In my opinion the proposed amendment has been adopted. *Page 626 This court properly assumed original jurisdiction in this case because of the absence of an adequate remedy in any other forum, and because of the great public interest in the question involved. There is in evidence before us, by stipulation of the parties, all the records of the election at which the question was submitted. The case has been twice orally argued and extensively briefed. I think this court has the legal authority and power to give complete relief, and that it is our duty to do so. Therefore, I feel justified in giving my reasons for withholding my assent to the majority opinion.
It may be conceded that under section 1, art. 24, of our Constitution, it is necessary that a proposed amendment receive a majority of the total number of electors voting at the election, and not merely a majority of electors voting on the amendment itself. But the question remains as to how we are to determine the total number of electors voting at the election. There are three possible methods, namely: (1) The highest vote cast for any office or measure at the election held November 3, 1936; (2) the total ballots issued throughout the state, less spoiled ballots; (3) the certificate of the election officers as to "the total number of electors voting in such election" as required under section 5886, O. S. 1931. The rotators contend for the first, the State Election Board applied the second, and the Legislature has prescribed the third.
The majority opinion criticizes the first method for the reason that it would exclude those persons who voted but who did not vote for the particular office or measure which was used as the test. I recognize this to be a legitimate criticism. However, the second method, which was used by the State Election Board, has been expressly rejected and disapproved in the cases of State ex rel. Saunders v. Clark (Neb.) 82 N.W. 8, and Hopkins v. City of Duluth (Minn.) 83 N.W. 536, and it is apparent, from the evidence in this case, that it does not represent correctly the number of electors voting at the election. For illustration, it is shown that the number of ballots issued, less spoiled, for Presidential Electors was 762,654, while the vote actually cast and counted for Presidential Electors was 749,740, which demonstrates that there were 12,914 ballots for Presidential Electors which were issued and not spoiled, but never accounted for. These were blank ballots, mutilated ballots, or otherwise illegal ballots, which were included, since they were not considered as spoiled, but which should not, by the holding of this court, have been counted in determining the total number of electors voting. In the case of Board of Education of Oklahoma City v. Woodworth (1923) 89 Okla. 192, 214 P. 1077, it was held that blank ballots and mutilated ballots should not be taken into account in determining the aggregate number of voters voting at an election where a determination of that question was necessary in determining whether a proposed bond issue had been adopted or rejected. "Spoiled ballots" are defined in section 5730, O. S. 1931, and they do not include blank ballots or mutilated ballots which are not marked "Spoiled" as prescribed in the statute. Thus the method used by the State Election Board which rejected only spoiled ballots and included blank ballots, mutilated ballots, and otherwise illegal ballots, is contrary to the decisions of this court. This method is not approved by the majority in this case, and it cannot be used without overruling our previous holding.
The third method is based upon section 5886, O. S. 1931, which was enacted in 1910. It requires the precinct election officials to certify to the county election board "the total number of electors voting at such election," and that the county election board transmit a similar certificate to the State Election Board. But this section does not lay down any yardstick for the election officials to follow in determining the number of electors voting in their precinct. It is left for them to arbitrarily make this determination as each shall see fit. In so far as it can be said that the majority suggests a means of determining this question, this method is the one receiving their sanction. But, irrespective of the assertion that there has been a substantial compliance with this provision, there was in fact no attempt to comply with this section and it never has been complied with since it was enacted in 1910.
I am of the opinion, therefore, that there is no way of determining to a mathematical certainty how many electors voted at the election. Under such circumstances, the only sensible thing to do is to make the determination as accurately as possible by examining the best evidence obtainable, and adopt the method which will most nearly *Page 627 approximate the correct number and afford a definite criterion for such determination. In using the language in the constitutional provision involved here, no doubt the framers of the Constitution were aware that it would be necessary to determine as a fact how many electors voted at the election at which a constitutional amendment would be submitted. They no doubt also realized the impossibility of making this determination to a mathematical certainty, and presumed that it would be done by common sense methods. This constitutional provision is not peculiar to Oklahoma. The identical language, at the time of the adoption of our Constitution, was found in the Constitutions of Mississippi, Arkansas, Nebraska, Illinois, and Ohio. 12 C. J. 694; Knight v. Shelton (1905) 134 F. 423. We must assume that the members of our Constitutional Convention, in framing the provision in question, had before them the Constitutions of these states and intended that we should adopt the construction previously placed thereon by the other states. In 12 C. J. 717, the rule is stated:
"When provisions have been adopted into the Constitution of the state, which are identical with or similar to those of other states, it will be presumed that the framers of such Constitution were conversant with, and designed to adopt also, any construction previously placed on such provisions in such other states."
