​Complaining Work

​Defending Work

Felix Bernard & Johnny Black


Hear Sound Recording

View Partial Score

Jerome Kern


Hear Sound Recording

View Partial Score


Plaintiff Fred Fisher, owner of the copyright to "Dardanella" -- a song popular in the early 1920's -- claimed Jerome Kern misappropriated the accompaniment pattern of "Dardanella" and used it in his song "Ka-lu-a". Judge Hand awarded Fisher the minimum statutory damages, finding that Kern did infringe Fisher's work -- albeit most likely unconsciously, and with no material damage to the plaintiff.

The sound recording clip for "Dardanella" is from TIME-LIFE Records' 1979 re-release of Bix Beiderbecke recordings of jazz works from the 1920's.

[ Back to Top ]


Opinion by Judge Learned Hand

The copyright to the composition "Dardanella" covered the piece as a whole; there were not several copyrights for each part of it. Nevertheless the plagiarism of any substantial component part, either in melody or accompaniment, would be the proper subject of such a suit as this. To sustain it, however, more must appear than the mere similarity or even identity, of the supposed infringement with the part in question. In this lies one distinction between a patent and a copyright. One may infringe a patent by the innocent reproduction of the machine patented, but the law imposes no prohibition upon those who, without copying, independently arrive at the precise combination of words or notes which have been copyrighted. The plaintiff therefore concedes that it must show that Kern, the composer, used "Dardanella" as the source of his accompaniment.

The argument is a strong one. Not only is the figure in each piece exactly alike, but it is used in the same way; that is, as an "ostinato" accompaniment, whatever. Further, the defendants have been able to discover in earlier popular music neither this figure, nor even any "ostinato" accompaniment whatever. The fact that "Kalua" appeared shortly after "Dardanella" had faded out, and was written by one who had necessarily known it, as a musician knew it, makes it still more hard to assume any independent provenience for "Kalua." Can I suppose that such parallelism could be the result of coincidence only?

Mr. Kern swears that he was quite unconscious of any plagiarism, and on the whole I am disposed to give him the benefit of the doubt. For this I rely, not only upon the impression which he made upon me, but upon the insufficiency of the motive. I cannot agree that the accompaniment was at all as important to the success of "Dardanella" as the plaintiff would ask me to believe. I admit that it was a good bass, and helped; but I think the piece won its success substantially because of the melody. It is of course possible that Kern might have lifted it bodily, hoping to escape detection. However, he has an established place among composers of light opera, and has already succeeded more than once. Certainly detection would be a matter of some moment to him. No producer willingly invites the suits which follow musical piracy. Once convicted in such a case, Kern's market might suffer. With the profit small and price high, it seems to me unlikely that he should have set about deliberate plagiarism.

Whether he unconsciously copied the figure, he cannot say, and does not try to. Everything registers somewhere in our memories, and no one can tell what may evoke it. On the whole, my belief is that, in composing the accompaniment to the refrain of "Kalua," Mr. Kern must have followed, probably unconsciously, what he had certainly often heard only a short time before. I cannot really see how else to account for a similarity, which amounts to identity. So to hold I need not reject his testimony that he was unaware of such a borrowing. This I understand to have been Judge Knox's conclusion, reached upon the affidavits alone.

On the issue of infringement this conclusion is enough. The point is a new one, but I think it is plain. The author's copyright is an absolute right to prevent others from copying his original collocation of words or notes, and does not depend upon the infringer's good faith. Once it appears that another has in fact used the copyright as the source of his production, he has invaded the author's rights. It is no excuse that in so doing his memory has played him a trick. In an indictment under Copyright Act, § 28 (Comp. St. § 9549), the excuse might be a defense, since the infringement would not be willful; but it is seldom that a tort, as this is, depends upon the purpose of the wrongdoer. Therefore I find with the plaintiff on the issue of infringement.

The defendant next asserts that the proof of authorship is insufficient. Fisher swears that Bernard said he was the author, and there was no objection to that evidence on the score of hearsay, which is competent, if not objected to. He further said that Black later claimed to have contributed to the composition, and that he (Fisher) bought him out. In view of the fact that the insufficiency of the proof was not raised upon the trial, or until reargument (London v. Biograph Co., 231 Fed. 696, 145 C.C.A. 582), the plaintiff's case might therefore stand, even if proof of authorship were necessary dehors the certificate of registration.

