part-petitions were reviewed again and ultimately certified by the secretary of state. Now those part-petitions are before this court.
{¶ 71} The secretary of state relied on this court’s decisions in explaining his policy on overcounts. We should follow our decisions, too. In State ex rel. Citizens for Responsible Taxation v. Scioto Cty. Bd. of Elections, 65 Ohio St.3d 167 , 602 N.E.2d 615 (1992), this court held that the board of elections had improperly rejected five part-petitions for overcounting; in each instance the circulator statements “indicated one more signature than each part-petition actually contained.” Id. at 171 . The court held that incorrectness in the signature count is not fatal to an entire part-petition:
R.C. 3501.38(E), however, does not expressly mandate a correct signature total, and [State ex rel.] Loss [v. Lucas Cty. Bd. of Elections, 29 Ohio St.2d 233, 233 , 281 N.E.2d 186 (1972)] implies that arithmetic error will be tolerated, but only if the error does not promote fraud. Indeed, Loss may explain why the Secretary of State instructed respondents here to reject an entire part-petition only where the circulator states a number “less than the total number of uncrossed out signatures” (emphasis sic) and to, in effect, overlook discrepancies in the number of signatures “in all other instances.”
*267 State ex rel. Citizens for Responsible Taxation at 172.
{¶ 72} This court “accept[ed] the Secretary of State’s reading of the signature-total requirement, see State ex rel. Beck v. Casey (1990), 51 Ohio St.3d 79, 81 , 554 N.E.2d 1284, 1286 , and conclude[d] that respondents improperly rejected the instant five part-petitions for noncompliance with R.C. 3501.38(E).” State ex rel. Citizens for Responsible Taxation at 173.
{¶ 73} A few months ago in OMA I, this court dealt with respondents’ argument that as a matter of law, part-petitions can never be rejected on the basis of an overcount. Respondents relied on Citizens for Responsible Taxation, but this court’s opinion stated that “Citizens for Responsible Taxation did not hold that an overcount can never promote fraud.” OMA I, 147 Ohio St.3d 42 , 2016-Ohio-3038 , 59 N.E.3d 1274, at ¶ 20 . At that time, this court recognized that whether an overcount promotes fraud should determine whether the overcount should result in the disqualification of an entire part-petition. Today, any overcount results in complete disqualification.
{¶ 74} In OMA I , the court reviewed this court’s jurisprudence on how to treat undercounts; again, the presence of fraud is a determining factor:
In cases in which the circulator’s statement slightly undercounts the signatures, this court has ordered the entire part-petition counted, so long as there is no indication of fraud or material misrepresentation. State ex rel. Curtis v. Summit Cty. Bd. of Elections, 144 Ohio St.3d 405 , 2015-Ohio-3787 , 44 N.E.3d 261, ¶ 8 ; State ex rel. Schwarz v. Hamilton Cty. Bd. of Elections, 173 Ohio St. 321 , 323, 181 N.E.2d 888 (1962). Only when the circulator knomngly submits an undercount has the court invalidated the entire part-petition. See, e.g., Rust v. Lucas Cty. Bd. of Elections, 108 Ohio St.3d 139 , 2005-Ohio-5795 , 841 N.E.2d 766 , ¶ 13-14.
(Emphasis sic.) OMA I at ¶ 19 .
{¶ 75} Has the longstanding practice of disqualifying undercounts and allowing overcounts been completely reversed? At the very least, shouldn’t the majority employ the “indication of fraud or material misrepresentation” test on the overcounts in his ease?