Prior to the adoption of our Constitution, Nebraska and Arkansas used the highest vote for any office or measure as the basis for determining the total number of "electors voting at such election."
Section 1, article 15, of the Constitution of Nebraska (1875), requires a "majority of the electors voting at such election," to adopt a constitutional amendment. In Tecumseh Nat. Bank v. Saunders (Neb. 1897) 71 N.W. 779, in the body of the opinion the court said:
"From a consideration of all the cases decided by this court which have a bearing upon this subject our conclusion is that to secure the adoption of an amendment to our Constitution it is necessary that the favorable votes be in excess of one-half of the highest aggregate number of votes cast at said election, whether such highest number be for the selection of an officer or upon the adoption of a proposition."
The majority opinion states that the question involved in the instant case was not considered by the Nebraska court. If the question referred to is what method shall be used to determine the number voting at the election, I cannot agree with that statement. After rejecting the argument that a majority of votes cast on the amendment itself was sufficient, the court took the total vote cast for Governor, as the office or measure receiving the highest vote, and required the amendment to reach a majority of that number, concluding with the above quotation. It appears clear that the court, by this procedure, interpreted and applied their constitutional provision as contended for by relators.
In applying the constitutional provision of Arkansas, which requires "a majority of the electors voting at such election" in order to adopt a constitutional amendment, the Supreme Court of that state, in Rice v. Palmer (1906) 96 S.W. 396, applied this method and cited with approval Knight v. Shelton (1905) 134 F. 423, construing the same constitutional provision, and which also applied this method.
The majority opinion makes the same criticism of the Arkansas case. In this case, the holding of the court was that the majority of electors voting at the election was the test as distinguished from a majority voting for the measure in question, but in determining that the measure did not pass, it was pointed out that it did not receive a majority of votes cast for Governor. Inferentially therefore, if the measure had received a majority of votes cast for Governor, it would have passed; so we may infer that this method was employed.
Also, since the adoption of our Constitution, this method has been recognized and employed in other states. Section 273 of the Constitution of Mississippi (1890) requires that "a majority of the qualified electors voting" shall be necessary to amend its Constitution. In State v. Brantley (Miss. 1917) 74 So. 662, in the second paragraph of the syllabus, it is said:
"In the absence of correct certification of the number of electors voting upon a constitutional amendment submitted at a general election, the court must presume that the highest number of votes cast for candidates for any one office represented the number of electors voting, so that a vote for the amendment to the Constitution (Laws 1916, c. 159) providing for the initiative and referendum *Page 628 was adopted where the vote therefor was 19,118 and the highest vote cast for any office was 37,583."
In this case it appears that after the returns from the various counties were received by the Secretary of State he requested the local election boards of each county to reconvene and ascertain and certify to him the total number of votes cast in each county, and that the election commissioners of all the counties except six did reconvene and certify the total number of votes cast in their counties, from which it appears that 40,070 qualified electors voted. Thus, it is seen that the total vote for the constitutional amendment of 19,118 was more than one-half the number of votes cast for the highest office, but was less than one-half the number of electors voting at the election as certified by the election officers. The court in declaring the amendment adopted, used the following language:
"Consequently, as was done in State v. Jones, 106 Miss. 522, 64 So. 241 (and, for that matter, as has heretofore been done each time an amendment has been inserted in the Constitution), we must presume that the highest number of votes cast for any officer represents the total number of votes cast at the election, from which it follows that we must hold that the amendment received the required number of votes."