However, the objection is independently bad under the act of 1909, because section 55 (Comp. St. § 9576) makes such a certificate, in fact offered in evidence, prima facie proof, among other things, of the name of the author. Bosselman v. Richardson, 174 Fed. 622, 98 C.C.A. 127 (C.C.A. 2), is made obsolete by this section. Indeed, Chief Justice Taney under the old act had held the copyright such prima facie proof in Reed v. Carusi, Fed. Cas. No. 11,642.

The most important point of law in the case is whether it is a defense that there was in the prior art substantially the same figure. First, I think there is a clear difference between its use at intervals, as in Wagner, Schumann, or Kummer, and its use by Bernard as an "ostinato." To the ear the two are tot coelo different, auditorially and emotionally, as different as eight notes when used alone and when used as part of a larger phrase. Musical melody is single, the sense of the earlier notes carrying over into those which succeed. Repetition is in substance the same in this respect, the effect upon the ear being entirely different when the figure is rolled over and over again. The only composition in the public domain which I should consider an anticipation, if this were a patent suit, is Landon's adaptation of Weber's "Mermaid Song."

The differences between Landon and the copyrighted "ostinato" are not enough to make the latter an "adaptation or arrangement" of the first, under section 6 (Comp. St. § 9522), had it been taken directly from Landon's work. There is a minimum of change which the law will disregard. Jollie v. Jaques, 1 Blatchf. 825, Fed. Cas. No. 7,437; At-will v. Ferrett, 2 Fed. Cas. 195, No. 640; Cooper v. James (D.C.) 213 Fed. 871. Yet there is no evidence to sustain the assertion that the Dardanella "ostinato" was in fact taken from Landon or from any other composition. For the purposes of this case it must be deemed to be original, if by original one means that it was the spontaneous, unsuggested result of the author's imagination. And so this case squarely raises the question whether it be a defense to a copyright that the precise work has independently appeared before it and is in the public domain.

Section 7 (Comp. St. § 9523) provides that "no copyright shall subsist in the original text of any work which is in the public domain." This is not new law, and means no more than that by taking such a text you may not get a copyright upon it. It is illustrated by such cases as Bullinger v. Mackey, 4 Fed. Cas. 649, No. 2,127, Kipling v. Putnam's Sons, 120 Fed. 631, 57 C.C.A. 295, 65 L.R.A. 873 (C.C.A. 2), and Eggers v. Sun Sales Co., 263 Fed. 373 (C.C.A. 2). The same rule was recognized in West Publishing Co. v. Lawyers' Co-operative Pub. Co. (C.C.) 64 Fed. 360, 25 L.R.A. 441, Mead v. West Publishing Co. (C.C.) 80 Fed. 380, Ed. Thompson Co. v. American, etc., Co., 122 Fed. 922, 59 C.C.A. 148, 62 L.R.A. 607, and Banks, etc., Co. v. Lawyers,' etc., Co., 169 Fed. 386, 94 C.C.A. 642, 17 Ann. Cas. 957. It has no application whatever to a work which is of original composition, because such a work is not the "original" text of any work in the public domain, but a second and equally "original" text of a work never published before its copyright.

The decisions relied upon by the defendant, other than those already cited, for the most part turn upon infringement, as, for example, Marks v. Feist (C.C.A.) 290 Fed. 959; Chautauqua School v. National School, 238 Fed. 151, 151 C.C.A. 227; Stevenson v. Harris (D.C.) 238 Fed. 432; Bachman v. Belasco, 224 Fed. 817, 140 C.C.A. 263; Haas v. Feist (D.C.) 234 Fed. 105; Vernon v. Shubert (D.C.) 220 Fed. 694; Howell v. Miller, 91 Fed. 129, 33 C.C.A. 407; Simms v. Stanton (D.C.) 75 Fed. 6; Burnell v. Chown (C.C.) 69 Fed. 993; Daly v. Palmer, 6 Blatchf. 256, Fed. Cas. No. 3,552; Reed v. Carusi, Fed. Cas. No. 11,642; D'Almaine v. Boosey, 1 Younge & C. 288; Chatterton v. Cave, L.R. 10 C.P. 572. In these cases it was indeed constantly discussed whether the defendant had copied the plaintiff's work, or whether he derived the alleged infringement, however similar, independently, but the similarity of his work with an earlier one in the public domain was used only as evidence in corroboration of his contention that that was its source. The cases do not decide that the copyrighted work is not valid, because there was a similar or identical work in the public domain. The quotations cited are to be read in the light of this rule governing infringement.

