Knowlton, J.
The defendant published in its newspaper an article describing the conduct of a prisoner brought before the Municipal Court of Boston, and the proceedings of the court in the case, designating him as “ H. P. Hanson, a real estate and insurance broker of South Boston.” He was, in fact, a real estate and insurance broker of South Boston, and the article was substantially true, except that he should have been called A. P. H. Hanson instead of H. P. Hanson. The plaintiff, H. P. Hanson, is also a real estate and insurance broker in South Boston, and in writing the article the reporter used his name by mistake.
In a suit for libel or slander, it is always necessary for the plaintiff to allege and prove that the words were spoken or written of and concerning the plaintiff. In Baldwin v. Hildreth, 14 Gray, 221, the declaration was adjudged bad on demurrer, because this allegation was wanting. The rule is reaffirmed, and authorities are cited, in McCallum v. Lambie, 145 Mass. 234. The form of declaration prescribed by the Practice Act in slander uses the phrase “ words spoken of the plaintiff,” and in libel, “ false and malicious libel concerning the plaintiff.” Pub. Sts. c. 167, § 94. It has often been held that it is a question of fact for the jury whether the words were or were not spoken or written “ of and concerning the plaintiff.” Van Vechten v. Hopkins, 5 Johns. 211, 221. Gibson v. Williams, 4 Wend. 320. Smart v. Blanchard, 42 N. H. 137. De Armond v. Armstrong, 37 Ind. 35. Goodrich v. Davis, 11 Met. 473, 480, 481, 484. Miller v. Butler, 6 Cush. 71. The defendant’s meaning in regard both to the person to whom [295]*295the words should be applied and the imputations against him is always to be ascertained. In Smart v. Blanchard, ubi supra, it is said that “ the meaning ... in this respect [as to the person to whom the libel applies] is undoubtedly a question of fact to be found by the jury.” It is also said that, when the meaning is ambiguous, it is incumbent on the plaintiff “to show that the defendant intended to apply his remarks to the plaintiff.” In Le Fanu v. Malcomson, 1 H. L. Cas. 637, which was an action for libel brought by copartners, the Lord Chancellor assumes that the plaintiff must prove “ that the party writing the libel did intend to allude to them.”
In Pub. Sts. c. 167, § 94, the rule is laid down as applicable “ to actions for written and printed, as well as oral slander,” that if the meaning is not clear there must be innuendoes to make the words intelligible, “in the same sense in which they were spoken.” Chenery v. Goodrich, 98 Mass. 224, 229, assumes that it must appear that the plaintiff was referred to in the publication, and Young v. Cook, 144 Mass. 38, is of similar import. Odgers on Libel and Slander, at page 127, discusses the topic “ Certainty as to the person defamed.” In Commonwealth v. Kneeland, 20 Pick. 206, 216, Chief Justice Shaw says that in actions of libel and slander it is the general rule that “ the language shall be construed ... in the sense in which the writer or speaker intended it.” In Smith v. Ashley, 11 Met. 367, the necessity of proving the defendant’s actual intention in regard to the person referred to was affirmed much more strongly than there is any occasion to affirm it, and perhaps more strongly than we should be prepared to affirm it in the present case. It was held that the publisher of a newspaper containing an article which he believed to be a fictitious narrative or mere fancy sketch was not liable to the plaintiff, although the article was libellous, and was intended by the writer to be applied to the plaintiff. The court said that in such a case the writer alone was responsible.
In every action of this kind the fundamental question is, What is the meaning of the author of the alleged libel or slander, conveyed by the words used interpreted in the light of all the circumstances? The reason of this is obvious. Defamatory language is harmful only as it purports to be the expression of the thought of him who uses it. In determining the effect of a slander the [296]*296questions involved are, What is the thought intended to be expressed, and how much credit should be given to him who expresses it? The essence of the wrong is the expression of what purports to be the knowledge or opinion of him who utters the defamatory words, or of some one else whose language he repeats. His meaning, to be ascertained in a proper way, is what gives character to his act, and makes it innocent or wrongful. The damages depend chiefly upon the weight which is to be given to his expression of his meaning, and all the questions relate back to the ascertainment of his meaning.