It is said in the majority opinion that there was no determinative rule of law intended to be laid down in the Mississippi case. The quotation in the majority opinion, taken from that case, appears in per curiam opinion, which was later supplemented on "suggestion of error" (analogous to our petition for rehearing) by the opinion of Chief Justice Smith in which the court definitely and unequivocally prescribed this test as the rule of law determinative of the case.
This rule was adhered to in State v. Cato (1923 Miss.) 95 So. 691, and the proposed constitutional amendment was declared not adopted because it did not receive one-half of the highest number of votes cast for any office or measure. In People v. Stevenson (1917) 281 Ill. 17, 117 N.E. 747, the court in construing a similar constitutional provision used language indicating that the method used was to take the highest vote cast at the election for any office or measure. For the failure to obtain a majority of the number on the poll list, or the vote for Presidential Electors, or the vote for Governor, even though it exceeded the vote for certain other state officers and members of the General Assembly, the constitutional provision failed.
In the states of Ohio and Illinois this rule was used in voting upon a special tax levy under a statute similar to the above constitutional provision (Enyart v. Trustees of Hanover Twp., 1874, 25 Ohio St. 618) and in an election upon a removal of a county seat under a constitutional provision similar to the above constitutional provision (People v. Wiant, 1868,48 Ill. 263).
Idaho, Wyoming, and Indiana have the same constitutional provision, which requires "a majority of the electors" to adopt a constitutional amendment, and it has been held in Idaho that all that is required is that it receive a majority of the votes east on the amendment (Green v. State Bd. of Canvassers, 1896, 47 P. 259), whereas in Wyoming it must receive a majority of all electors voting at the election (State v. Brooks, 1909, 99 P. 874). In Indiana, the court for 50 years held that in order to adopt a constitutional amendment it must receive a majority of all the electors voting at the election, to be determined by the highest vote cast for any office or measure (In re Denny, 59 N.E. 359; In re Boswell, 100 N.E. 833.); but recently, in the case of In re Todd (1935) 193 N.E. 865, the Indiana court reversed the former decisions, and held that under their constitutional provision a majority of votes cast on the amendment itself was sufficient.
Thus it is seen that the impossibility of determining "a majority of the electors" has caused the courts of Idaho and Indiana to adopt an interpretation that will assure accuracy and certainty. I am not suggesting, however, that such interpretation be applied to our Constitution, but it does indicate our need for a simple and understandable method.
The practice of the executive officers of the state is persuasive in determining the best method to be used, and in this connection the majority opinion is misleading. Governor Haskell, who was a member of the Constitutional Convention, at the first election in 1908, in his proclamation on the result of the vote on two constitutional *Page 629 amendments, used the votes for Governor as the total number of electors voting at the election. He applied the same rule in 1910. Under the agreed statement of facts in this case, it appears that at every election from the adoption of the Constitution until the general election in 1916, the Governor and the State Election Board adopted this method of determining the number of electors voting at the election at which a constitutional amendment was submitted. The record shows that since 1916, no uniform rule has been followed in determining the number of electors voting at the various elections at which amendments were submitted, but it appears that in several of such elections they have used the highest number of votes cast for any office or measure as the test, specifically in the general election of 1918, 1920, 1930, and 1934. This construction, applied during the first eight years after statehood, strongly supports our theory of the law. The interpretation of the greatest weight is that applied within a reasonable time after the adoption of the Constitution. Foote v. Town of Watonga (1913) 37 Okla. 43, 130 P. 597; 12 C. J. 712; Knight v. Shelton, supra.
Due to the inconsistency of the practice since 1916, in that no one method for determining the vote has been used, it is advisable that we establish some practical method of ascertaining the facts. If the tion 5899, O. S. 1931, provides in part that then the returns made thereunder as to the "number of electors voting" probably would be prima facie correct, no matter how they were compiled. It is my opinion, however, that in carrying into effect the provisions of this statute, the highest number of votes for any office or measure in each precinct should be used as the test, as being the best evidence available. But that question is not before us now for the reason that there was no attempt to comply with the statute, and under these circumstances I believe that the method which I have advanced and supported by authority should be adopted as the test.