{¶ 76} Instead, this court decides for the first time in this case that any overcount invalidates an entire part-petition. This court makes no attempt to determine whether fraud occurred in any instance or whether an error may have been one of arithmetic or a simple oversight. As an example, consider Lucas County part-petition No. 000285. The part-petition contains 27 signatures (19 *268 ultimately were determined to be good by the board of elections), but the circulator attested to witnessing 28 signatures. The signatures of all 19 registered voters who signed that part-petition do not count. What about Ashland County part-petition No. 000015? It contains at least 27 signatures (20 of which were ultimately determined to be valid by the board), with what is perhaps the start of a 28th signature marked out. The circulator attested to 28 signatures, and this court has invalidated the entire part-petition. Franklin County part-petition No. 000230 has 26 signatures (21 were determined to be valid) and two other signature spaces with an “X” through them. The circulator attested to 28 signatures; all 21 valid signatures are now discounted. Carroll County part-petition No. 000001 contains 27 signatures and has three “X”s placed on signature line 14. The circulator attested to 28 signatures; the 25 found valid by the board of elections have now been disqualified by this court. Warren County part-petition No. 000066 contains 25 signatures, and a large “X” fills the last three signature lines on the part-petition. The circulator attested to 28 signatures; the 16 deemed valid by the board of elections have been disqualified by this court. This quick, nonexhaustive count of just these part-petitions involves over 100 different registered voters whose signatures were collected by five different circulators in five different counties. Those signers’ desire to place the proposal petition in front of their elected representatives has been thwarted by this court’s one-size-fits-all mandate on overcounts.
{¶ 77} Instead of reviewing the individual part-petitions, this court has relied on an exhibit submitted as evidence by the challengers. That exhibit, according to a supporting affidavit, is a spreadsheet created by the “Litigation Support Manager” at Bricker <& Eckler, L.L.P. (“Bricker”), the challengers’ counsel, “based upon Bricker’s review of the part-petitions.” The spreadsheet notes the instances in which Bricker has determined that a circulator attested that he or she witnessed 28 signatures when the part-petition did not contain 28 signatures. The spreadsheet also includes the number of valid signatures contained on each of those part-petitions, to make this court’s invalidation of signatures of registered voters easy.
{¶ 78} There are also a large number of part-petitions in this case containing just one signature, when the circulator has attested to witnessing 28. If the requirement that a circulator state the number of signatures personally witnessed “ ‘is a protection against signatures being added later.’ State ex rel. Loss v. Lucas Cty. Bd. of Elections, 29 Ohio St.2d 233, 234 , 281 N.E.2d 186 (1972),” lead opinion at ¶ 44, why is the court disqualifying all the part-petitions that contain only one signature? Where are the signatures that could have been added later? They are not on the part-petitions.
*269 {¶ 79} The court’s decision is “too technical, unreasonable and arbitrary,” a mistake this court has warned boards of elections against. See State ex rel. Schwarz v. Hamilton Cty. Bd. of Elections, 173 Ohio St. 321 , 323, 181 N.E.2d 888 (1962). The court’s decision has implications at every level: every person running for offices—from school board member to governor to coroner, and those who support them—-should live in fear that small mistakes will lead to the invalidation of entire part-petitions and the end to candidacies.
Circulator Addresses
{¶ 80} The court disqualifies 632 signatures on the part-petitions submitted by Kacey Yeliquette, a transient circulator. The lead opinion, paying extreme short shrift to respondents’ argument that the permanent-address requirement violates the First Amendment to the United States Constitution, concludes that “her part-petitions could be deemed valid only if we were to rule that it is unconstitutional to require circulators to provide some means of locating them.” (Emphasis sic.) Lead opinion at ¶ 40. But Veliquette has provided a way to locate her: paid circulators must provide “the name and address of the person employing the circulator to circulate the petition, if any.” R.C. 3501.38(E)(1). She provided that address. And her affidavit has been filed in this case, so she has been located. Only the 632 verified signatures that she collected from registered voters are missing.
