Under the first error assigned, it has been con tended that the referees had no power to proceed to make a second report after the recommitment, because their authority had been revoked by the plaintiifs in error. It was resolved in. Vinyor’s case, cited in the argument, that an authoiity counter-maudable by the law, cannot in any way be made irrevocable. Hence, it was there decided that if one becomes bound to abide the award of an arbitrator agreed upon, and afterwards revokes the submission, such revocation is good, although the bond is forfeited. And this principle has been recognised in subsequent cases. But in Milne v. Gratrix, cited from 7 East 608, Lord Ellenborough says, after the submission is made a rule of court, the party cannot rescind it, without incurring a breach of that rule. It would seem therefore from this authority, that a submission once made a rule of court is no longer countermandable by the law ; the party attempting to countermand it being liable to an attachment for a contempt ; which is the coercive process by which rules of court are enforced in England, And in Haskell v. Whitney 12 Mass. 47, it was decided that where an actio» has been referred by a rule of court, neither party has a right, without the consent of the other, to rescind or discharge it.
The authority of the court to recommit generally, without consent of parties is controverted ; it being urged that their *464power to recommit can only be exercised for the purpose of correcting mistakes in matters of form. In support of this position, cases have been referred to in the English books, where courts have refused to sustain alterations made in awards by arbitrators, after the time limited for the exercise of their authority had ceased, or after they had once executed the powers delegated to them. Of this kind also is the case of Woodbury v. Northey. But all these were cases of submission to arbitrators out of court, in some of which they were expressly restricted as to time, and in others their power was held to be at an end, after it had been once exercised. But recommitments of reports made under a rule of court, or under a submission before a justice, in regard to which the common pleas has, by statute, the same power as it has over its own rules, have been uniformly made, both in this court and in the common pleas, whenever, in the opinion of the court, the purposes of justice required such a course. Nor has this practice been confined to the amendment of mere matters of form, but has extended to the substantial merits of the matters in controversy, whenever a reexamination of the whole subject has been deemed expedient. And nothing is more common than an award of referees, after re-commitment, presenting results differing materially from their first report. Where the court, from any cause, not arising from the misconduct of the referees, deems it improper to accept the report first made, it is generally much more convenient to the parties to recommit it for revision, than to discharge the rule. The power of the courts to do this is fully recognized in the cases of Whitney v. Cook and Boardman v. England, cited by the counsel for the defendants in error.
One of the errors assigned is, that it does not appear for what reason the report was recommitted. It is not usual, nor is it necessary, to spread upon <he record the reasons which induce a recommitment. Whether the report shall, or shall not, be thus disposed of, depends upon the sound discretion of the judge 5 whose determination upon this point is conclusive.
From the assignment of errors, and from the record before us it appears that Nathan Elden, the chairman of the referees, did *465not act with his colleagues after the recommitment; he having re* fused tonotify a meeting, or to meet at the time and place by them appointed. Aud a majority of the court are constrained to determine that this error is well assigned ; and that the judgment must be reversed. The sum awarded against the plaintiff's in error, in the second report, signed by two of the referees, being nearly double the amount awarded against them in the first report, signed by them all, evinces that the subject matter was re-examined, and that it resulted in a great change of opinion, on the part of those who signed the second report. The authority of referees to adjudicate between the parties, can originate only from their assent. The assent of the plaintiffs in error to this submission, may have depended upon the confidence they reposed in the judgment and integrity of the chairman of the referees. Unless he was appointed, and consented,to act, they might have declined the reference altogether. To oblige them therefore to abide the award of the two other referees, made and concluded at a meeting at which he was not present, and without the benefit of his assistance and advice, even though he might refuse to act, which might happen without the fault of the plaintiffs in error, would be to subject them to the determination of a tribunal differently constituted from that to which they had submitted. It is true they had consented to be bound by the report of a major part of the referees; but that must be intended to mean upon a final difference of opinion, after a hearing by all, and after availing themselves of the aid, which each could afford, in the consideration and discussion of the merits. That the views and arguments of one may justly and fairly have an important influence upon the opinion and judgment of others, is a fact which will not be controverted. If therefore two have come to a certain result, 'without the assistance of the third, it by no means follows that they would have come to the same result, if they could itave had the benefit of his advice. It would not probably have been contended that, upon a reference to three, the original award of two, ihe third not having been present at the hearing, could have been binding upon the parties. A majority of the court are unable to perceive any difference in principle between such an award and w» made by two, in the abscuco aud without the assistance of *466the third, after a recommitment, substantially varying from the first award, to which all had assented.