Moreover, it is consistent with other provisions of our Constitution. Section 2, art. 5, of the Constitution, after specifying the per centum of legal voters required to initiate a proposed legislative measure or constitutional amendment, or to refer a legislative enactment, provides: "The ratio and per centum of legal voters hereinbefore stated shall be based upon the total number of votes cast at the last general election for the state office receiving the highest number of votes at such election." I think this is a reasonable test to apply in determining the "majority of all the electors voting at such election" under section 1, art. 24, of our Constitution, in the absence of a compliance with section 5886, O. S. 1931.
It is asserted in the majority opinion, however, that even under this method, as appears from a certificate filed herein pursuant to the special order of this court, the amendment was not adopted. I cannot agree with this assertion, but before proceeding further, it is well to make clear just what has transpired. This action originated here upon writ of certiorari sought by relators. Pending determination of the writ, and after oral argument of counsel, this court made the following order:
"Whereas, it is stipulated by the parties in the above styled and numbered cause as follows: 'It is expressly stipulated that all of the county returns or certificates of the election held on November 3, 1936, may be considered as being in evidence in this case.' It is ordered by the court that relators file anamendment to their pleadings by alleging the facts, to be determined either by stipulation of the parties or by certificates from proper county authorities, showing the total maximum number of votes to be computed by taking the highest vote cast and counted for any office or measure either national, state, county or precinct, in each precinct and theState Election Board is hereby directed to certify such number, after computation and compilation thereof to this court, as a part of the record of this case."
Thereupon, the State Election Board filed a certificate showing such total number of votes to be 760,055, ascertained from returns which were certified to them from the county election boards. This certificate of the State Election Board, having been made pursuant to the directions of this court, is not an official act of that body, but partakes of the nature of a referee's report and is of no greater weight. Relators then filed an amendment to their pleadings and a stipulation in which they agreed to the number as to all the precincts except 81, and as to these remaining *Page 630 81 precincts they presented the figure which they contend represents the total vote sought under this order, ascertained from records in evidence in this case, and excepted to the figure presented by the State Election Board as to those 81 precincts. If the number of votes in these 81 precincts presented by relators is taken as correct over the number presented by the State Election Board, the amendment carried. The majority, in its opinion, refuses to determine the correct number of votes in these disputed precincts, and conclude that relators have not sustained the burden of proving that they are clearly entitled to the writ. This eludes the true issues of this case which, by the stipulation of the parties and the action of the court itself, are now properly before us.
The majority opinion does not pass upon the question of whether the action of the State Election Board in canvassing the returns and certifying to the Governor the result of said election is purely ministerial or whether judicial or quasi judicial. Section 5899, O. S. 1931, provides in part that it shall be the duty of the State Election Board "to certify to the Governor, immediately upon receipt of all the returns upon any proposition, the result thereof". It may be conceded that the mere process of tabulating the votes cast at an election is a purely ministerial act, but to determine whether a constitutional amendment has been properly adopted according to the requirements of the Constitution calls for an interpretation of the language of the Constitution. I take it to be well settled under the following authorities, that the determination of whether a proposed amendment has been engrafted upon the Constitution is the exercise of a judicial function. State ex rel. Short v. State Board of Equalization (1924) 107 Okla. 118, 230 P. 743; McConaughy v. Secretary of State (Minn.) 119 N.W. 408; Bott v. Wurt (N.J.) 43 A. 744; Rice v. Palmer, supra.
Thus certiorari would lie, but if this proceeding be considered as being based strictly upon that writ, then it may be that the majority opinion is correct in refusing to determine the correctness of the figure presented by the Election Board as representing the total votes cast in the 81 precincts in dispute. Under this writ the court could only review jurisdictional questions and it can be said that if the Election Board has jurisdiction or power to declare the amendment defeated, it can exercise that power correctly or incorrectly. No matter how erroneous it might be in its selection of a method of arriving at the total number of electors voting, it still would not exceed its jurisdiction, provided it had acted in accordance with the constitutional and statutory provisions.