Remedy
{¶ 81} The court gives respondents ten days to gather additional signatures in support of their proposal petition, but to what end? First, we must overcome the absurdity of the court’s continued belief that the processes under Article II, Section lg apply to proposal petitions. Pursuant to Article II, Section lg, the secretary of state must determine the sufficiency of the additional petitions “not later than sixty-five days before the election.” When is “the day of the election?” It is unknowable at this point, because the election date is measured from the date of the filing of the supplementary petition—pursuant to Article II, Section lb, the proposal shall be submitted to the electorate at the “next regular or general election occurring subsequent to one hundred twenty-five days after the supplementary petition is filed.” There has been no supplementary petition filed. So by what date must the secretary of state make his sufficiency determination regarding the additional signatures? How does one count 65 days from never? Under Article II, Section lg, any challenge to the additional signatures must be made “not later than fifty-five days before the day of the election.” Again, it is unknowable what that date is, other than that it is probably ten days closer to never.
*270 {¶ 82} But let us assume, as the court does, that the Article II, Section lg challenge process does apply to proposal petitions. That section grants initiative proponents ten additional days to file additional signatures “if the petition or signatures are determined to be insufficient.” The court gives respondents that extra ten days, but then what? The lead opinion states that if, after respondents’ submission of additional signatures, the secretary of state determines that there are a sufficient number of valid signatures, “then he shall resubmit the initiative to the General Assembly, in accordance with the terms of Ohio Constitution, Article II, Section lb.” Lead opinion at ¶ 47. I assume that this directive means that the secretary of state will transmit the part-petitions to the General Assembly as soon as any challenge to the additional signatures has run its course—though measured by a timeline with benchmarks related to the “day of the election,” (which will have to be ignored) the collection, review, and challenge process can consume up to 40 days—likely sometime in September 2016. But in the meantime, respondents have been seriously prejudiced.
{¶ 83} Respondents have given up trying to get their initiative on the November 2016 ballot; they wrote in their “Notice of continued circulation of supplemental petition,” filed in this court on July 13, 2016:
Although Respondent Secretary’s actions adversely affected Petition Respondents’ ability to submit their Supplementary Petition before the deadline to appear on the November 2016 ballot, Petition Respondents have until September 2, 2016 to collect a sufficient number of signatures to place the Ohio Drug Price Relief Act on the 2017 general election ballot, which they plan to do.
{¶ 84} The four-month period for consideration of their proposal initiative by the General Assembly under Article II, Section lb did not end until June 4. Respondents could not begin to collect signatures on supplementary part-petitions until that date. Pursuant to Article II, Section lb, they had 90 days to collect signatures from an additional 3 percent of the electorate. However, because of the month-long delay in transmitting the proposal petition to the General Assembly at the front end, respondents had just over 30 days to obtain their supplementary signatures if they wanted to get their initiative petition on the November 2016 ballot. This is because the initiative goes on the ballot at the next scheduled election “occurring subsequent to one hundred twenty-five days after the supplementary petition is filed.” Ohio Constitution, Article II, Section lb. The 125-day deadline for the 2016 ballot was July 6, 2016. Despite the fact that they now are not going to seek to reach the ballot until 2017, respondents have another deadline looming. Their supplementary petition—with its attend *271 ant signatures from an additional 3 percent of the electorate—has to be submitted to the secretary of state within 90 days of the expiration of the General Assembly’s four-month consideration period, which ended on June 4, 2016. That 90-day deadline expires September 2, 2016. Without knowing how this court was going to respond to today’s challenge, respondents had to keep collecting signatures. If this court had announced a decision rejecting all of the challenges today, respondents would have had only 18 days to collect an additional 91,000 plus valid signatures. Now, all the work that respondents have done since June 4, 2016, has been wasted. They have been told by this court that the curative period to collect additional signatures required for the proposal petition can actually cure very little. They have to resubmit the proposal petition and then gather another additional 91,000 signatures on the supplementary petition. This is fundamentally unfair and the wrong interpretation of Article II, Section lg.
