Jiri Mucha v. Charles King

792 F.2d 602

• Court: Court of Appeals for the Seventh Circuit
• Decided: June 23, 1986
• Docket: 85-2773
• Status: Published

Opinion

POSNER, Circuit Judge.

This suit raises interesting legal questions, mainly of the common law of bailments, set against an exotic factual background involving the international art trade. The dispute is over ownership of a painting that, having been given away seven years ago, may today be worth several hundred thousand dollars. The extraordinary if irrelevant coincidence that this is the second case in little more than a year to confront the same three judges of this court with a dispute rooted in the movement known as Art Nouveau should not go unremarked. See Piarowski v. Illinois Community College Disk 515, 759 F.2d 625, 627 (7th Cir.1985); Buffet-Challié, The Art Nouveau Style 9 (1982). In Piarowski the artist was an imitator of Aubrey Beardsley; in this case it is Alphonse Mucha. A Czech who lived for many years in Paris, and at the turn of the century was one of its most prominent resident artists, Mucha figures importantly in the history of modern art. In some quarters he is regarded as the originator of Art Nouveau, which indeed for a time was called “le style Mucha.” He is best known for posters and (like Beardsley) illustrations. See Rennert & Weill, Alphonse Mucha: The Complete Posters and Panels (1984).

In 1983, in a federal district court in Chicago, Jiri Mucha, Alphonse Mucha’s only son and a citizen of Czechoslovakia, sued Charles King, basing federal jurisdiction on 28 U.S.C. § 1332(a)(2). The suit sought the return of a painting that Alphonse Mucha had painted in about 1904, and that as late as 1980 was believed lost. See Mucha: 1860-1939, at no. 307 (Editions des musées nationaux, Paris 1980). A seven foot by seven foot oil painting known as “Quo Vadis” or “Petronius and Eunice,” and loosely based on the novel that many years later was made into a movie by MGM, see Sienkiewicz, Quo Vadis (1955 [1896]); Halliwell’s Film and Video Guide 665-66 (4th ed. 1983), it depicts with the sinuous linearity characteristic of Art Nouveau a man and an adoring female slave, in classical pose. The man is a self-portrait of Alphonse Mucha.

In 1979 the Newcomb-Macklin art gallery in Chicago, to which Alphonse Mucha had consigned the painting in 1920, gave it away to a “hot tubs” merchant who in turn sold it to an art dealer from whom King bought the painting in 1981 for $35,000 in cash and merchandise. The painting was in poor condition, and King hired an expert to restore it at a price of $16,500, of which $8,500 has been paid. King testified that the painting might fetch as much as $800,-000 today; Jiri Mucha’s estimate is $100,-000. The painting is in the restorer’s custody pending the final decision of this suit and the payment of the balance of her fee.

After a bench trial, the district judge ruled in favor of the plaintiff and ordered King to return the painting and Mucha to reimburse King for the $8,500 in expenses that King had incurred in restoring it. King appeals, arguing that:

1. Alphonse Mucha and his heirs were dispossessed of the painting as a result of its conversion by the Newcomb-Macklin art gallery in Chicago more than five years before this suit was brought; hence Jiri Mucha’s right to recover the painting was extinguished by the five-year Illinois statute of limitations applicable to suits for the possession of personal property. Ill.Rev. Stat. ch. 110, 1113-205.

2. Jiri Mucha abandoned all rights in the painting to the gallery in 1958.

3. Jiri Mucha is at most a 50 percent owner of the painting; King, if he doesn’t own the whole thing (his first two defenses), owns the other 50 percent.