But it is not necessary to decide that question, for this case has long passed the stage where it can be said to be here strictly upon writ of certiorari. The petition of relators not only prayed for the writ, but asked this court to "determine whether said amendment has become a part of the Constitution of the State of Oklahoma; and further prays for such other and further relief as may to the court seem just and proper in the premises". Furthermore, the parties, on February 24, 1937, stipulated that certain attached exhibits might be "considered by the court in determining the issues raised by the pleadings in this case", and that all "county returns or certificates of the election held on November 3, 1936, may be considered as being in evidence", and further that "any of the records in the offices of the State Election Board and Secretary of State may be used by either of the parties hereto, or may be examined by the court, in so far as the same may be material to any of the issues in this case." With this pleading and stipulation before it, this court made the order quoted above directing the relators to amend their pleadings and allege, and the State Election Board certify, the highest vote cast for any office or measure. This is a command by the court that the parties bring before it certain items of evidence which is in the record by stipulation, pertaining to the method of computing the total vote as herein propounded. This is not a jurisdictional matter, and therefore the court has already considered this proceeding in a broader sense then pure certiorari as reflected by the majority opinion. We have ordered that this evidence be brought before us. Why do we decline to determine if it is in fact the correct evidence in the record on this subject? The majority opinion states that we are impotent to act and that it will be "resolved into a glorified election contest".
But we are by no means powerless to act, and where can be found a better forum to *Page 631 stage such a free-for-all combat, if the majority of the court so considers this proceeding, than in this court? The essential purpose of this proceeding is to determine whether an amendment to the Constitution has been rejected or approved by the people, and this court is open and has acted toward that end. In section 2 art. 7, of our Constitution it is provided that "The original jurisdiction of the Supreme Court shall extend to a general superintending control over all inferior courts and all commissions and boards created by law". Under this section the Supreme Court has power, in the exercise of its original jurisdiction, to issue writs of habeas corpus, mandamus, quo warranto, certiorari, prohibition, and such other remedial writs as may be provided by law, and hear and determine the same.
In Matney v. King (1908) 20 Okla. 22, 93 P. 737, the court, in discussing this power, said:
"This provision practically places the Supreme Court in the same relation to the inferior courts of the state as the court of the King's Bench bore to the inferior courts of England under the common law. Chief Justice Marshall, in an early case (Ex parte Crane et al., 5 Pet. (U.S.) 190, 8 L. Ed. 92), discussing the supervisory powers of the Supreme Court of the United States over the proceedings of the inferior courts by the writ of mandamus according to the principles of the common law, said: 'In England the writ of mandamus is defined to a command issued in the King's name from the court of King's Bench and directed to any person, corporation, or inferior court of Judicature within the King's dominions, requiring them to do some particular thing therein specified, which appertains to their office or duty, and which the court of King's Bench has previously determined, or at least supposes, to be consonant of the right and justice. Blackstone adds that: "It issues to the judges of any inferior courts, commanding them to do justice according to the powers of their office whenever the same is delayed. For it is the peculiar business of the court of King's Bench to superintend all other inferior tribunals, and therein to enforce the due exercise of those judicial or ministerial powers with which the Crown or Legislature have invested them, and this, not only by restraining their excesses, but also by quickening their negligence and obviating their denial of justice." ' "
In State v. Kight (1915) 49 Okla. 202, 152 P. 362, the court said that "in some states the courts have held that this superintending control is as broad as the exigencies of the case demand (State ex rel. Bayha v. Philips, 97 Mo. 331, 10 S.W. 855, 3 L. R. A. 476; State v. Dist. Court of First Judicial Dist., 24 Mont. 539, 63 P. 395); and where the writ, issued in the ordinary course of the common law, is not sufficiently broad or effective to accomplish the result necessary, the courts of some states hold that the same will be enlarged or new writs will be devised to meet the particular needs of the case (State v. District Court, First Judicial Dist., supra; State ex rel. New Orleans v. Civil Dist. Judge, 52 La. Ann. 1275, 27 So. 697, 51 L. R. A. 71."
