HURIN,-J.
This action was brought originally in the probate court to contest a local option election in Putnam county, which county, on the face of the returns, was declared to have voted “dry” om October 13, 1908.
[516]*516Soon after filing the petition to contest the election a motion was also filed by the contestors asking the probate judge to certify the case over to the court of common pleas on the ground that he was prejudiced against the side of the contestors as shown by his activity in the campaign preceding the election. The motion being overruled by the probate court, an action in mandamus was commenced in the circuit court seeking to compel the probate judge to so certify the case.
A majority of this court, believing that such an action would lie, granted such writ, and while exceptions were taken to this action and a proceeding in error was prosecuted to the Supreme Court, which resulted in that court’s denial of the right to compel, by mandamus, a probate court to certify to the common pleas court a case in which it claimed jurisdiction, yet while such proceedings were pending in the Supreme Court, the probate judge obeyed the mandate of this court, and having done so, it was held by the Supreme Court, no opinion, Heidlebaugh v. Recker, 81 Ohio St. 514, that he had waived the right to rely upon his right of jurisdiction, and the Supreme Court therefore refused to take, further action in the case.
In the meantime the case had been tried on its merits in the •court of common pleas. That court found on the evidence that the results of the election in the south precinct of the village of Ottawa had been erroneously certified to the board of elections and that the vote had been improperly tabulated by the clerks of election; in this respect, that the number of voters voting for the prohibition of the sale of intoxicating liquors had been recorded on the line designated for those voting against such prohibition and that those voting against such prohibition had been recorded on the line designated for those voting for such prohibition, so that, in that precinct, the vote which had been recorded as 119 against prohibition, or “wet,” and 196 for prohibition, or “dry,” as it is commonly •called, should have been recorded as 196 “wet”’ and 119 “dry” •and that, as a consequence of that error, the vote of the whole •county was changed, and instead of there being a majority of 21 votes in the county for the prohibition of the sale of intoxi[517]*517eating liquors, there was, in reality, a majority of 133 votes against such prohibition.
In a very able, learned and painstaking opinion the judge of the court of common pleas found that such was the actual result of such vote as shown by the evidence and that a tri^l court on the trial of a contest of an election could and should go behind the first returns and, having found the results of the election to have been improperly recorded, should set aside suck return and cause a correct return to be recorded.
From this decision error is prosecuted to this court.
It is evident that there are but two important questions its this case and that we must first inquire whether, on a contest of an election, the trial court had the power to go behind 'the face of the returns and inquire into the facts, and, second, if we find that the court had such power, we must inquire whethei” the evidence in this ease justified the conclusion reached by the trial court.
The law (99 O. L. 35; Gen. Code 6108 et seq.) providing fora county local option election also provides for a contest of suck election.
By Sec. 9 of act 99 O. L. 38, the so-called Bose law, it is-provided that any qualified voter of the county * * * may-contest the validity of such election by filing a petition in the probate court within ten days after the election, setting forth, the grounds of contest.
Gen. Code 6118 further provides thaf “the probate court shall have final jurisdiction to hear and determine the merits of the proceedings, and in other respects, in the procedure of the: ease, shall be governed by the law providing for contesting the; election of a justice of the peace so far as such law is applicable. ’ ~
By referring to the law applicable to the contesting of an election of a justice of the peace we find that, among other-things, it provides, Gen. Code 5162, for a notification of the probate judge by the contestor specifying the points on whick the contest shall be based.
It provides further, Gen. Code 5163, for the summoning witnesses.
[518]*518■ It provides, further, Gen. Code 5166, that no election shall be set aside because illegal votes were cast at the election, if it appears that the person whose election is contested has the greatest number of legal votes given at such election.
These provisions point to contests on various grounds which, as is also required by the Bose law, shall be specified in the notification of contest, and they also specifically refer to illegal votes as a ground of contest — a question not raised in this case.
Some light may be obtained by way of analogy from other similar statutes relating to other contests of election.