London v. Biograph Co., 231 Fed. 696, 145 C.C.A. 582 (C.C.A. 2), was a case which may have meant to go further, though the precise ratio decidendi is not wholly clear. The plaintiff had written a story based upon a very old and often repeated plot, modernized in its incidents. The infringing picture play contained the same plot, but only unimportant resemblances to the new details of the story. An injunction was denied, but the grounds are not certain. It nowhere appears that the play was copied from the story, and the familiarity of the plot was good reason for supposing that it was not. Judge Lacombe, on page 698 (145 C.C.A. 584), said:

"The plot is common property; no one, by presenting it with modern incidents, can appropriate it by copyrighting."

"Appropriate" would seem to indicate conscious lifting of it from its old form. Perhaps the court thought that the earlier existence of the plot was alone enough to defeat the copyright, but the opinion is not clear upon it, and does not seem to me to establish a rule which would, as I shall show, overthrow a doctrine fundamental in copyright.

Eichel v. Marcin (D.C.) 241 Fed. 404, was a similar case. The defendant had denied copying, and his denial was held sufficient on preliminary injunction. There was, indeed, some discussion of the fact that the common elements in both plots were old, and that an old plot could not be copyrighted; but nowhere was it clearly said that an old plot, independently conceived, might be copied with impunity. In my own decision of Stodart v. Mutual Film Corp. (D.C.) 249 Fed. 507, affirmed without opinion in 249 Fed. 513, 161 C.C.A. 439, I had the same kind of case. To be candid, I think that I understood London v. Biograph Co., supra, as going further than I now believe it need be taken to go; but my decision did not in the least involve the point now in question, because the plaintiff had a decree.

These cases go the furthest of any I have found to suggest that the rule of the patent law governing validity is to be carried over into copyrights. It appears to me very obvious that the rule as to infringement has, and indeed must have, as its correlative, the rule that originality is alone the test of validity. Any subsequent person is, of course, free to use all works in the public domain as sources for his compositions. No later work, though original, can take that from him. But there is no reason in justice or law why he should not be compelled to resort to the earlier works themselves, or why he should be free to use the composition of another, who himself has not borrowed. If he claims the rights of the public, let him use them; he picks the brains of the copyright owner as much, whether his original composition be old or new. The defendant's concern lest the public should be shut off from the use of works in the public domain is therefore illusory; no one suggests it. That domain is open to all who tread it; not to those who invade the closes of others, however similar.

If it were not so, I cannot see how possibly to account for well-established instances of valid copyright. In the first place, consider the necessary results of section 6 which authorizes the copyright of "compilations or abridgments, adaptations, arrangements, dramatizations, translations, or other versions of works in the public domain." Take, for example, two faithful compilations or translations. While it may be rare that they should be identical, obviously even that is possible over substantial parts. It could not be maintained that the earlier version destroyed the copyright of the later, and yet, if copyright be analogous to patents, this must result. Certainly, the labor of the second translator or compiler is not lost, so he do not use the work of the first.

Directories constitute a familiar instance of such compilations. No one doubts that two directories, independently made, are each entitled to copyright, regardless of their similarity, even though it amount to identity. Each being the result of original work, the second will be protected, quite regardless of its lack of novelty. But the best instance is in the case of maps. Here, if each be faithful, identity is inevitable, because each seeks only to set down the same facts in precisely the same relations to each other. So far as each is successful, each will be exactly the same. While I know no case which involves the point, Bowker says on page 255 (Copyright, Its History and Law):

"Two map makers, collecting at first hand, would naturally make the same map, and each would equally be entitled to copyright. In this respect copyright law differs from patent law, where a first use bars others from the same field."

In the field of photography the element of personal arrangement and selection may be reduced to very narrow compass. Pagano v. Beseler (D.C.) 234 Fed. 963 (Mayer, J.). Suppose a second photographer selects the same aspect of a public building as in that case. Certainly, if he do so independently, he will have as much title to protection as the first.