There are two preliminary issues to be got out of the way. The first is choice of law. The district judge said, “As this Court’s jurisdiction is grounded upon the parties’ diversity of citizenship, Illinois law governs this case.” This implies that in a diversity suit the substantive law of the state where the federal court is located always governs. Not so. The federal court in a diversity suit applies the choice of law rules of the forum state, and those rules may or may not make the substantive law of that state the governing law for the suit. See Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496-97, 61 S.Ct. 1020, 1021-22, 85 L.Ed. 1477 (1941); Casio, Inc. v. S.M. & R. Co., 755 F.2d 528, 531 (7th Cir.1985). But this is a harmless error. Where, as in this case, the parties either explicitly or by implication agree to be governed by the substantive law of the forum state, their agreement will be enforced. See id. at 531.

The other preliminary issue is the standard of appellate review, on which the appellant’s brief is silent. It conspicuously lacks any references to the “clearly erroneous” rule that governs our review of findings of fact in a bench trial. Fed.R.Civ.P. 52(a). When pressed at argument on this omission, counsel noted that the facts are uncontested and that decision turns on the legal effect of documents, which he characterized as an issue of law. This is a recurrent misunderstanding and it is worth taking a moment to try to straighten the matter out.

King’s counsel was using the word “facts” to describe historical events. This is a common usage. It is a fact that Alphonse Mucha consigned “Quo Vadis” to Newcomb-Macklin in 1920, that Mucha died in 1939, that in 1979 Newcomb-Macklin gave away “Quo Vadis” for nothing, and that in 1981 King bought it from another dealer. None of the material “facts” in this case is in dispute. The only question is their legal significance — a question of law, King argues, and therefore not governed by the clearly-erroneous rule but by a different rule, deferential but less so: that a district judge’s view on a question of the law of the state in which he sits is entitled to substantial weight. See, e.g., Goldstick v. ICM Realty, 788 F.2d 456, 466 (7th Cir. 1986), and cases cited there.

But one cannot answer the question what is a “fact” without first considering the purpose of the question. The purpose here is to draw the line between the trial judge’s responsibility and our responsibility. Rule 52(a) of the Federal Rules of Civil Procedure, in providing that findings of fact made by the district judge in bench trials (and, by interpretation, in all other settings in which district judges are called on to make such findings, Casio, Inc. v. S.M. & R. Co., supra, 755 F.2d at 530) must be upheld on appeal unless found to be clearly erroneous, assigns to the trial judge the responsibility of determining not only the historical events that are relevant to how the case should be decided but also the legal significance of those events. There are exceptions for some constitutional issues, see, e.g., Miller v. Fenton, — U.S. -, 106 S.Ct. 445, 88 L.Ed.2d 405 (1985), and cases cited there, but they are immaterial to this case. Here the judge had to determine whether the historical events showed that Alphonse Mucha possessed “Quo Vadis” at specified times. Although possession is a legal concept, whether particular “facts” show possession is itself a “fact” for purposes of separating the trial judge’s function from our own. Negligence is another such fact. Facts of this sort, which are found by applying a legal standard to a descriptive or historical narrative, are governed by the clearly-erroneous rule.

Admittedly there is much waffling on this point in the cases, which led the Supreme Court to note in Pullman-Standard v. Swint, 456 U.S. 273, 289 n. 19, 102 S.Ct. 1781, 1790 n. 19, 72 L.Ed.2d 66 (1982) (itself a case involving an inferential fact— discriminatory intent — rather than a legal characterization such as possession), that there “is substantial authority in the Circuits on both sides of” “the much-mooted issue of the applicability of the Rule 52(a) standard to mixed questions of law and fact.” In particular, the Second Circuit had long adhered to the view that a finding of negligence is not subject to the clearly-erroneous standard. See Mamiye Bros. v. Barber S.S. Lines, Inc., 360 F.2d 774, 776-78 (2d Cir.1966) (Friendly, J.), and cases cited in 5A Moore’s Federal Practice ¶ 52.05[1], at p. 52-117 n. 14 (2d ed. 1986). But most courts treat legal characterizations (negligence, possession, ratification, principal place of business, etc.) as facts to which the clearly-erroneous standard applies. See 5A id., 11 52.05[1], at p. 52-113 n. 11; 9 Wright & Miller, Federal Practice and Procedure §§ 2589-90 (1971); Piper Aircraft Corp. v. Wag-Aero, Inc., 741 F.2d 925, 936-36 (7th Cir.1984) (concurring opinion). That is certainly the view of this circuit. See, e.g., Smith v. Chesapeake & Ohio Ry., 778 F.2d 384, 390 (7th Cir.1985); Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 385 (7th Cir.1984); Continental Group, Inc. v. Lincoln Land Moving & Storage, Inc., 710 F.2d 368, 372 (7th Cir.1983); Riverway Co. v. Trumbull River Services, Inc., 674 F.2d 1146, 1150 (7th Cir.1982); Glenview Park Dist. v. Melhus, 540 F.2d 1321, 1323 (7th Cir.1976).

