In this case there must be a new trial. We shall state the grounds on which we come to this conclusion, and shall discuss such of the rulings as dealt with questions which are likely to come up again. Some matters not likely to recur we shall pass over. The first question which we shall consider is raised by the presiding judge’s refusal to rule that the articles were privileged. The requests referred to each article as a whole. Each article contained direct and indirect allegations of fact touching the plaintiff, and highly detrimental to him, charging him with being a party to alleged frauds in the New York custom-house. Some or all of these allegations we must take to be false. In our opinion the rulings asked were properly refused.
We agree with the defendant, that the subject was of public interest, and that in connection with the administration of the custom-house the defendant would have a right to make fair comments on the conduct of private persons affecting that administration in the way alleged. But there is an important distinction to be noticed between the so called privilege of fair criticism upon matters of public interest, and the privilege existing in the case, for instance, of answers to inquiries about the character of a servant. In the latter case, a bona fide statement not in excess of the occasion is privileged, although it turns out to be false. In the former, what is privileged, if that is the proper term, is criticism, not statement, and however it might be if a person merely quoted or referred to a statement as made by others, and gave it no new sanction, if he takes upon himself in his own person to allege facts otherwise libellous, he will not be privileged if those facts are not true. The reason for the dis*243tinction lies in the different nature and degree of the exigency and of the damage in the two cases. In these, as in many other instances, the law has to draw a line between conflicting interests, both intrinsically meritorious. When private inquiries are made about a private person, a servant, for example, it is often impossible to answer them properly without stating facts, and those who settled the law thought it more important to preserve a reasonable freedom in giving necessary information than to insure people against occasional unintended injustice, confined as it generally is to one or two persons. But what the interest of private citizens in public matters requires is freedom of discussion rather than of statement. Moreover, the statements about such matters which come before the courts are generally public statements, where the harm done by a falsehood is much greater than in the other case. If one private citizen wrote to another that a high official had taken a bribe, no one would think good faith a sufficient answer to an action. He stands no better, certainly, when he publishes his writing to the world through a newspaper, and the newspaper itself stands no better than the writer. Sheckell v. Jackson, 10 Cush. 25, 26.
The distinction to which we have referred has been brought out more clearly in England than it has been in our own decisions. Thus, in Davis v. Shepstone, 11 App. Cas. 187, 190, Lord Herschell says: “ It is one thing to comment upon or criticise, even with severity, the acknowledged or proved acts of a public man, and quite another to assert that he has been guilty of particular acts of misconduct. In the present case the appellants, in the passages which were complained of as libellous, charged the respondent, as now appears without foundation, with having been guilty of specific acts of misconduct and then proceeded on the assumption that the charges were true, to comment upon his proceedings in language in the highest degree offensive and injurious ; not only so, but they themselves vouched for the statements by asserting that though some doubt had been thrown upon the truth of the story, the closest investigation would prove it to be correct. In their Lordships’ opinion there is no warrant for the doctrine that defamatory matter thus published is regarded by the law as the subject of any privilege.” Popham v. Pickburn, 7 H. & N. 891, 898. Lefroy v. Burnside, 4 L. R. Ir. 556, *244565, 566. Campbell v. Spottiswoode, 3 B. & S. 769, 779. Walker v. Brogden, 19 C. B. (N. S.) 65. Regina v. Flowers, 44 J. P. 377, 378. Negley v. Farrow, 60 Md. 158. See Hamilton v. Eno, 81 N. Y. 116; State v. Schmitt, 20 Vroom, 579, 586; Eviston v. Cramer, 57 Wis. 570 ; Scripps v. Foster, 41 Mich. 742, 746 ; Sheckell v. Jackson, 10 Cush. 25, 26.
The foregoing language is applicable to the case at bar. The defendant in all the articles makes statements of fact on its own behalf, and in the second article fairly may be understood to intimate that the private sources of information alleged by the words “ we say this on authority,” apply not merely to the existence of corruption in the New York custom-house, but to the plaintiff’s connection with it. The articles published by the defendant, so far as they contained false statements, were not privileged. We should add, however, with reference to another trial, that there was evidence that some of the charges in the articles were true, and so far as the jury might find them to be so, inasmuch as the matter under discussion was a matter of public concern, the defendant would be justified not only in making those charges, but in free and open comment and criticism in regard to them.
The next question, the first which is raised by the bill of exceptions, is whether the court below rightly excluded a letter from the Secretary of the Treasury of the United States, and an ex parte report on the same subject made to the Treasury Department, containing similar charges against the plaintiff, coupled with evidence that the writer of the articles had these documents before him, and believed the statements contained in them. The evidence was offered to show that the defendant acted in good faith, and, as it is commonly said, to negative express malice, in support of its plea of privilege; also as bearing on damages, and generally for any purpose for which it might be admissible.
We already have considered how the defendant stands in respect of privilege. It is said that the report tended to prove that the defendant had reasonable cause to believe the charges to be true, and that it tended to show that the plaintiff was less damaged than he otherwise would have been by reason of the fact that similar charges had been made and published before. *245As to the former of these suggestions, it is enough to say that it is not a justification that the defendant had reasonable cause to believe its charges to be true. A person publishes libellous matter at his peril. Watson v. Moore, 2 Cush. 133, 140. Parkhurst v. Ketchum, 6 Allen, 406. Clark v. Brown, 116 Mass. 504, 507.