Gen. Code 5152, which relates to the contest of an election mf a county officer provides that ‘ ‘ On the trial, either party may introduce oral testimony or depositions of witnesses taken as provided in civil actions. When any omission, defect or error occurs in the proceedings of an officer in declaring or certifying that a person was duly elected to-an office, it may be corrected by oral or other testimony, offered at the hearing of any preliminary proceeding or at the trial, ’ ’ while Gen. Code 5148-5151 give specific authority for the taking of testimony and the compelling of the production of books, papers, etc.
And Gen. Code 5157, 5158, which relate to the contesting of the validity of a vote on the removal of a county seat provides for a commissioner to take testimony in writing as to the validity of the votes cast at such election upon such question and as to the validity of the result thereof.
Provision is made for a review of the commissioner’s report by the judge of the court of common pleas and by Gen. Code 5160, it is provided that “If, upon the hearing, the court or judge finds that illegal votes were cast at the election upon such question by reason whereof or for any other reason found by the court or judge, the result of the election or vote so returned :a.nd certified is contrary to what it would have been but for such illegal votes or other reason, the court or judge shall enter and certify such findings on the records of the court.”
By these last three sections it appears that in other contests «of elections the legislature has clearly provided for going behind [519]*519the face of the returns in order to arrive at an understanding of the facts.
The section last cited provides for setting aside the result of an election either because illegal votes were cast or for any other reason; and Gen. Code 5152 provides that “any omission, defect or error occurs in the proceedings of an officer in declaring or certifying that a person was duly elected to an office, it may be corrected by oral or other testimony, offered at the hearing of any preliminary proceeding or at the trial.”
It thus clearly appears that, in similar contests, errors in certifying returns may be inquired into by oral testimony.
By analogy, then, the same rule would seem to apply to contests of local option elections, although the specific statute providing for such contest does not in terms so declare.
The constitution of the state, Art. 2, Sec. 21, has provided that ‘ ‘ the general assembly shall determine, by law, before what authority, and in what manner, the trial of contested elections shall be conducted. ’ ’
This the general assembly has done in the statutes previously quoted. '
It is apparent at once that there is a sharp distinction between the powers and duties of the board of deputy supervisors of elections in declaring the results of the election and the powers and duties of a court in considering a contest of election. The board of deputy supervisors has ministerial powers only. The court has judicial powers. The board of deputy supervisors is without authority to hear evidence to contradict or explain the tally sheets: while a court, on a contest of the election, is clothed with such authority. See State v. Tanzey, 49 Ohio St. 656 [32 N. E. Rep. 750], and this distinction is clearly recognized by Gen. Code 5110 which specifically provides that “in making the abstracts of votes, the board shall not decide on the validity of the returns, but it shall be governed by the number of votes stated in the returns.”
The Supreme Court of Ohio in the famous case of Dalton v. State, 43 Ohio St. 652 [3 N. E. Rep. 685], a case where there seemed to be no doubt that the most infamous frauds had been [520]*520practiced in Cincinnati, held that the duties of canvassers of election returns under R. S. 2981 now Gen. Code 5110, were ministerial merely and that they had no power to decide such returns invalid by reason of fraud at the election or in the returns thereof as made to the clerk, and that the court could not by mandamus compel such canvassers to omit certain votes from the abstracts and that the court could not in the mandamus proceeding, decide on the validity of such returns but that the remedy in such case was by contest as provided by law.
The statute there under consideration expressly forbade the canvassers to decide on the validity of the returns and the court held that in a mandamus proceeding it could not compel the canvassers to perform a judicial duty when the law made them ministerial officers only.
In a later case growing out of the same election, Dalton, Ex parte, 44 Ohio St. 142, 156 [5 N. E. Rep. 136; 58 Am. St. Rep. 800], the court again emphasized the fact that in a contest of election the facts could be thoroughly inquired into, and, referring to the contest then in progress before the house of representatives over the election of some of its members, the court repeats this strong assertion:
“In a contest in either house the broadest range is given contestants to purge the ballot and returns of the consequences of neglect, mistake, fraud, and crime, from the opening of the polls to the final declaration of the result.”