In the present case we are concerned only with the meaning of the defendant in regard to the person to whom the language of the published article was to be applied, and the question to be decided is, How may his meaning legitimately be ascertained? Obviously, in the first place, from the language used; and in construing and applying the language, the circumstances under which it was written and the facts to which it relates are to be considered, so far as they can readily be ascertained by those who read the words, and who attempt to find out the meaning of the author in regard to the person of whom they were written. It has often been said that the meaning of the language is not necessarily that which it may seem to have to those who read it as strangers, without knowledge of facts and circumstances which give it color and aid in its interpretation, but that which it has when read in the light of events which have relation to the utterance or publication of it.
For the purposes of this case it may be assumed, in favor of the plaintiff, that if the language used in a particular case, interpreted in the light of such events and circumstances attending the publication of it as could readily be ascertained by the public, is free from ambiguity in regard to the person referred to, and points clearly to a well known person, it would be held to have been published concerning that person, although the defendant should show that through some mistake of fact, not easily discoverable by the public, he had designated in his publication a person other than the one whom he intended to designate. It may well be held that where the language, read in connection with all the facts and circumstances which can be used in its interpretation, is free from ambiguity, the defendant will not [297]*297be permitted to show that through ignorance or mistake he said something, either by way of designating the person, or making assertions about him, different from that which he intended to say; but his true meaning should be ascertained, if it can be, with the aid of such facts and circumstances attending the publication as may easily be known by those of the public who wish to discover it.
Whether the defendant should ever be permitted to state his undisclosed intention in regard to the person of whom the words are used, may be doubtful. If language purporting to be used of only one person would refer equally to either of two different persons of the same name, and if there were nothing to indicate that one was meant rather than the other, there is good reason for holding that the defendant’s testimony in regard to his secret intention might be received, but perhaps such a case is hardly supposable. Odgers, in his book on Libel and Slander, at page 129, says: “So, if the words spoken or written, though plain in themselves, apply equally well to more persons than one, evidence may be given both of the cause and occasion of publication, and of all the surrounding circumstances affecting the relation between the parties, and also any statement or declaration made by the defendant as to the person referred to.” In Regina v. Barnard, 43 J. P. 127, when it was uncertain whether the libel referred to the complainant or not, and when the language was applicable to him, Lord Chief Justice Cockburn held the affidavit of the writer that he did not mean him, but some one else, to be a sufficient reason for refusing process. In De Armond v. Armstrong, 37 Ind. 35, evidence was received of what the witnesses understood in regard to the person referred to. In Smart v. Blanchard, 42 N. H. 137, it is stated that extrinsic evidence is to be received “ to show that the defendant intended to apply his remarks to the plaintiff,” when his meaning is doubtful. Goodrich v. Davis, 11 Met. 473, 480, 484, 485, and Miller v. Butler, 6 Cush. 71, are of similar purport. See also Barwell v. Adkins, 1 M. & G. 807; Knapp v. Fuller, 55 Vt. 311; Commonwealth v. Morgan, 107 Mass. 199, 201.
If the defendant’s article had contained anything libellous against A. P. H. Hanson, there can be no doubt that he could have maintained an action against the defendant for this publication. The name used is not conclusive in determining the [298]*298meaning of the libel in respect to the person referred to; it is but one fact to be considered with other facts upon that subject. Fictitious names are often used in libels, and names similar to that of the person intended, but differing somewhat from it. A. P. H. Hanson could have shown that the description of him by name, residence, and occupation was perfect, except in the use of the initials “H. P.” instead of “A. P. H.,” that the article referred to an occasion on which he was present, and gave a description of conduct of a prisoner, and of proceedings in court, which was correct in its application to him and to no one else. The internal evidence when applied to facts well known to the public would have been ample to show that the language referred to him, and not to the person whose name was used.
