George R. Romine, bankrupt, having applied for his discharge, on February 6, 1905, certain creditors filed specifications in opposition thereto, and thereupon, on that day, my predecessor entered an order referring the case to referee George W. Johnson to ascertain and report the facts touching said matter of discharge. This action is authorized by section 3 of No. xii of the General Orders in Bankruptcy (89 Fed. vii; 32 C. C. A. xvi) prescribed by the Supreme Court, reading as follows:
“(3) Applications for a discharge, or for the approval of a composition, or for an injunction to stay proceedings of a court or officer of the United States or of a state, shall be heard and decided by the judge. But he may refer such an application, or any specified issue arising thereon, to the referee to ascertain and report the facts.”
In compliance with this order, the referee, on the 13th day of February, 1905, commenced the taking of testimony with a view to ascertaining such facts, and up to April 18, 1905, had made a record of 342 full typewritten pages, consisting largely of wholly immaterial matters, captious objections, remarks of counsel, reiterations of the same questions and demands, to such an extent as to trespass to the last limit upon the patience of any court required to read it. It is absolutely safe to say, in my judgment, that every fact adduced in this record that in the remotest degree could be deemed material could and ought to have been clearly and fully presented in a record of 50 such typewritten pages. This criticism, kindly made, is justified’by the fact that the end of it has by no means, apparently, been as yet reached, for these “depositions” stand uncompleted and the referee’s duty unperformed, because, on petition and application of the protesting creditors, an order was entered on the-day of April, 1905, in the nature of a rule by this court, against said referee, to cause him to answer and show cause for his alleged misconduct in refusing to admit testimony offered, and refusing to certify for review questions arising before him touching the admissibility of such evidence, and by his rulings practically causing certain witnesses to refuse both to answer certain questions and produce certain written evidence. To this petition and rule the referee has made answer, and has certified the evidence taken, and this matter is now before the court for its consideration.
It seems that the parties, all represented by counsel, started out under the idea that the referee was to sit as a court and determine upon what testimony was admissible and what was inadmissible, and that they were to be bound by his rulings in this particular. A large amount of the testimony was taken under this understanding, when, it appears, counsel for the opposing creditors objected, and, *839supporting the objection with authorities that convinced the referee that he had been proceeding wrongly, he reversed his ruling, and determined that, notwithstanding his opinion and judgment was-against the admissibility of certain evidence, he would nevertheless-permit the witness to answer the questions and allow the evidence to go to the court. To correct the error of his first method of procedure, the referee recalled the witness Prewett, about whose testimony the trouble had arisen, and allowed all rejected matters to be' inquired of under his statement and puling that the objection should be sustained. Thereupon the witness Prewett refused to produce certain books demanded of him, and the referee refused to compel him to do so, and further refused to certify the matter to the court for revision, on the ground that, before so certifying, all the testimony should be taken, and all questions of objection certified at one and the same time. It should be added that the protesting creditors insist that his expressing his opinion touching the admissibility of the testimony encouraged witnesses to refuse to answer and produce such testimony. Thus it will be seen that substantially three practical points have arisen touching the practice to be observed by the referee in taking testimony before him: (1) How-far has he the power to pass upon and determine the admissibility of evidence presented to him? (2) When is he required to certify objections made to his rulings to the court for revision ? (3) What power has he to determine as to whether a witness is recalcitrant and in contempt or not, and, if held to be so, what proceeding should he take against him ?