Then as to damages. The damages recovered are measured in all cases by the injury caused. Vindictive or punitive damages are never allowed in this State. Therefore, any amount of malevolence on the defendant’s part in and of itself would not enhance the amount the plaintiff recovered by a penny, and reasonable cause to believe the charges or absolute good will would not cut it down. Watson v. Moore, 2 Cush. 133. Parkhurst v. Ketchum, 6 Allen, 406. Clark v. Brown, 116 Mass. 504. Apart from the statute, (Pub. Sts. c. 167, § 80,) and possibly from privilege, the defendant’s motives and intent are totally immaterial. Its liability is the usual liability in tort for the natural consequences of a manifestly injurious act. No doubt there might be cases, especially of slander, where it is impossible photographically and phonographically to reproduce the overt act complained of, where the defendant’s motive would afford an inference as to the character of the act and the impression made by it. No doubt a manifestation of malevolent motives might enhance damages under our rule allowing damages for injured feelings. Mahoney v. Belford, 132 Mass. 393, 394. Hastings v. Stetson, 130 Mass. 76, 78. Markham v. Russell, 12 Allen, 573. See Ford v. Ford, 143 Mass. 577, 579; Leonard v. Allen, 11 Cush. 241. But there is nothing of that sort in this case. Watson v. Moore, 2 Cush. 133, 140. Clark v. Munsell, 6 Met. 373, 388.
As a general proposition, the defendant cannot show that the plaintiff’s damages are less than they otherwise would have been because the charge has been made and published before. Mahoney v. Belford, 132 Mass. 393. Peterson v. Morgan, 116 Mass. 350, and cases cited. Alderman v. French, 1 Pick. 1, 18. See Scott v. Sampson, 8 Q. B. D. 491; Hatfield v. Lasher, 81 N. Y. 246, 249, 250. But with reference to what we shall have to say concerning the fourth count, it should be added that probably a different rule would apply if the defendant’s publication pro*246fesses on its face to be based upon other publications which are referred to, and the fact is so.
The last ground on which it was argued that the report should have been admitted was that it was made a part of the libel by reference. We are of opinion' that the articles, with the exception of that set out in the fourth count, do not refer to the report in such a way as manifestly on their face to need the report in order to explain them. The report was laid before us by agreement of counsel. It is a very long document, which it would have been useless to read through to the jury, and, from such inspection as we have been able to make, does not in any way' explain, qualify, or mitigate the expressions used by the defendant. On the contrary, its only effect could be to raise either a belief or a strong suspicion that the defendant’s charges were true. For that purpose it was of course incompetent, and therefore as a practical matter the only effect of allowing it to go in would have been an injustice to the plaintiff. But the article set out in the fourth count stands somewhat differently from the others. This refers to the report in terms, and then proceeds to say that the affidavits in the report and the later affidavits denying the former, when marshalled together, make a terrible indictment against the plaintiff, — that the new affidavits evidently are intended to shield him, — and implies that the public, on reading the two sets, will believe the former. The jury, if they should compare the documents, might agree with the inferences of the defendant, and to that extent might find the allegations of the article to be true. Under this count alone the report should have been admitted, for the above reason, and with reference to the damages, as we have stated earlier.
Another exception is to the exclusion of evidence that the polariscope used in the Boston custom-house did not give too high a valuation. One evidence of fraud in the Mew York custom-house was that sugar was valued lower there than in Boston. The plaintiff admitted the fact, and gave as one reason for it that the Boston examiner overestimated the quartz plate by which he tested his instrument. He gave this reason when attempting to secure the removal of the Boston examiner from New York, after the latter had been sent there to reform the alleged New York practices. The evidence offered tended to *247prove the existence of the fraud, and perhaps that the plaintiff knew of its existence, and should have been admitted.
The court was asked to rule that the plaintiff could not recover for repetitions of the defendant’s libels by others, and several other rulings tending in the same direction were asked. The court instructed the jury as follows: “ One who publishes a libel is not responsible for the publication of it by others; that is to say, he is not responsible for the injurious act of another. But there is a general principle which runs through all tort cases, which is generally stated in this way: that a man is presumed to intend the natural and probable consequences of his acts, and that request for instruction therefore should be qualified in that way. He is not responsible for the injurious acts of another in publishing, but he is under obligation to the plaintiff to take into account and into consideration what will be the natural and probable consequences of his act in putting a libel into circulation. To that extent he is responsible, and only to that extent.” The general proposition laid down is correct, no doubt, if rightly understood, and it was applied to libel, under what circumstances and with what meaning does not appear, in Miller v. Butler, 6 Cush. 71. But if applied to libel or slander without further explanation, it is likely to be misleading, and when put as a qualification of the ruling asked hardly can fail to be so. The meaning which naturally would be conveyed to the jury is, that, although a particular republication cannot be recovered for, damages may be enhanced by the general probability of unlawful republications. This is not the law. Wrongful acts of independent third persons, not actually intended by the defendant, are not regarded by the law as natural consequences of his wrong, and he is not bound to anticipate the general probability of such acts, any more than a particular act by this or that individual. Hastings v. Stetson, 126 Mass. 329, 331. Shurtleff v. Parker, 130 Mass. 293, 296. Hayes v. Hyde Park, 153 Mass. 514. Leonard v. Allen, 11 Cush. 241, 246.
Hxeeptions sustained.