And in the contested election case of Howard v. Shields, 16 Ohio St. 184, Judge Welch in the opinion, page 189, says:
“The policy of the law seems to be, that until the contrary is: shown, the tally sheet shall be taken and considered as a true' statement of the number of legal votes cast for each candidate. Of course, it is open to be impeached by the other party. ’ ’
And further on in the same opinion the court said:
“The question is, can the court, in trying a contested election, go behind the poll books and tally sheets, to supply and correct mere omissions and mistakes in them, by parol evidence ? We have no hesitation, either upon principle or authority, in answering the question in the affirmative. * * * We [521]*521apprehend the true rule to be that both the abstract and the poll boobs and tally sheets when' substantially correct upon their face, are prima facie sufficient, but may be impeached -by evidence aliunde, showing their falsity or sufficiency; and that when not so substantially correct upon their face, they may be sustained in the same way. To hold that, when an election has been in fact held, and the majority of the legal voters have in fact, and according to the prescribed forms of law, cast their ballots for the candidates of their choice, the constitutional rights of the voters and of their candidates can be defeated by a mere misprision or omission of the judges or clerks, would be manifestly unjust and contrary to the plain intent and spirit of our election laws.”
And in the case of Ingerson v. Berry, 14 Ohio St. 315, the Supreme Court, per Scott, J., declares that the court of common pleas is “clothed with full * * * power to judge of the validity of the returns as shown by the poll books, and to go behind them and inquire into the legality of every vote which they exhibit.”
The case of Phelps v. Schroder, 26 Ohio St. 549, is to the same effect.
There would seem, then, to be no room for doubt that the statutes of Ohio, relating to contests of election generally, passed in pursuance of the authority vested in the general assembly by the constitution of Ohio, and as interpreted by our Supreme Court, do give to courts, in contested election cases, the power to go behind the face of the returns and inquire into any neglect, mistake, fraud, or crime, which has interfered with the true expression of the will of the people and that, in such inquiry, oral testimony may be received to support either side of the contest.
But in the case at bar the contestors rely in part upon the testimony of the judges and clerks of the election in the south precinct of the village of Ottawa to prove that their own certificate as to the results of the election was erroneous — and that therefore the result of the election was the reverse of what they had officially certified it to be.
In other words, these contestors seek to have these judges [522]*522and clerks of elections impeach their own certificates by their oral testimony. Can this be done ?
We' have little hesitancy in saying that, as a general rule and as applied to officers generally, this cannot be done. The authorities are overwhelmingly against their power to do so.
In the early case of Hill v. Kling, 4 Ohio 136, the court held that “the sheriff cannot be permitted, either in pleading or by evidence, to falsify his return,” citing Gardner v. Hosmer, 6 Mass. 325, and Purrington v. Loring, 7 Mass. 388.
The case of State v. Moffitt, 5 Ohio 358, while not strietly analogous, decided that the record of an election by the legislature where it was claimed that by mistake the record showed that Samuel Moffitt had been elected judge whereas it was in fact Lemuel Moffitt who was elected, could not be corrected by parol, because members of the legislature could not be heard to prove that the action of that legislature had really been, as against the actual record of that action.
The case of Sinks v. Reese, 19 Ohio St. 306, was a contested election ease wherein the tally sheet, poll book and ballots might have been offered in evidence but were not so offered, and oral evidence was offered to prove that an error was made in the count.
The court held that this could not be done because of the rule requiring the best evidence to be produced vffiere it is possible to produce it.
The case of Taylor v. Wallace, 31 Ohio St. 151, was a case involving the election of state and county officers.
The court held that the declaration of the clerk and justices showing who were duly elected ought to be certified in writing and the certificate ought also to show the day on which the declaration was made in order to fix the time for taking an appeal.
The court then held that “if, taking the certificate in connection with the returns and the abstract, there is no ambiguity or uncertainty as to the date of the declaration, parol evidence, in-the absence of fraud, is inadmissible for the purpose of fixing the time for taking an appeal.”
[523]*523In a ease decided! by the court of appeals of Kentucky it was held that:
“An officer of election will not be permitted to contradict his solemn certificate to the returns signed by him at the close of the election by parol testimony in an election contest.’’ Browning v. Lovett, 29 Ky. L. 692. [94 S. W. Rep. 661].