The first question presents little difficulty. In Re Wilde’s Sons, 11 Am. Bankr. Rep. 714, 131 Fed. 142, it is held that a referee acting in his character of referee or as special commissioner has the right to exclude evidence which he deems inadmissible. But many other cases hold the contrary. In Re Lipset, 9 Am. Bankr. Rep. 32, 119 Fed. 379, it is held that hearings before referees are substantially the same as in equity, subject, in effect, to equity rule 67, and therefore it is the duty of the referee, although he must rule on any objections made to testimony offered, to take all excluded testimony and make the same a part of the record, with his ruling on the objections, and also the exceptions which may be taken noted in connection with such testimony. It is clearly set forth in this case that the reason for this procedure is to enable the judge on a review not to reverse a decision made because of the error of the referee in excluding evidence, but enable such judge to at once, without reference back to take such testimony, to determine the issue upon the proper testimony, disregarding that which was improper. In re Natelle De Gottardi, 7 Am. Bankr. Rep. 723, 114 Fed. 328; Dressel v. North State Lumber Co., 9 Am. Bankr. Rep. 541, 119 Fed. 531; In re Covington, 6 Am. Bankr. Rep. 373, 110 Fed. 143; Blease v. Garlington, 92 U. S. 1, 23 L. Ed. 521.
Referees are judicial officers clothed with judicial powers. They-are, however, subordinate to the court above them, and should so-conduct their proceedings and make up their records that a full and. fair review may be made of their actions. Their decisions will not *840be lightly- treated, but given the consideration due to conclusions reached by conscientious officers seeking to discharge their duties to the best of their ability. “That they sometimes err is to be expected — so do the ablest judges of all the courts — but they should not be reversed except upon clear and convincing proof of error, especially as to the findings of fact, when they have seen the witnesses and heard them testify.” In re Covington, 6 Am. Bankr. Rep. 374, 110 Fed. 143; In re Shriver, 10 Am. Bankr. Rep. 746, 125 Fed. 511.
But it is needless to adduce further authority touching this point, for, in my judgment, General Orders in Bankruptcy No. xxii (89 Fed. x; 32 C. C. A. xxv), fully determines it. The latter clause of this order provides:
“The referee shall note upon the deposition any question objected to, with his decision thereon; and the court shall have power to deal with the costs of. incompetent, immaterial, or irrelevant depositions, or parts of them, as may be just.”
It is clear to me that in taking testimony the referee must have it taken down, preferably in narrative form, but, upon objection raised, it is his duty to require the matter to be presented by question, to which the objection and reason thereof is to be clearly but briefly noted; then to enter his ruling thereon as to whether proper or not, and, although he may rule it to be' improper, yet allow it to be answered. I am persuaded, however, that he is not called upon to suffer and allow counsel, as in this case, to ask and permit witnesses to answer the same question over and over again, whereby time is unnecessarily consumed and costs incurred; but that upon his noting the fact that the question has been once answered, or the demand to answer has been once positively refused, the court will justify him in preventing vain repetition. Applying these rules to this case, it is clear that the referee and the attorneys engaged all had a wrong conception of his power in the start; that they discovered their error; that the referee corrected all possible injury by recalling the witness and permitting the questions to be propounded again which he had ruled out; and that, if the witness was influenced to refuse to answer because of his ruling the questions to be improper, the referee was not blamable, because it was his duty to rule, and the witness alone could at his peril refuse to answer. There was therefore in this first proposition nothing to warrant this rule against the referee.
(2) When is a referee required to certify objections made to his rulings to the court for revision ? Must he do so every time a question, is asked which he rules is objectionable, or every time he may express an opinion during its taking touching the evidence-? Certainly not. The very reason for the establishment of the first proposition — that he is to take down testimony which he believes and rules improper — is for the very purpose of preventing constant and vexatious certificates for revision. In any matter wherein he is by the law empowered to enter orders that under the law may become final when he has entered such order, a revision may be had — a re*841vision of his judicial act, which, unrevoked, binds parties and becomes the law of the case. General Order No. xxvii settles this:
“When a bankrupt, creditor, trustee, or other person shall desire a review by the judge of any order made by the referee, he shall file with the referee his petition therefor setting out the error complained of; and the referee shall forthwith certify to the judge the question presented, a summary of the evidence relating thereto, and the finding and order of the referee thereon.”
See In re Hawley, 8 Am. Bankr. Rep. 632, 116 Fed. 428; Blease v. Garlington, 92 U. S. 1, 23 L. Ed. 521.