Further support for it is found in the Supreme Court’s recent decision in Anderson v. City of Bessemer City, 470 U.S. 564, 105 S.Ct. 1504, 1512, 84 L.Ed.2d 518 (1985), which emphasized that the clearly-erroneous rule applies “even when the district court’s findings do not rest on credibility determinations, but are based instead on physical or documentary evidence or inferences from other facts.” See also Scandia Down Corp. v. Euroquilt, Inc., 772 F.2d 1423, 1427-29 (7th Cir.1985). The language we have quoted does not, it is true, address the question when is a particular finding one of fact and when one of law, but does make clear that appellate review is not plenary simply because the court of appeals may in the particular case have had the same access to facts as the district court. In this case, for example, the only live witness was King, and his testimony was peripheral; the main evidence was documentary, consisting of correspondence plus transcripts of depositions. As we suggested in Brock v. TIC Int’l Corp., 785 F.2d 168, 171 (7th Cir.1986), Bessemer City teaches that the main reason for appellate deference to the findings of fact made by the trial court is not the appellate court’s lack of access to the materials for decision but that its main responsi bility is to maintain the uniformity and coherence of the law, a responsibility not engaged if the only question is the legal significance of a particular and nonrecurring set of historical events. This set will not recur — that is for sure.

Once Rule 52(a) is understood to rest on notions of the proper division of responsibilities between trial and appellate courts, see Miller v. Fenton, supra, 106 S.Ct. at 451-52, rather than just on considerations of comparative accessibility to the evidence, the concern expressed by Judge Friendly in Mamiye — that it is shocking to imagine that two identical cases might be decided in opposite ways by two different district judges, yet each decision be affirmed — subsides. Review is deferential precisely because it is so unlikely that there will be two identical cases; the appellate court’s responsibility for maintaining the uniformity of legal doctrine is not triggered.

The issue raised by this appeal, therefore, is not when Alphonse Mucha or his heirs lost possession of “Quo Vadis” but whether the district judge’s determination that this did not happen till 1979 at the earliest, and therefore within the statute of limitations, is clearly erroneous. To answer this question we shall have to trace the history of what — if the district judge was right — was a 59-year bailment.

In 1920 Alphonse Mucha, who is conceded to have owned “Quo Vadis” at the time, consigned it along with more than 20 other paintings and many other art works to the Newcomb-Macklin gallery. As explained in a letter to Mucha from Thomas Dunbar of the gallery, “these paintings, posters [‘Quo Vadis’ is described in the letter as a ‘very large square poster’] and drawings, are consigned to Mr. John Suster [the owner of the gallery] on sale at the net prices stated on the previous page.” The net price listed for “Quo Vadis” is $10,000. That was a very high price for the time, but Mucha was a very famous artist — more so than today. He was particularly well known in Chicago, having among other things lectured at the Art Institute of Chicago. See Rennert & Weill, supra, at 24. Most of the items listed in Dunbar’s letter also carried suggested sale prices, but apparently these were not intended to bind Suster.