{¶ 85} The lead opinion’s remedy runs contrary to the curative process of Article II, Section lg, which exists to allow for deficiencies in petitions to be corrected. The cure relates back to the time of the deficiency. For instance, a supplementary petition must be filed within 125 days of the next election. Article II, Section lg provides that “[i]f the petitions or signatures are determined to be insufficient, ten additional days shall be allowed for the filing of additional signatures to such petition.” This gives initiative supporters ten days to cure their deficiency by gathering additional signatures. Under Article II, Section lg, the appeal-and-review process can last until 45 days before the election. Once that is complete and the additional signatures are judged sufficient to make up the deficiency, the initiative goes on the ballot. That is, despite the fact that the deficiency is corrected within 125 days of the election, it is as if the supplementary petition had sufficient signatures before the 125-day deadline. The curative fix relates back to the original filing of signatures.
{¶ 86} Likewise, the additional signatures in this case, if certified by the secretary of state, should relate back to the secretary of state’s February 4, 2016 transmission of the proposal petition to the General Assembly. The fix would establish that the proposal was indeed properly before the General Assembly for its four-month review. The four-month review did happen, and it ended on June 4. The rest of the timetable should continue forward from that—leaving respondents until September 2, 2016, to file their supplementary petition to get on the 2017 ballot.
{¶ 87} But no, the lead opinion concludes, the proposal petition is fixed only as of the date the additional signatures are verified. But what if we were to apply that logic to a case in which a supplementary petition did not contain sufficient signatures 125 days before the election but the initiative supporters corrected that in the ten-day curative period? According to the lead opinion, the supple *272 mentary petition would not meet the signature threshold until the curative additional signatures were added, measured from the date that they were added. The initiative would fail to qualify for the ballot at the next election, because the 125-day deadline would have been missed. We know that is not the case because under Article II, Section lg, the challenge process continues, measured from the date established as the election day by the originally defective supplementary petition.
Bricker & Eckler, L.L.P., Kurtis A. Tunnell, Anne Marie Sferra, Nelson M. Reid, and James P. Schuck, for challengers.
{¶ 88} The curative nature of the additional signatures relates back to the deficiency in the petition that they are curing. So respondents should have ten days to fix any deficiency, and that fix should relate back to the proposal submission to the General Assembly of February 4, 2016.
Conclusion
{¶ 89} The lead opinion states, “It is generally our obligation to defer to the secretary of state’s reasonable interpretation of an election statute. State ex rel. Cornerstone Developers, Ltd. v. Greene Cty. Bd. of Elections, 145 Ohio St.3d 290 , 2016-Ohio-313 , 49 N.E.3d 273, ¶ 16 .” Lead opinion at ¶ 26. The secretary of state has determined that there were sufficient registered voters from the requisite number of counties who actually signed the part-petitions that were timely filed and twice checked by local, properly trained election officials to assure that the signatures matched the voter-registration cards. We should defer to that judgment.
{¶ 90} The sweeping reach of this per curiam opinion—handed down without parentage and without oral argument—marks a sea change for the Ohio Supreme Court. In ProgressOhio.org, Inc. v. JobsOhio, 139 Ohio St.3d 520 , 2014-Ohio-2382 , 13 N.E.3d 1101 , this court made it clear that Ohio citizens have little hope to meet the requirements to challenge enactments of the General Assembly in the courts of Ohio. Now, the court chooses to make it extremely difficult for Ohio citizens to even politely exercise their fundamental constitutional right to petition the General Assembly to take action on important matters affecting millions of Ohio citizens. The fallout from today’s decision will undoubtedly multiply the challenges to both state and local citizen petitions of all types as well as to the efforts of any citizen seeking to stand for election to public office. Any suggestion that our court’s decision is protecting Ohio citizens from election fraud is a complete fiction. What we have is a government protecting itself from its people. Today, the walls got higher.
*273 McTigue & Colombo, L.L.C., Donald J. McTigue, J. Corey Colombo, and Derek S. Clinger, for respondents William S. Booth, Daniel L. Darland, Tracy L. Jones, and Latonya D. Thurman.
Michael DeWine, Attorney General, Steven T. Voigt, Senior Assistant Attorney General, and Brodi J. Conover, Assistant Attorney General, for respondent Ohio Secretary of State Jon Husted.