So, in the present suit, the court had no occasion to rely on the testimony of the writer as to the person to whom the language was intended to apply. The language itself, in connection with the publicly known circumstances under which it was written, showed at once that the article referred to A. P. H. Hanson, and that the name H. P. Hanson was used by mistake. As the evidence showed that the words were published of and concerning A. P. H. Hanson, the finding that they- were not published of the plaintiff followed of necessity. The article was of such a kind that it referred, and could refer, to one person only; when that person was ascertained, it might appear that the publication as against him was or was not libellous, and his rights, if he brought a suit, would depend upon (he finding in respect to that. No one else would have a cause of action, even if, by reason of identity of name with that used in the publication, he might suffer some harm. For illustration, suppose a libel is written concerning a person described as John Smith of Springfield. Suppose there are five persons in Springfield of that name. The language refers to but one. When we ascertain by legitimate evidence to which one the words are intended to apply, he can maintain an action. The other persons of the same name cannot recover damages for a libel merely because of their misfortune in having a name like that of the person libelled. Or, if the defendant can justify by proving that the words were true, and published without malice, he is not guilty of a libel, even if, written of other persons of the same name of whose existence very likely he was ignorant, the [299]*299words would be libellous; otherwise, one who has published that which by' its terms can refer to but one person, and be a libel on him only, might be responsible for half a dozen libels on as many different persons, and one who has justifiably published the truth of a person might be liable to several persons of the same name of whom the language would be untrue. The law of libel has never been extended, and should not be extended, to include such cases.
Whether there should be a liability founded on negligence in any case when the truth is published of one to whom the words, interpreted in the light of accompanying circumstances easily ascertainable by those who read them, plainly apply; and where, by reason of identity of names, or similarity of names and description, a part of the public might think them applicable to another person of whom they would be libellous, is a question which does not arise on the pleadings in this case. So far as we are aware, no action for such a cause has ever been maintained. It is ordinarily to be presumed, although it may' not always be the fact, that those who are enough interested in a person to be affected by what is said about him, will ascertain, if they easily can, whether libellous words which purport to refer to one of his name were intended to be applied to him or to some one else.
The question in this case, whether the words were published of and concerning the plaintiff, was one of fact on all the evidence. Unless it appears that the matters stated in the report would not warrant a finding for the defendant, there must be judgment for him, even if the finding of fact might have been the other way. We are of opinion that the finding was well warranted, and there must be,
Judgment on the finding.
Holmes, J.
I am unable to agree with the decision of the majority of the court, and as the question is of some importance in its bearing on legal principles, and as I am not alone in my views, I think it proper to state the considerations which have occurred to me.
The first thing to determine is what question is presented. If we were to stop with the words in which the conclusion of the report is couched there would be no question at all. “ The court found as a fact that the alleged libel declared on by the [300]*300plaintiff was not published by the defendant of or concerning the plaintiff.” But it is not to be supposed that a justice of the Superior Court would send a report to this court in which he did not intend to present a question of law. The so called finding either is a ruling on the effect of the facts previously found, or at least, putting it in the most favorable way for the defendant, is a conclusion drawn from those facts alone. Whether the conclusion be one of fact or of law, the question is whether it is justified by the facts set forth, without other facts or evidence.
The facts are that libellous matter was published in an article by the defendant about “ H. P. Hanson, a real estate and insurance broker of South Boston,” that the plaintiff bore that name and description, and, so far as appears, that no one else did, but that the defendant did not know of his existence, and intended to state some facts about one Andrew P. H. Hanson, also a real estate and insurance broker of South Boston, concerning whom the article was substantially true.