But the court in that case cites an earlier Kentucky case where the parol evidence of such an election officer was admitted to explain how an error in the additions had been made on the poll books which were in evidence.
In a few eases, such as Hamilton v. Young, 26 Ky. L. 447 [81 S. W. Rep. 682], we find that the courts of other states have permitted officers of elections to testify orally as to the circumstances attending the préparations of their certificates — so that it appears that this exaction question has arisen in very few cases and the courts are not at all uniform in holding for or against the admissibility of the parol testimony of an officer of election, impeaching his own certificate. The rulings appear to be governed more by the circumstances of each case than by any clearly defined principle of law; and in most of the cases where such parol evidence was admitted, it appears to have been admitted to explain discrepancies in the certificates and returns or to explain errors alleged to exist in them, their admission being based on claims of fraud or mistake.
With the question thus undecided by the courts of our state and with no definite rule to be derived from the adjudicated cases in the courts of other states, we must resort to the general principles of law in the decision of this case.
It may be said to be well established that the certificates and returns of election officers are prima facie evidence of the fact certified to.
Cooley, Const. Lim. p. 940; Ewing v. Thompson, 43 Pa. St. 372; Hartman v. Young, 17 Ore. 150 [20 Pac. Rep. 17; 2 L. R. A. 596; 11 Am. St. Rep. 787]; 2 Elliott, Evidence Sec. 1294, and many other authorities so declare.
This is equally true of the tally sheets. Howard v. Shields, 16 Ohio St. 184, 189; State v. Donnewirth, 21 Ohio St. 216.
[524]*524In the ease at bar the tally sheets, certified to by the officers of election, show one state of facts — the five other certificates of the same officers of election show exactly the contrary state of facts. Both are, as we have seen, prima facie evidence of the facts therein recited.
But neither the tally sheets nor the certificates are conclusive. Both are signed and certified as correct by exactly the same officers. If either is correct, the other must be wrong. It is of the greatest importance that the question as to which is more probably right shall be settled. In this dilemma the court must look to whatever evidence is available to aid it in its determination.
"Where the same men have .signed two contradictory certificates, one of which must be wrong, sound sense would seem to demand that they should be permitted and required to explain the discrepancy. This court is interested not so much in what the certificates say as in what the people actually did at the polls.
Courts of laws and courts of equity both are compelled to allow a wide latitude where questions of mistake or fraud are involved and there is a large class of cases which hold that in cases of fraud and mistake a certificate, otherwise binding, may be contradicted by parol, and even by the oral testimony of the person making the certificate, where such certificate appears to be clearly a mistake or fraudulent. See Elliott, Evidence Sec. 594; Young v. Duvall, 109 U. S. 573 [3 Sup. Ct. Rep. 414; 27 L. Ed. 573].
In view of these principles we have read and-reread, with the greatest care, the evidence offered in this case — looking first to-evidence other than that of the election officers and finally to that, of the officers themselves.
"What actually took place at the voting place of the south precinct of Ottawa, on October 13] 1908, can only be known to-eight persons — four .judges, two clerks and two inspectors.
The judges and clerks all signed two absolutely con.tradictory sets of certificates, two of them attached to the tally sheets, and five of them purporting to state the facts disclosed [525]*525by those tally sheets. The tally sheets themselves are, of course, the best evidence, for they are the record of the votes as originally called off, and noted before the final result could possibly be known. They were kept by the clerks, and both show that 196 votes were east for the prohibition of the sale of intoxicating liquor's and 119 votes were cast against such prohibition. Immediately after these were signed five other certificates were filled out and distributed as provided by law, though it appears that these had been somewhat irregularly signed before being filled out. These five other certificates all agree in stating the result as exactly contrary to the tally sheets.
Of the eight witnesses to the events of that evening only two — the inspectors — did not sign the certificates. Of these two men A. M. Brown represented the dry side and Andrew Brink-man represented the wet.