After the referees had once undertaken the execution of the trust confided to them, if they, or any one of them, should refuse to re-examine the subject, the court might enforce obedience to their order of recommitment, by mandamus or attachment. It is believed however that no precedent can be found of a resort to such a process, in a case of this kind. It would without doubt be deemed a more eligible course to discharge the rule, and leave the parties to the ordinary modes of trial at law. In Boardman v. England, Parsons C. J. in stating the practice of the courts in regard to rules of reference, says, if either of the referees refuse to execute the trust, the rule is discharged. In Short v. Pratt 6 Mass. 496, it was decided that upon the recom-mitment of a report, it must appear that all the referees heard the parties ; although if théy disagree, the award of two is binding. That was the case of a report made to the common pleas, upon a submission before a justice ; but as reports of that kind are by statute to be treated precisely like reports made under a rule of court, it is an authority directly in point.
There is nothing in the case of May v. Haven or of Peterson v. Loring, referred to in the argument, at variance with the authority last cited. The reports originally made in these cases, had been signed by all the referees. After this recommitment, two of them, in the absence of the third, in each case, made a report conforming to the first. As the same results had been assented to by all, they were deemed to have been substantially made by three, and not by the two only, who had last signed. Peterson v. Loring was decided upon the authority of May v. Haven, and in the opinion of the court in the former case, both are declared to be consistent with the case of Short v. Pratt. In Walker v. Melcher, the referees met and heard the parties on the thirtieth of November, which they set forth under that date ; on the same day one of them,certified that he was present at the hearing, hut gave no opinion as to the damages ; and on the seventh of March following, the other two referees made up and signed their report; the third not being present. One of the errors assigned *467was, that the assessment of damages was made on the seventh of March ; and that it did not appear that the referee, who did not subscribe, had any notice of the meeting on that day, or of their assessment of damages. Upon this point, the court say that the referee, who did not join in the report, appeared to have been present at the heaving ; but did not agree with his brethren, as to the amount of the damages. If he did not agree in the amount, he must have known what it was. And it is apparent, from the opinion of the court, that it was founded upon the assumption, that every thing was virtually and substantially agreed by the two in November ; the third dissenting as to the amount of damages ; and that the result, to which they had then arrived, was put into form and signed by the two in March. The court, did not overrule this error ; but denied the fact upon which it was predicated.
Mellen C. J.After the most serious consideration of this cause, in all its bearings, I have not been able to agree with my learned brothers in the opinion which has just been delivered ; and though I regret that an ultimate difference should exist, still I must pursue that course which my judgment and sense of duty prescribe. Though my own opinion will not affect the decision of the court, yet, injustice to myself, I deem it proper to state distinctly the reasons and principles on which it is founded ; observing at the same time that I agree with the other members of the court in their opinion, so far as it overrules several of the errors assigned, and objections urged by the plaintiffs’ counsel.
I consider the law to be well understood and settled, that when one referee refuses to accept his appointment, the others have no authority to proceed.
When he has accepted and entered on the duties of his appointment, he must be considered as retaining his authority, until he shall refuse to proceed any further; and give notice to the court under whose commission he has been acting, or at least to both the parties in the cause.
As nothing of this kind has been done in the case we are examining, FMen’s powers as a referee were continuing at the *468time the last report was made ; and, of course, the powers of the other referees also.
There is then no objection to the report by reason of a determination of the authority of any of the referees.
Is it objectionable on any other ground ?
The power given to referees to hear and try a cause, is a joint pdwer. — All agree in this.
The power given them to decide is not a joint one ; because the decision of a majority is legal and sufficient.
If, after a joint hearing of the parties and their proofs, in the first instance,and before a recommitment, one of the referees absent himself immediately, or refuse to consult with his brethren, or to give any opinion, then the other two have full power to decide the cause upon the evidence previously produced, and heard by all.
After such a hearing, each referee has an unquestionable right to express and continue to hold his own opinion ; and neither of the other two, while continuing in office, can by any act of his, defeat, impair or control this right.
So after a report'has been recommitted, this right continues; and if no further evidence be offered, or hearing of the parties had, a majority has the same authority to decide the cause on its original facts and merits, as they had when the first report was made; because the order of recommitment does not take away any of the original powers of the referees ; but only authorises them to reexamine the cause upon the former facts, or upon additional facts also, as circumstances, or the wishes of the parties, may render proper or require.
If no additional evidence be offered, or hearing of the parties had by the referees, then the only power exercised by them under the order of recommitment, is that of reviewing, and, if thought proper, of correcting their former opinion and report.
In doing this, each referee may and must judge and decide for himself. The operations of the minds of the referees, are not and oannot be joint; they may reason and arrive at their conclusion separately, as well as together ; and when any tivo of them agree in thejr conclusions, they may, after due notice to the third referee to join them in making a report, decide the cause. without or against his opinion.