The article described the subject of it as a prisoner in the criminal dock, and states that he was fined, and this makes it possible to speak of the article as one describing the conduct of a prisoner. But this mode of characterization seems to me misleading. In form it describes the plight and conduct of “ H. P. Hanson, a real estate and insurance broker of South Boston.” The statement is, “ H. P. Hanson, a real estate and insurance broker of South Boston, emerged from the seething mass of humanity that filled the dock,” etc. In order to give it any different subject, or to give the subject any further qualifications or description, you have to resort to the predicate, to the very libellous matter itself. It is not necessary to say that this never can be done, but it must be done with great caution. The very substance of the libel complained of is the statement that the plaintiff was a prisoner in the criminal dock, and was fined. The object of the article, which is a newspaper criminal court report, is to make that statement. The rest of it amounts to nothing, and is merely an attempt to make the statement amusing. If an article should allege falsely that A. murdered B. with a knife, it would not be a salisfncmry answer to an action by A. that it was a description of the conduct of the murderer of B., and was true concerning him. The [301]*301] ublic, or all except the few who may have been in court on the day in question, or who consult the criminal records, have no way of telling who was the prisoner except by what is stated in the article, and the article states that it was “ H. P. Hanson, a real estate and insurance broker of South Boston.”
If I am right so far, the words last quoted, and those words nlone, describe the subject of the allegation, in substance as well as in form. Those words also describe the plaintiff, and no one else. The only ground, then, on which the matters alleged of and concerning that subject can be found not to be alleged of and concerning the plaintiff, is that the defendant did not intend them to apply to him, and the question is narrowed to whether such a want of intention is enough to warrant the finding, or to constitute a defence, when the inevitable consequence of the defendant’s acts is that the public, or that part of it which knows the plaintiff, will suppose that the defendant did use its language about him.
On general principles of tort, the private intent of the defendant would not exonerate it. It knew that it was publishing statements purporting to be serious, which would be hurtful to a man if applied to him. It knew that it was using as the subject of those statements words which purported to designate a particular man, and would be understood by its readers to designate one. In fact, the words purported to designate, and would be understood by its readers to designate, the plaintiff. If the defendant had supposed that there was no such person, and had intended simply to write an amusing fiction, that would not be a defence, at least unless its belief was justifiable. Without special reason, it would have no right to assume that there was no one within the sphere of its influence to whom the description answered. The case would be very like firing a gun into a street, and, when a man falls, setting up that no one was known to be there. Commonwealth v. Pierce, 138 Mass. 165, 178. Hull’s case, Kelyng, 40. Rex v. Burton, 1 Strange, 481. Rigmaidon’s case, 1 Lewin, 180. Regina v. Desmond, Steph. Cr. Law, 146. So, when the description which points out the plaintiff is supposed by the defendant to point out another man whom in fact it does not describe, the defendant is equally liable as when the description is supposed to point out nobody. On [302]*302the general principles of tort, the publication is so manifestly detrimental that the defendant publishes it at the peril of being able to justify it in the sense in which the public will understand it.
But in view of the unfortunate use of the word “malice” in connection with libel and slander, a doubt may be felt whether actions for these causes are governed by general principles. The earliest forms of the common law known to me treat slander like any other tort, and say nothing about malice. 4 Seld. Soc. Pub. 40, 48, 61. Probably the word was borrowed at a later, but still early date, from the malitia of the canon law. By the canon law, one who maliciously charged another with a grave sin incurred excommunication, ipso facto. Lyndw., Provinciate, lib. 5, tit. 17 (De Sent. Excomm. c. 1, Auctoritate Dei). Oughton, Ordo Judiciorum, tit. 261. Naturally malitia was defined as cogitatio malee mentis, coming near to conscious malevolence. Lyndw., ubi supra, note/. Naturally also for a time the common law followed its leader. Three centuries ago it seems to have regarded the malice alleged in slander and libel as meaning the malice of ethics and the spiritual law.