Mr. Brinkman testifies that he was present at the counting of the votes, watched the count, saw the record made and knows that the ballots were then burned; that, after the count, the number of votes was called off and that 196 votes were called off as “wet”; that a certificate of the result was made out after the count was over and given to him, as inspector — being certificate known as exhibit “I” and that it showed then, as it does now, that there were 196 wet votes and 119 dry votes. He says, that in counting the votes, there were first 7 bunches of 5. wet votes each before there were any drys; that they put the wet votes in the dry column, and afterwards put the dry votes on the line below.
Mr. A. M. Brown testified that he was an inspector for the dry side, was present during all the time of the court, but did not see how the votes were being tallied by the clerks and did not .see any inspection, either by any of the judges or by the other inspector of the tally sheets as they were made by the clerks, and that he made none himself; that all in the room were busy with the particular work assigned to each. He says that he, too, received a certificate being exhibit “ J, ” which exhibit is in exactly the same form as the other four certificates in evidence; that is, it .shows 196 “wet” votes and 119 “dry” votes. Strange to say, [526]*526neither the attorneys for the contestors nor for the contestees: asked Mr. Brown whether, as a fact, the wet votes were recorded on the dry line and the dry votes on the wet line, nor did they ask him whether he knew on the evening of October 13 that there was a difference between the two kinds of certificates.
Later on Mr. Brown was again called as a witness and testifies that on the morning of the election one of the judges was absent and his place was filled by election and that Mr. Bennett received all the votes for that office, but one, which was cast by Mr. Brown himself. An attempt was made, and we think was improperly prevented by the court to show that the complexion of the board and that of the crowd that elected Mr. Bennett was plainly “wet.” That the crowd was composed in part at least of saloon keepers and bartenders and that Mr. Brown was the only “dry” man present. But whether this was improperly excluded or not'is now immaterial. Of all those present at the counting of the ballots on the night of the election, Mr. Brown was the one most interested in seeing that the vote of the ‘ ‘ drys, ’ * if in fact the larger, was properly recorded. He is unable to tell anything about the way the record was made, or whether the votes were recorded on the proper lines of the tally sheet or on the wrong lines. He, like the other inspector, received a certificate showing that the “wet” side had received 196 votes to 119 for the “dry.” His attention was thus immediately challenged to the announced result. He made no protest, made no claim that this was an error and appears to have been perfectly satisfied with this statement of the result. If it was wrong, it was his plain duty to see that it was at once corrected. That is what he was there for. His silence carries with it an almost overwhelming, presumption that the result was correctly announced, that he understood it, and that he was satisfied with that announcement.
As to what the tally books actually showed he appears to have been absolutely ignorant. He was not asked by either side to state from his own knowledge what the vote actually was. He had not watched the tally made by the clerks, being engaged in watching the ballots as they were called off, and he was not [527]*527asked to testify as to whether the statement of his fellow inspector was true, that is, that the wet votes were first recorded on the line .reserved for the dry. If there were no other evidence than the testimony of these two inspectors, neither of whom could possibly be held to be an incompetent witness, for neither of them had signed the certificates or poll books, the court of common pleas would probably have been justified in deciding as it did. But if we give any credence at all to the testimony of the judges and clerks, the proof is overwhelming that the five certificates were right and that the tallies were erroneously recorded on the tally sheets. Every one of these six officers testifies. While there are some discrepancies in the details of their testimony, they substantially agree on all of the important facts. Four of these six officers swear that the first 35 votes counted, which were apparently the first votes-east, were “wet” votes — thus agreeing with the testimony which was sought to be brought out on the other side, and was excluded to the effect that all of those who organized the polls in the morning were “wet” sympathizers. The other two officers were not asked this question. Four of these six officers testify, the others not being asked the question, that these first 35 wet votes were recorded on the first line of the tally sheet where the dry votes should have been recorded, and that from that time on, all the wet votes were recorded on that line and that the totals were always about the same number ahead of the total dry votes.
All but one of these six judges and clerks, the other not being asked the question, swear that at the conclusion of the count, public announcement was made at the door of the voting place and that in this announcement the vote was publicly declared to be 196 wet and 119 dry. Five of these, six'witnesses,- the other not being asked as to this, declare that a certificate stating this as the result was immediately tacked on the outside of the door of the voting place and that certificate is identified and attached to the bill of exceptions as exhibit “F,” and it like all the other certificates (except those on the tally.sheets), shows 196 wet votes and 119 dry votes. All of these officers of election testify that in fact there were 196 wet votes cast and- only 119 dry votes.