*469Hence his absence, after due notice, cannot be of more importance than his presence and express dissent from the opinion of the. majority ; and, most certainly, that cannot affect the validity of the report.
I am not able to discover any fallacy or unsoundness in cither df the foregoing propositions, or incorrectness in the results, as I have stated them.
It is not denied that, upon these principles, the majority of the referees may, after a recommitment, report the same sum, which was originally reported ; such was the decision in May v. Haven 9 Mass. 325, and Peterson v. Loring 1 Greenl. 64; but it is denied by the plaintiffs in error that, in such circumstances, two of the referees can legally report a larger sum in damages than the sum named in the first report. It appears, in both the above mentioned cases, that the majority of the referees, in the absence of the third, made a report, however, by which, though they did not increase the damages, they allowed additional costs; but both reports,notwithstanding this increase, were sanctioned and accepted. It is not easy, at least for me, to perceive why two of the referees, in the circumstances mentioned, had not as good a right to increase the amount recovered, in the form of damages, as in the form of costs ; the latter are as much a part of the report as the former. In both cases new liabilities and additional obligations are created by the second report. It has been said, by way of reply to this suggestion, that costs are only a consequence or incident ; but, though in some cases they are, in case of a decision by referees, they are not so. The allowance and recovery of cost arising before them, depend on their reports. If they do not make such costs a part of their report, they cannot be taxed or recovered. In principle, then, there is no difference between an increase of damages, and an increase of costs only, in a report made by two referees, after a recommitment; and I am unable to see why one should be made in fact.
The case of Walker in error v. Melcher 14 Mass. 148, seems to tnc to be worthy of consideration. By the printed report of it, and an inspection of the record in the clerk’s office, it appears that all the referees met and fully heard the parties November 30 *4701813, as they state in their report ; and then it is stated thus ; “ and now on this seventh day of March 1814, report” &c. This report was signed by two only of the referees; but the third certified at the bottom of the report, under date of November 30,1813, in these words ; — " I William Hawes, one of the referees within named, do not give any opinion as to the damage ; but say said Melcher 3d, has cause for action against said Walker for a libel.” The natural implication from the whole is, that nothing was said, or at least, decided, on the day of the hearing. In March following, two, in the absence of the third, agreed and signed the. report. In that case two of the referees decided, upon the evidence which all had heard, and decided in the absence of Hawes. In the case at bar two did no more, in the absence of Elden. They only corrected an error in their former opinion upon the evidence, and thereby increased the damages ; and in Walker v. Melcher, the two formed an original opinion upon the evidence as to the amount of damages. What sound distinction is there, in principle, between the two cases ? I do not perceive any.
The supposed error which is assigned, viz. that the last meeting of the referees was not notified by the chairman, cannot be a circumstance of any kind of importance. If referees meet by a mutual understanding, without formal notice from any one, it is sufficient ; the object of the notice is to convene the referees. The chairman, as he is called, possesses no more or greater authorities and powers than either of his associates. It is merely a matter of courtesy that he should, as he generally does, notify the other referees, and the parties, of the time and place of hearing ; but, in legal contemplation, one of them may as well do this, as the other ; or else there might be a failure of justice. Surely, if the chairman refuse to notify a meeting, he cannot thereby arrest and frustrate all proceedings under the submission ; this would be extending courtesy to an unreasonable and dangerous length. In the present case, for some unexplained reasons, best known to Mr. Elden himself, he declined to notify a new meeting, pursuant to the order of recommitment, though he was particularly requested so to do; but by the record it appears that such meeting was notified ; and that all the referees and *471both the parties had notice, and were requested to attend at the time and place appointed; and it also appears that no new evidence was offered of any kind, or any hearing of the parties or their counsel had ; and that in the absence of Elden, the other referees agreed upon and signed the report in question ; and that it is predicated on the original facts and merits, and on them only. They had no other sources of information, or basis of calculation and judgment, than all had on the former hearing. All that was done by the two was, to review and revise the evidence and first report, and to exercise the unquestioned right of judging for themselves, and drawing their own conclusions upon such review and revision. But it is said that they had no authority to do this, in the absence of Elden, because, had he been present, his arguments might have changed their opinion, and convinced them that the damages ought not to have been increased. Why then did he not attend as requested, and give his brethren the benefit of his arguments and opinion ? His absence is not chargeable to them, nor to the town of North Yarmouth ; and why should his absence under such circumstances, and his consequent silence upon the subject of the submission, affect the validity of the report, any more than his presence and total silence would have done ? In the latter case, surely, the report would be liable to no objection. For one, I do not feel at liberty to establish distinctions, where I can discern no difference.
In my opinion the report was properly accepted, and judgment thereon properly rendered ; and, of course, that it ought to be affirmed. Judgment reversed.