In the-famous case where a parson in a sermon repeated, out of Fox’s Book of Martyrs, the story “ that one Greenwood, being a perjured person, and a great persecutor, had great plagues inflicted upon him, and was kilted by the hand of God, whereas in truth he never was so plagued, and was himself present at that sermon,” and afterwards sued the parson for the slander, Chief Justice Wray instructed the jury “ that, it being delivered but as a story, #,nd not with any malice or intention to slander any, he was not guilty of the words maliciously; and so was found not guilty.” Greenwood v. Prick, stated in Brook v. Montague, Cro. Jac. 90, 91. See also Crawford v. Middleton, 1 Lev. 82, ad fin.
But that case is no longer law. Hearne v. Stowell, 12 A. & E. 719, 726. The law constantly is tending towards consistency of theory. For a long time it has been held that the malice alleged in an action of libel means no more than it does in other actions of tort. Commonwealth v. York, 9 Met. 93, 104, 105. Gassett v. Gilbert, 6 Gray, 94, 97. Abrath v. Northeastern Railway, 11 App. Cas. 247, 253, 254. See Commonwealth v. [303]*303Pierce, 138 Mass. 165, 175 et seq.; White v. Duggan, 140 Mass. 18, 20. Indeed, one of the earliest cases to state modern views was a case of libel. Bromage v. Prosser, 4 B. & C. 247, 255. Accordingly, it recently was laid down by this court that the liability was the usual liability in tort for the natural consequences of a manifestly injurious act. Burt v. Advertiser Newspaper Co. 154 Mass. 238, 245. A man may be liable civilly, and formerly, at least by the common law of England, even criminally, for publishing a libel without knowing it. Curtis v. Mussey, 6 Gray, 261. Commonwealth v. Morgan, 107 Mass. 199. Dunn v. Hall, 1 Ind. 344. Rex v. Walter, 3 Esp. 21. Rex v. Grutch, Mood. & Malk. 433. See also Rex v. Cuthell, 27 St. Tr. 642. And it seems he might be liable civilly for publishing it by mistake, intending to publish another paper. Mayne v. Fletcher, 4 Man. & Ry. 311, 312, note. Odgers, Libel and Slander, (2d ed.) 5. So, when by mistake the name of the plaintiff's firm was inserted under the head “ First Meetings under the Bankruptcy Act,” instead of under “ Dissolution of Partnerships.” Shepheard v. Whitaker, L. R. 10 C. P. 502. So a man will be liable for a slander spoken in jest, if the bystanders reasonably understand it to be a serious charge. Donoghue v. Hayes, Hayes, 265. Of course it does not matter that the defendant did not intend to injure the plaintiff, if that was the manifest tendency of his words. Curtis v. Mussey, 6 Gray, 261, 273. Haire v. Wilson, 9 B. & C. 643. And to prove a publication concerning the plaintiff, it lies upon him “ only to show that this construction, which they Ve put upon the paper, is such as the generality of readers must take it in, according to the obvious and natural sense of it.” The King v. Clerk, 1 Barnard. 304, 305. See further Fox v. Broderick, 14 Ir. C. L. 453; Odgers, Libel and Slander, (2d ed.) 155, 269, 435, 638. In Smith v. Ashley, 11 Met. 367, the jury were instructed that the publisher of a newspaper article written by another, and supposed and still asserted by the defendant to be a fiction, was not liable if he believed it to be so. Under the circumstances of the case, “ believed ” meant “ reasonably believed.” Even so qualified, it is questioned by Mr. Odgers if the ruling would be followed in England. Odgers, Libel and Slander, (1st Am. ed.) 387, (2d ed.) 638. But it has no application to this case, as [304]*304here the defendant’s agent wrote the article, and there is no evidence that he or the defendant had any reason to believe that H. P. Hanson meant any one but the plaintiff.