The referee, therefore, was right in refusing to certify for revision his rulings upon this testimony. In fact, this being a matter referred to him specially to ascertain facts alone designed to aid the court in performing its duty in determining whether the bankrupt should be discharged or not, no possible revision could be had. The utmost extent the court could go would be to discharge the referee from further consideration of the matter, and take it up and decide it without his assistance. There was therefore no excuse for this rule against the referee on this ground.
This brings us to the third proposition: What power has a referee to determine as to whether a witness is recalcitrant and in contempt or not, and, if held to be so, what proceeding should he take against him? This proposition may be dissolved into two heads: (a) his power to determine as to whether contempt exists; (b) his method of procedure. This last presents no difficulty; in fact, is well settled. In a case where a referee believes a witness improperly refuses to testify or produce written testimony — in other words, to be in contempt for any reason — it is his plain duty to set forth the contempt upon his record, certifying the facts to the district judge, who will then deal with the question as if the contempt had originally arisen in his court. In re Miller, 5 Am. Bankr. Rep. 184, 105 Fed. 57; In re McCormick, 3 Am. Bankr. Rep. 340, 97 Fed. 566; Loveland, Bley. (2d Ed.) 671, § 240.
But the question as to the referee’s power to determine as to whether contempt exists in a given case presents a question of very serious difficulty. It is always to be borne in mind that the power to punish for contempt should by all courts be exercised with caution; that it is in the nature of a criminal proceeding, and the evidence to sustain it should be beyond reasonable doubt. In re McCormick, supra. It is also always to be borne in mind that it is an inherent constitutional right that a witness has to refuse testimony that is improper, just the same as it is his absolute duty to give proper testimony if in his knowledge or possession. It is true, in ordinary cases, he will not .be suffered in the courts to determine for himself whether the evidence is proper or not. The court will decide that for him, and he must then answer or suffer the consequences. Still the books'have given us many cases where parties have refused to answer, have appealed to a superior court, and, because the evidence was determined by it to be improper, have been released and acquitted from the charge of contempt. What shall a referee do in a case like this when his conscience and judgment tell him that the evidence is improper, immaterial, and not pertinent, and *842he has so ruled? Must he nevertheless stop, and certify that a contempt has been committed, which he does not believe to be true, simply because the parties may demand it? When the right of protection guarantied by all courts to a witness is taken into consideration, I am clearly of the opinion that the referee has a right to and must determine judicially in the first place whether a contempt has been committed, and, if he thinks not, to refuse, no matter how strongly urged, to certify the matter for contempt proceedings-to the judge. But what recourse has the party aggrieved who is-seeking the testimony? I answer, that he may file his petition for revision, or his exception to the referee’s report, in due and proper season, to have his judgment reviewed as in other matters. It seems-to me that any rule less cautious than this would not properly secure the protection of a witness from wholly improper demands for disclosure of his own private matters, and from vexatious prosecution for failure to make such disclosures.
This case, it seems to me, presents a very striking illustration.. I desire to refrain from discussing the evidence presented, as far as possible, at this time, because it is apparently not yet completed, and it must all be considered carefully upon the hearing of the motion for discharge. Briefly, the trouble has largely, if not altogether, arisen over the fact that witness Prewett would not comply with a general demand of the protesting creditors to furnish for their examination “the books” belonging to a partnership existing, between him and Jno. W. Romine, bankrupt’s son, and which by the articles of partnership was conducted alone in the name of Prewett. The procedure in the case, to say the least, has been remarkable. There are two contesting creditors, the Bank of Ravenswood and R. H. Douglass. They join in presenting a specification and an amended specification against the discharge of the bankrupt. The amended specification sets forth 16 different acts-on the part of the bankrupt, any one of which should, if true, prevent his discharge, and several should cause his criminal punishment. These specifications are not made upon “information and belief.” They enter into details; are charged absolutely and set forth as positive facts. They are sworn to by Douglass, who, under oath, says that he has read them, and that they are true, except so far as stated to be made upon information, which is not done in any line of the document, so far as I have found. Here is a man who is full of absolute knowledge touching the facts, or else he has presented the most reckless sworn pleading I have ever known presented in a court of justice. He is plaintiff in the issue, yet has never so far-testified to these material facts. Instead, the first witness he calls isPrewett, who, in effect, he charges positively with having been in partnership with G. R. Romine, the bankrupt, instead of with J. W. Romine, the son, and with having taken a conveyance of the bankrupt’s interest in the partnership fraudulently, in order to conceal it from creditors. He makes Prewett his witness. Prewett denies-most positively these allegations; denies that the bankrupt had any interest in such partnership; brings forward the original written agreement whereby said partnership was established between him*843self and Jno. W. Romine (not Geo. R. Romine), dated January 1, 1902, duly acknowledged by both before a notary, near three years before this bankruptcy proceeding; also the written agreement, dated January 2,1904, between himself and Jno. W. Romine, whereby said partnership was dissolved. He produced his cashbook, journal, ledger, the notes that had been executed that were demanded, a statement of the bankrupt’s account with the firm, and made the positive statement that he had disclosed all that his books would show concerning his relations with either Jno. W. or Geo. R. Romine. Nevertheless, these protesting creditors, without even specifying what special thing they could prove pertinent to the issue, made a general demand for all the books, reiterating that demand time and again after the referee had ruled it improper for them to be produced.