Thus it is clear, under this grant of power in the Constitution, that this court has authority to employ whatever writ is appropriate to meet the exigencies of the case, within the legal bounds of the writ chosen. In view of the fact that the issues and the evidence are here before us and we have acted to further assimilate that evidence, we need not stifle ourselves with the strict limits prescribed for the writ of certiorari. In California, the court, when prevented from acting upon a question of great public importance, by the strict rules of certiorari, held, in Board of Trustees, etc., v. State Board of Equalization (1934 Cal.) 37 P.2d 84, that having all facts before it by stipulation of the parties, it may give appropriate relief warranted by the record, though the petition demanded the wrong relief. The proceeding before us now has been called certiorari, but we have not treated it as such, nor must we be now compelled to do so, when under the power of superintending control we may in effect consider such writ as abandoned and apply the remedy which will effectuate justice without doing violence to any established principles of law. It is evident that if there is a remedy by which the proposed amendment may be declared adopted or rejected by the method which will best determine the "total number of electors voting" as prescribed by the Constitution, we should not be powerless to examine the evidence submitted pursuant to that method to ascertain how that remedy shall be exercised. In making such determination we should consider the best evidence.
If we accept the number of votes for the highest office or measure as certified by the State Election Board pursuant to the order of this court, the amendment was defeated under this method by 623 votes. But the relators contend that there are clerical errors in 81 precincts which are apparent from the *Page 632 face of the record and which are more than sufficient to change the result of the election. We find that this contention is fully supported by the evidence before us; in fact, it appears that the figure used by the State Election Board as the total number of votes cast for the highest office or measure in some precincts exceeds even the total number of ballots issued in those precincts.
There are several types of clerical errors. First, errors due to the posting of the vote in one precinct in the wrong column so that it represents the vote in another precinct. For example, the figure used by the Election Board for 13 precincts in Cherokee county, taken from a copy of the tally sheet on file in the office of the county clerk, exceeded the total number of ballots issued in those precincts by 845. It exceeded the total number of votes for the highest office or measure, taken from the same tally sheet and the returns in the State Election Board, by 895 votes. The error of the State Election Board arose in the following unique manner: The county election board in making the copy of the tally sheet placed the figures representing the votes received by the county assessor in the wrong column as illustrated below:
Tah.
County Court Ward 3 Ward 4 Critt. 1 Critt. 2
Clerk Pct. 2
_______________________________________________________________
X _________( 239 ( 280 ( 109 ( 105
Y _________( 54 ( 69 ( 26 ( 61
___ ___ ___ ___
( (293) ( (349) ( (135) ( (166)
( ( ( (
( ( ( (
( ( ( (
County Ass'r ( ( ( (
X __________( 221 ( 279 ( 102 (
Y __________( 59 ( 56 ( 28
___ ___ ___
( (280) ( (335) (130)
etc.
Thus it is seen that the vote for county assessor in ward No. 4 was placed in the column representing the vote in Critt. No. 1. Ward No. 4 is a large precinct and the vote for assessor totaled 335, while Critt. No. 1 is a small precinct and the total vote there for assessor totaled only 130. By using the figure 335, the vote in the large precinct, as the vote cast for assessor in the small precinct, it became the highest vote for any office or measure in that precinct, whereas actually the highest vote in that precinct was that for Presidential Electors and was only 145. Thus it is seen that in this particular instance this error made a difference of 190 votes. If, however, the vote for assessor which was carried over to the wrong column was less than the actual highest vote for any office or measure, then the correct vote and not the vote for assessor would be used, so the only effect of this error was to increase the vote in the small precincts which were listed to the right of a large one. This was the result in the 13 precincts referred to by relators.
A similar error occurred in two precincts in Custer county, two in Garfield county, and two in Payne county, where the vote in one precinct was interchanged with the vote in another precinct, resulting in an increase in the vote for the highest office or measure in the smaller precinct in each instance, totaling 409 votes.
From this type of error alone there were 1,304 votes which were included in computing the total of 760,055 certified by the election board, which did not exist, and by deducting those votes from that figure, the total number of electors voting would be 758, 751. The favorable vote for the amendment being 379,405, it received a majority of the total number of electors voting. This error was not caused by miscounting, misadding, or misreading, but resulted merely from having the correct figures in the wrong column. There is no merit to the argument that there is no authority to change the vote recorded for any candidate, for no votes need be changed; they need only to be considered in their proper column.