The foregoing decisions show that slander and libel now, as in the beginning, are governed by the general principles of the law of tort, and, if that be so, the defendant’s ignorance that the words which it published identified the plaintiff is no more an excuse, than ignorance of any other fact about which the defendant has been put on inquiry. To hold that a man publishes such words at his peril, when they are supposed to describe a different mail, is hardly a severer application of the law, than when they are uttered about a man believed on the strongest grounds to be dead, and thus not capable of being the subject of a tort. It has been seen that by the common law of England such a belief would not be an excuse. Hearne v. Stowell, 12 A. & E. 719, 726, denying Parson Prick’s case.
I feel some difficulty in putting my finger on the precise point of difference between the minority and majority of the court. I understand, however, that a somewhat unwilling assent is yielded to the general views which I have endeavored to justify, and I should gather that the exact issue was to be found in the statement that the article was one describing the conduct of a prisoner brought before the Municipal Court of Boston, coupled with the later statement that the language, taken in connection with the publicly known circumstances under which it was written, showed at once that the article referred to A. P. H. Hanson, and that the name of H. P. Hanson was used by mistake. I have shown why it seems to me that these statements are misleading. I only will add on this point, that I do not know what the publicly known circumstances are. I think it is a mistake of fact to suppose that the public generally know who was before the Municipal Criminal Court on a given day. I think it is a mistake of law to say that, because a 'small part of the public have that knowledge, the plaintiff cannot recover for the harm done him in the eyes of the greater part of the public, probably including all his acquaintance who are ignorant about the matter, and I also think it no sufficient answer to say that they might consult the criminal records, and find out that probably there was some error. Blake v. Stevens, 4 F. & F. 232, 240. [305]*305If the case should proceed further on the facts, it might appear that, in view of the plaintiff’s character and circumstances, all who knew him would assume that there was a mistake, that the harm to him was 'merely nominal, and that he had been too hasty in resorting to an action to vindicate himself. But that question is not before us.
E. jD. Loring, for the plaintiff.
6r. M. Palmer, for the defendant.
With reference to the suggestion that, if the article, in addition to what was true concerning A. P. H. Hanson, had contained matter which was false and libellous as to him, he might have maintained an action, it is unnecessary to express an opinion. I think the proposition less obvious than that the plaintiff can maintain one. If an article should describe the subject of its statements by two sets of marks, one of which identified one man and one of which identified another, and a part of the public naturally and reasonably were led by the one set to apply the statements to one plaintiff, and another part were led in the same way by the other set to apply them to another, I see no absurdity in allowing two actions to be maintained. But that is not this case.
Even if the plaintiff and A. P. H. Hanson had borne the same name, and the article identified its subject only by a proper name, very possibly that would not be enough to raise the question. For, as every one knows, a proper name always purports to designate one person and no other, and although, through the imperfection of our system of naming, the same combination of letters and sounds may be applied to two or more, the name of each, in theory of law, is distinct, although there is no way of finding out which person was named but by inquiring which was meant. “ Licet idem sit nomen, tamen diversum est propter diversitatem personae.” Bract, fol. 190 a. Commonwealth v. Bacon, 135 Mass. 521, 525. Cocker v. Crompton, 1 B. & C. 489. In re Cooper, 20 Ch. D. 611. Mead v. Phenix Ins. Co. 158 Mass. 124, 125. Kyle v. Kavanagh, 103 Mass. 356. Raffles v. Wichelhaus, 2 H. & C. 906.
Mr. Justice Morton and Mr. Justice Barker agree with this opinion.
The article was as follows: “ He Waxed Eloquent. H. P. Hanson fined ten dollars for refusing payment of car fare. . . . H. P. Hanson, a real estate and insurance broker of South Boston, emerged from the seething ■mass of humanity that filled the dock and indulged in a wordy bout with •policeman Hogan, who claimed to have arrested Hanson on the charge of evading car fare and being drunk at the same time. The judge agreed that ■the prisoner was sober, but on the charge of evasion of car fare the evidence warranted the fining of the eloquent occupant of the dock ten dollars without costs, which he paid.”