It is clear to my mind that these protesting creditors had absolutely no right, under the circumstances, to demand these books. They had no right to a discovery; they needed none. Douglass by his affidavit showed himself with full and positive knowledge. Prewett was their witness; they were bound by his statement that he had produced all that in any way pertained to his relation and transactions with both Romines, until, at least, they had clearly contradicted him. The effort to do this by the witness Stone proved utterly futile. His testimony was vague, uncertain, contradictory, and unsatisfactory, if not on its face incredible. He manifestly was a willing witness, for he says, “Oh, there are some other statements I would like to make, but would rather not” — manifestly because he could not truthfully make them; at any rate, when pressed, he would not risk making them. In the single pertinent matter testified to by him he is flatly contradicted by Moss, a wholly disinterested witness. Upon analysis, I do not think the evidence of witness Gates presents itself in any better plight, and these are the two relied on to contradict their other witness Prewett and base a motion to produce books which he says related only to his private affairs, and the exposure of which might injure him and in no way aid the creditors or other parties.
It has been settled beyond peradventure for very many years that courts do not compel production of books simply to gratify curiosity, or permit “fishing” excursions into them to see what can be found that may or may not be of advantage to the parties making the demand. “A party cannot obtain a roving commission for the inspection or production of books or papers in order that he may ransack them for evidence to make out his case.” “He is entitled to production and inspection only when the same is material and necessary to establish his cause of action.” “The application will not be granted where the facts to be proved by the books can be otherwise established.” “It will therefore be denied when the party has in his possession or under his control the means of acquiring all the information he seeks to obtain, or when the books do not in themselves contain evidence, but merely information by which evidence can be obtained. It is not permitted to enable a party to ascertain whether he has cause of action or defense, or to *844ascertain the evidence on which his opponent’s action or defense rests.” These quotations are from 14 Cyc. pp. 370 to 382, where they and many other similar principles touching the production of private books, their use, the manner of their examination, etc., are laid down, and where more than 400 cases are collated and cited. See, also, In re Carley (D. C.) 106 Fed. 862; Southern Ry. Co. v. North Car. Corp. Com. (C. C.) 104 Fed. 700; Henry v. Ins. Co. (C. C.) 35 Fed. 15; In re Pacific Ry. Com. (C. C.) 32 Fed. 250; Triplett v. Bank, 24 Fed. Cas. No. 14,178; Abrahams v. Swann, 18 W. Va. 274, 41 Am. Rep. 692.
Without further comment or discussion, it is sufficient for me to say that I do not regard these protesting creditors to be entitled to the books of Prewett for examination under the circumstances disclosed by the testimony. Further, I do not regard any of the complaints made against the ruling and action of the referee as well founded; but, on the contrary, I regard the rule against him to have been improvidently awarded, and it will now be discharged, with direction for him to complete his reference in this case so soon as he can reasonably do so.