Although the correction of this type of error alone is sufficient to change the result of the election, it may be of interest to point out briefly another type of error, which is equally apparent. Such errors occurred in posting the vote upon certain tally sheets, or work sheets, in the office of the county election boards, which tally sheets were used in computing the figure certified by the State Election Board under order of this court. These errors make a difference of 2,705 votes over the total vote for the highest office or measure actually cast and counted in those precincts, as shown by the precinct certificates on file with the State Election Board. By section 5809, O. S. 1931, these certificates are made prima facie correct. Aside from the ballots themselves, they are the best evidence for the obvious reason, if for no other, that the more hands through which the tabulations pass, the greater will be the probability of clerical errors. To illustrate: The total vote for justice of the peace, div. 1, dist. 4 in precinct 16, Tulsa county, appears *Page 633 upon the official certificate of that precinct to be 565. Whereas the vote shown on the tally sheet of the county election board is 865 for this office in this precinct. It appears thus:
Precinct Official Certificate Justice of the Peace, Div. 1, Dist. 4 W.L. Gray (Dem.) Three Eighty One (381) Fred A. Faust (Rep.) One Hundred Eighty four.
(Total_____565)
County Tally Sheet Justice of the Peace, Div. 1, Dist. 4 ( ( W.L. Gray (Dem.) ( 381 ( Fred A. Faust (Rep.) ( 484 ( (Total_____865)
The total of 865, was used by the State Election Board as the vote for highest office or measure in this precinct, whereas actually the vote for Presidential Electors totaling 658 was the highest vote. Thus it is seen that this error made a difference of 207 votes in computing the total number of electors voting in that precinct.
A similar error occurred in Seminole county, Econtucka precinct 4, where the precinct certificate showed a total of 389 for the office of Representative and the county tally sheets showed 789 for the same office in that precinct. The actual vote for the highest office or measure we find to be 474, which was the vote for Presidential Electors. This error made a difference of 315 votes. There are numerous other errors of this same type, which, taken with these illustrations, reduce the total vote by 2,705, but there is no need for further discussion in this regard. The relators assert that actually the total number of electors voting at said election as determined by highest vote in each precinct for any office or measure pursuant to the order of this court herein referred to is 756,455 votes. My examination of the record convinces me that they are, correct in this computation.
The State Election Board does not deny that these errors exist. In fact it has stipulated that they do exist. Therefore, the only remaining question is to determine by what procedure they can be compelled to certify the result of the election as it actually existed.
The proper remedy is the writ of mandamus, which can be issued by this court under its grant of superintending control and can be effectively used in this case to compel the correct determination of the will of the people in certifying the result of the election according to the dictates of our Constitution, without doing violence to any established principles inherent in the exercise of the writ.
First, the board must be compelled to use the method of determining the total number of electors voting at the election which can be said to be an interpretation of the constitutional requirement within the limits of legal construction. Although it is a general rule that mandamus will not issue to control the exercise of discretion, yet when there is an arbitrary abuse of discretion, the courts recognize that this is an exception to the general rule and mandamus may issue if there is no other adequate remedy, though the result is that the court is called upon to review the exercise of a discretionary power. 18 Rawle C. L. 126, sec. 39; 38 C. J. 608, sec. 85. In Board of Com'rs of Seminole County v. State ex rel. Cobb (1912)31 Okla. 196, 120 P. 913, the first syllabus of the court is as follows:
"Where there has been an exercise in good faith of judgment or discretion by an officer upon whom a duty involving discretion is imposed, the writ of mandamus will not lie to compel him to act again; but, if by a mistaken view of the law or by an arbitrary exercise of his authority there has been in fact no actual exercise in good faith of the judgment or discretion vested in the officer, the writ will lie to compel such officer to act within the limits of the law."