[528]*528The testimony of these six officials is thus overwhelmingly in support of that of Mr. Brinkman, the wet supervisor, to the effect- that the wet vote was actually recorded on the line intended for the dry vote, and that the dry vote was recorded on the line intended for the wet vote. It further substantiates the statement which neither Mr. Brown, nor any other witness, denies, that the result of the vote was understood by all present to be and was certified to be and was publicly announced to be 196 wet and 119 dry, and that Mr. Brown, as well as all others entitled to them, were that night given certificates showing this result; and that no protest was made and no one questioned the truth of this certificate. The clearest explanation of the whole matter is given by Mr. Crawfis, one of the clerks. He testifies as follows,, pages 88 and 89, and I quote his testimony on this point in full. He says:
‘ ‘ In the early part- of the day I looked over the tally book as usual, investigating the form of it, read what was in it and paid po more attention to the printing on the left hand side. When the votes were counted out there was no one called fortihe two clerks; we were instructed not to tally any votes till called off by a certain judge who was to call them off in batches of five at a time; and there was 7 batches of wet votes called off before there was a.batch of dry votes; for some reason I know not what ,it is, any more than we can explain any other-error that we make day by day, I started in on the top line putting the first ones down there and of course continued to do that until I had 7 batches of fives; that thoroughly fixed in my mind which I was going to use for the wets without referring whatever to the printing on the left; and then after the 7 batches had been called off, one batch of drys, making 35 at the top and five, at the bottom; from that time on till after the top line had gone above 100 there remained just 30 between the two; the other clerk and myself remarked several times during the tally the peculiar coincidence of the remaining just 5 apart after this 35 wet vote and the 5 dry and then there was 5 wet votes and then 5.drys and I went from the top to the bottom and the top- to the bottom without any reference whatever to the printing on the left. ’ ’
[529]*529The other clerk, Roy Deck, merely says that he tallied the wet votes on the upper line and the dry votes on the lower line. Mr. Crawfis says that they called off their totals to each other and they agreed. This explanation is a simple one, and not unnatural. It is easy to believe that such a mistake could have occurred.
The hard thing to understand is how it could possibly remain undiscovered that eight presumably intelligent men could stand by all evening and see such a mistake carried out, and then in addition, see those figures erroneously copied into the statement at the bottom of each tally sheet — that six of these men could deliberately sign those erroneous statements which were directly contrary to their understanding of the facts, seems almost incredible. Yet they all say that they did it and Mr. Brown, who was certainly interested in proclaiming the truth, 'if these two statements were true, accepted in silence a certificate directly contradicting them in the vefy vital matter of the whole day’s work, heard that new certificate, which is now called false, proclaimed in public as the truth, and saw it tacked to the door to further proclaim that truth, and all without a word of protest. Apparently all of these men relied entirely on the accuracy of the two clerks, and must now ask us to believe that they signed the clerk’s report without reading it. If. the clerks and judges were careless, the inspectors Avere almost equally careless.
If the clerks and judges were deliberately falsifying the returns, as we are asked by the other side to believe, then they at least gave due notice of their design'to,the very man who was there to prevent such trickery, and he took no action to prevent it. We are loath to believe that these men would deliberately perpetrate such a fraud against the public, or that they Avould presume to carry out such a fraud by perjury on the Avitness stand. But whether avc could believe it or not the evidence does not prove it. That they were grossly careless — astonishingly careless — is the only other alternative. The evidence, if believable to all, admits of no other conclusion. It is clear, explicit and without any contradiction. Such carelessness is more than [530]*530unfortunate. It has been expensive, not only in money but in the reputations of all who participated in the transaction. It has kept the whole community in turmoil and has aroused suspicion and bitterness among citizens who ought to be on terms of cordial friendship. Such things are not good for a community. It is to be hoped that this unfortunate strife will now cease and that what cannot be cured will be forgotten.
The judgment of the court of common pleas will be affirmed.
Donnelly and Kinder,' JJ., concur.