When the letter was sent, Mucha’s pictures were on exhibition at the Brooklyn Museum. (This celebrated exhibition is said to have been seen by 600,000 persons. Id. at 352.) The letter states that when the pictures are returned to the Newcomb-Macklin gallery in Chicago, Suster “becomes responsible for them as against theft and damage,” and “it is also understood that Mr. Suster agrees to defray all express and cartage charges, storage etc. after they have been returned to him at his place in Chicago.” In other words, Suster would recover his costs in the difference between the prices at which he sold to the public and the net prices he was required to remit to Mucha. This is an unusual form of consignment in the art world. Usually the gallery takes a fixed percentage of the gross sales price, such as 40 or 50 percent. DuBoff, The Deskbook of Art Law 785 (1977). Under his arrangement with Mucha, Suster if he anticipated that the paintings would appreciate would have an incentive to store them rather than sell them, since the appreciation would benefit him exclusively. And as a matter of fact, several months after this letter was written Suster placed many of the items, including “Quo Vadis,” in storage in a warehouse in Chicago rather than exhibiting them.

The next document in the case is a letter that Suster wrote Mucha in 1929, by which time Mucha had returned to Czechoslovakia. The letter, which begins “My dear Brother,” apparently was prompted by a cablegram from Mucha to Suster instructing the latter to ship all of Mucha’s art works in Suster’s custody to Oslo for an exhibition. (Suster and Mucha were not related, though Suster was of Czech extraction; the “My dear Brother” mode of address may reflect the fact that both were free-masons.) Suster’s letter promises to do this after exhibiting some of the items in an attempt to sell them, and contains a detailed discussion of the “stuff,” which includes many items (such as 128 sketches and some furniture) not listed in the 1920 letter. King fastens on the following passage: “The picture that I am going to keep for all the trouble (no trouble, Mr. Mucha; it was all a pleasure!) but as you said, is the blue picture — the two pictures in blue____ It is not your wife’s portrait, much as I would like to have it.” King argues that this shows a breach of the consignment agreement and therefore a termination of the bailment. Suster had promised to pay all storage expenses; now he was confiscating two of Mucha’s pictures to recoup those expenses. Neither of the pictures was “Quo Vadis,” but King’s point is that the breach of contract, by ending the bailment, dispossessed Mucha and set the five-year statute of limitations on a suit for replevin running. King appeals to the principle that “where there is no limit of time or definite time fixed for the return, or termination is to be at the will of the bailor, there must be a demand made for the return and a refusal or some other act on the part of the bailee of which the bailor has notice which is hostile or inconsistent with the bailment and such as would constitute a denial of the bailment and a conversion of the property.” Slack v. Bryan, 299 Ky. 132,137,184 S.W.2d 873, 876 (1945).

We find King’s argument hard to understand. Even if the premise is conceded— that Suster was breaking his contract — -the conclusion does not follow that he had converted “Quo Vadis.” On the contrary, he said in the letter that he was sending all of Mucha’s “stuff,” therefore including “Quo Vadis,” to Oslo, where Mucha wanted it sent. There was no dispossession. The premise of the argument is also questionable. From salutation to farewell and postscript, the letter fairly oozes warmth and friendship. There is no suggestion that Suster was breaking his nine-year-old contract. The language in which he introduces the topic of the picture to be retained —“The picture that I am going to keep for all the trouble (no trouble ...) but as you said ...” (emphasis added) — implies that Mucha had told Suster to keep a picture for his trouble, and an earlier letter from Mucha is consistent with this interpretation.