I am aware, of the rule that mandamus will not lie to correct an erroneous exercise of discretion, but in my opinion the action of the State Election Board in certifying to the Governor that the amendment was defeated for the reason that it did not receive a majority of the number of ballots issued, less spoiled, was an arbitrary action and a clear abuse of discretion. I have discussed the only three methods of construction which were available. The statutory provisions enacted to provide a satisfactory compliance with the constitutional requirement had not been complied with, so the board could not choose that method. Because of that fact, that course Was not open, and the board was compelled to certify the result of the election by using the vote for Presidential Electors, or by using the total number of ballots issued, less ballots spoiled. The method which they did choose is an interpretation which is specifically rejected in other states and which considers as votes *Page 634 of electors blank and mutilated ballots, which under the decision in Board of Education of Oklahoma City v. Woodworth, supra, are not to be so considered. Such an interpretation is not within the limits which the law will permit to be applied to this constitutional amendment. Moreover, the record discloses an opinion by the Attorney General advising that "a constitutional amendment proposed by the Legislature and voted upon at a regular election, is required to receive a favorable vote equal to a majority of the total number of votes actually cast and counted on an office or measure for which all the electors of the state were entitled to vote and for which the greatest number of votes were actually east and counted at said election, in order to carry." It appears from the certificate of the State Election Board transmitting to the Governor the result of this election that a copy of the Attorney General's opinion was filed with the board, and there further appears in the certificate a letter from the Governor quoting from this opinion and requesting that the board certify to him the number of votes so cast and counted, together with the votes on the amendment, in order that he may declare the result of the election as required by law. The certificate then recited: "That the following is the greatest number of votes which were actually east and counted at said November 3, 1936, regular general election, on an office or measure for which all the electors of the state were entitled to vote, to wit, 749, 740, same being the total vote cast and counted at said election for Presidential Electors. We certify that 767, 745 is the number of electors voting at the November 3, 1936, regular general election, which represents a grand total for the state of the number of ballots issued in each precinct less those ballots spoiled, in the respective precincts, and is interpreted by the State Election Board as the number of electors voting at such election, and the number that must be taken into consideration in determining whether the measure lost or carried."
Thus I am of the opinion that the action of the board was arbitrary and was a clear abuse of discretion, and was in effect no exercise of discretion within the limits of the law. It was within the province of the board as between the two methods to determine the total number of electors voting only by the highest vote cast and counted for any office or measure for which all electors of the state were entitled to vote. In this case it was Presidential Electors. Mandamus should therefore lie to compel the board to certify the result of the election on that basis, and under this view the amendment carried.
On the other hand, if the method required by the order of May 26, 1937, is used as the test, the board must be compelled to ascertain the correct number of electors voting in the 81 precincts where the clerical errors occurred, and certify the result of the election as it actually existed. This can clearly be done by writ of mandamus. We have demonstrated that the errors occurred merely in the process of tabulation. This is purely a ministerial duty and requires no discretion whatsoever 20 C. J. 200; Stelarns v. State (1909) 23 Okla. 462,100 P. 909; and if the vote is correctly tabulated, we find from the evidence herein that the amendment carried.
Probably we cannot, however, direct the discretion of the board in choosing one of these two methods, but mandamus will lie to compel them to proceed upon either one or the other. The number of votes for Presidential Electors has already been ascertained, and the board can be compelled to ascertain the correct number representing the highest vote cast for any office or measure in each precinct. It is immaterial which course they take, for under either method the amendment was adopted. It is obvious from the record before us that the will of the majority has been defeated by the arbitrary action of the election board and because of mere clerical errors in tabulating the votes, and it is the duty of this court to exercise every power and remedy at its command to rectify such an error. This court has no higher duty than to determine whether a proposed amendment has been engrafted upon the Constitution, since it deals with the sovereign rights of the people. State ex rel. Short v. Board of Equalintion, supra; State ex rel. Postel v. Alarcus. (Wis.) 152 N.W. 419. 1 cannot understand how this court can say as it does in the majority opinion "it appears that relators have not shown that the proposed constitutional amendment received a vote equal to 'a majority of all the electors voting at such election.' " To refrain from issuing the appropriate writ in this case is a regrettable neglect of our clear duty. *Page 635