I can find no case in the United States which rightly presents the point, except perhaps Blume v. Spear (C.C.) 30 Fed. 629. That involved also a musical copyright, and one of the defenses was that an earlier air had been in fact the source of the copyrighted work. Wheeler, J., thought that the similarity was not close enough to indicate the provenience of the copyrighted work and overruled the defense. Apparently the defendant did not urge that the copyright composition must be new, and so the case hardly decides the point; but, if the law be otherwise, at least neither court nor counsel thought so.

However, the text-writers have dealt with it and agree, so far as I have found. Thus, Drone in his well-known work says, on page 205, in a passage cited with approval in Chautauqua School v. National School, 238 Fed. 151, 151 C.C.A. 227, supra (C.C.A. 2):

"Works alike may be original. It is not essential that any production, to be original or new within the meaning of the law of copyright, shall be different from another. Whether the composition for which copyright is claimed is the same as or different from, or whether it is like or unlike, an existing one, are matters of which the law takes no cognizance, except to determine whether the production is the result of independent labor or of copying. Two or more authors may write on the same subject, treat it similarly, and use the same common materials in like manner or for one purpose. Their productions may contain the same thoughts, sentiments, ideas; they may be identical. Such resemblance or identity is material only as showing whether there has been unlawful copying. In many cases the natural or necessary resemblance between two productions which are the result of independent labor will amount to substantial identity. … But, notwithstanding their likeness to one another, any number of productions of the same kind may be original within the meaning of the law, and no conditions as to originality are imposed upon the makers, except that each shall be the producer of that for which he claims protection."

And again, at page 208 of the same work, a passage not quoted in the case just mentioned:

"In all cases whatever may be the kind or the character of the work for which protection is claimed, the true test of originality is whether the production is the result of independent labor or of copying."

Weil, in his work in Copyright, says at section 998:

"If A. produce identically the same work as B. by independent thought, in good faith, without hearing of or seeing these works, both A. and B. would be entitled to individual copyrights of their individual works."

And again, at page 184 of the edition of 1917:

"All that is required to afford the needed degree of originality is independent thought and an absence of copying the work of others. To this somewhat limited and nonexacting extent writings, to be copyrightable, must be original."

And on page 186:

"Since such materials are open to any one to use as he sees fit, the result may be achieved independently, and, if so, each will be entitled to copyright in his resultant work. This result, inevitable from the very nature of the usual and permissible method for manufacturing the great bulk of copyright works, is still, from time to time, overlooked by judges when considering the exceptional, novel, or palpably meritorious work before them. Copyright law, accordingly, differs radically from patent law, as the monopoly given by the latter is unqualified, while the right given by the former is preventative in many respects, rather than positive."

I conclude, therefore, that the existence of Landon's "ostinato," though substantially the same as the "Dardanella" accompaniment, did not invalidate the copyright pro tanto, there being no evidence, or, indeed, any reasonable possibility, that it was the source of Bernard's conception. There remains only the question of relief.

The plaintiff may, of course, take the usual injunction, though under the circumstances it will apparently be of no service. As for damages, it seems to me absurd to suggest that it has suffered any injury. "Dardanella" had faded out before "Kalua" appeared; but, if it had been at the peak of its popularity, I do not believe that the accompaniment to the chorus of "Kalua" would have subtracted one copy or one record from its sales. The controversy is "a trivial pother" (Hugh, J., dissentiente in Jewelers', etc., Co. v. Keystone Pub. Co. (C.C.A.) 281 Fed. 83, 95), a mere point of honor, of scarcely more than irritation, involving no substantial interest. Except that it raises an interesting point of law, it would be a waste of time for every one concerned.

However, section 25 (Comp. St. § 9546) fixes a minimum of $250, which is absolute in all cases. Since Westermann Co. v. Dispatch Co., 249 U.S. 100, 39 Sup. Ct. 194, 63 L. Ed. 499, any doubts reserved in Hendricks Co. v. Thomas Pub. Co., 242 Fed. 37, 154 C.C.A. 629 (C.C.A. 2), are laid. Therefore I must and do award that sum as damages. The plaintiff is likewise absolutely entitled to a full bill of costs, but I will make no allowance for counsel fees, since that is discretionary. Such victories I may properly enough make a luxury to the winner.

Settle decree on notice.


[ Back to Top ]