Although Suster may actually have been trying to slip one over on Mucha, who living in Czechoslovakia may not have had much practical ability to enforce the contract, nine years had passed since the consignment had been made and apparently Suster had sold none of Mucha’s art and had therefore borne the entire expense of storage without compensation. This would not, however, justify Suster in unilaterally altering the contract, especially when, as noted earlier, the contract gave him an incentive to keep Mucha’s art work in storage in the expectation of a future increase in value from which Suster alone would benefit. That fact makes the principle that the bailee may be liable to the bailor for extraordinary expenses in taking care of the bailed goods, see Tate v. Sutherland, 27 Wash.2d 573, 577-78, 179 P.2d 313, 315 (1947); Williamson v. Phillipoff, 66 Fla. 549, 553-54, 64 So. 269, 271 (1914); Lawson, The Principles of the American Law of Bailments § 25, at p. 53 (1895), doubly irrelevant: the storage expenses were not extraordinary and they could have been predicted from the terms of the bailment. Moreover, the fact that Mucha did not object to Suster’s keeping a picture would not create a binding modification of the consignment agreement, for Suster offered no fresh consideration, and certainly in 1929 a contract modification not supported by consideration would not have been enforceable. See, e.g., Lindsey v. Rosen, 255 Ill. App. 21, 26, rev’d on other grounds, 335 Ill. 402, 167 N.E. 89 (1929); Smith v. Gray, 316 Ill. 488, 496, 147 N.E. 459, 463 (1925); Farnsworth, Contracts § 4.21 (1982).

The essential point, though, is that the alleged breach of the consignment agreement that defined the bailment, while possibly if improbably a conversion of the blue pictures, was not a conversion of “Quo Vadis.” Far from trying to retain “Quo Vadis,” Suster was assuring Mucha that all of the remaining works, of which “Quo Vadis” was one, would be shipped to Oslo in accordance with Mucha’s directions. Not every breach of a contract of bailment entitles the bailee to sue for the value of the bailed goods. In Johnson v. Weedman, 5 Ill. 495 (1843) (a case in which Abraham Lincoln represented the winning party), a horse had been bailed to the defendant, who rode the horse contrary to the terms of the bailment; the horse died after, but not because of, the ride. The plaintiff sued for the value of the horse, alleging that the defendant had converted it. The court held that the plaintiff was entitled only to nominal damages, since the breach of contract had caused no injury to the horse. It would seem to follow that if the breach of the bailment contract concerns bailed good A, the bailor cannot recover damages for conversion of unharmed bailed good B, which in this case would be “Quo Vadis.” In Brandenburg v. Northwestern Jobbers’ Credit Bureau, 128 Minn. 411, 151 N.W. 134 (1915), a case much like this one in principle, the bailee accidentally sold a small number of the bailed goods; this was held not to be a conversion of the entire stock of bailed merchandise.

Nothing was sent to Oslo in response to Mucha’s request. In 1931 he wrote Suster directing him to turn over certain paintings to a Sam Woods, who would pick out which paintings he wanted. Woods picked out ten paintings and took them back to Mucha in Prague, where Woods was a consular officer. Next comes a letter from Mucha to Suster, written in 1935, in which Mucha explains that since he has reached the autumn of his life (having just turned 75) it is necessary to put his affairs in order. In Suster’s English translation of the Czech original, Mucha says that “to close the account with all my life” he is planning a retrospective exhibition in Paris next year. “For this exhibition, I shall need all things that are up to this time in your kind hands and under your precious care..... From the list which I have before me, I can see all the pictures, studies and drawings. Some of them I have already received, but the rest are still there in your way.” The letter, which does not mention “Quo Vadis,” says that Mucha had asked Woods to pick up Mucha’s art works on a recent trip to the United States but that Woods had failed to do so because Woods had gotten no answer to a letter he had written Suster. “You were perhaps on your summer travels. I hope you didn’t meet with any mischief.” The letter adds that Mucha’s son, Jiri, will probably visit the United States in October.

Replying to this letter, Suster wrote, “I am just as anxious to go to work and get that stuff to you, as you are to get it; but, the idea is that I am afraid of the duty if I ever should let loose of them pictures and they got into the hands of the American or Bohemian government. There certainly will be some duty to pay, which I don’t think you should have to pay, and the only way to eliminate that, I think, is that somebody must come here.” The reference to duty is puzzling. As far as we can determine, the American government did not impose a duty on art exports, and it seems unlikely that the Czech government would have imposed a duty on a famous Czech artist’s bringing his own work back into his country — but who knows? King does not argue that Suster was lying about the possibility of heavy duties.

Suster goes on in the letter to urge Mucha to have his son pick up the stuff when he comes to the United States. The rest of the letter, like the 1929 letter, describes the works that Suster has in storage and assures Mucha that they are being well looked after. “Quo Vadis” is not mentioned.

Again King argues that this letter is a repudiation of the bailment. Suster was unwilling to pay any duty that might be charged on art works sent to Czechoslovakia; King argues that such duty was encompassed by the reference in Dunbar’s letter to the charges that Newcomb-Macklin had agreed to defray. But there is no firm indication that those charges included import duties. The contract contemplated that the paintings and other art works would be sold in Chicago or elsewhere in the United States and the artist’s net price remitted, with Newcomb-Macklin defraying its expenses out of the difference between the artist’s price and the retail price. By this time, moreover, the bailment was 15 years old. The Great Depression had intervened and, as a previous letter from Suster to Alphonse Mucha makes dramatically clear, had caused the collapse of the American art market (as Suster put it, you couldn’t even sell a dollar for 75 cents). Maybe in these circumstances the payment of duty would have been an extraordinary expense for which the bailor would be responsible, according to the authorities cited earlier. In any event, as we said before, even if Suster was breaking his contract of bailment by refusing to pay duty it would not follow that he was converting “Quo Vadis.”

King has a separate argument — that Suster’s failure to ship Mucha’s works to Oslo as promised in 1929 was a conversion. A letter from Suster to Mucha in 1929 suggests that Mucha himself had countermanded the instructions but had done so out of concern for the cost of transportation. This could indicate that Mucha had been coerced to drop his demand by Suster’s refusal (maybe in breach of the consignment agreement) to pick up the cartage charges. But Mucha’s countermanding letter does not support this interpretation— not clearly anyway. In it he says, shortly before the Oslo exhibition, that Suster should “not hurry at this time with the shipment at all because I could need the stuff only next spring. So I would think it would be preferable to try to sell of them as much as possible and wait until we would find an occasion of somebody going to Europe who would take the boxes with him, so it could make the shipping.” None of Mucha’s letters contains any hint of recrimination. Apparently he was content to have Woods pick up some of the paintings in person two years after Oslo. Even in 1935 he seems to have been content to wait till his son visited the United States in a few months and brought back the rest of the paintings and other items, for there is no indication that Mucha wrote back to Suster complaining about Suster’s refusal to ship the items.

Jiri Mucha did not come to the United States. More years passed. In March 1939 Czechoslovakia passed under the Nazi yoke. Alphonse Mucha, a fervent Czech nationalist, was arrested and questioned by the Gestapo, though released. A friend of his, Mrs. Friedrich, wrote Suster inquiring about Mucha’s art works; she explained that she was writing on Mucha’s behalf because the Muchas’ mail was being heavily censored. Suster’s successor as the head of the Newcomb-Macklin gallery — his son-in-law, C. Dewey Imig — replied to Mrs. Friedrich in a letter that reviewed the history of the gallery’s dealings with Mucha. Imig explained that in 1935 Suster had received instructions from the Czech consul, Smetanka, to ship the balance of sketches, lithographs, and drawings to Prague, and that this had been done, but that Smetanka had not wanted to ship “the two pieces of furniture and two large murals,” of which, judging by the dimensions given, one must have been “Quo Vadis.” Imig’s letter also explains that Suster had kept another painting “to take care of such expense incurred in all of these storage charges,” which had been accumulating at $6 per month since 1921. Imig told Mrs. Friedrich that he didn’t think the murals were salable. He asked for instructions regarding the disposition of the four items, as “we would prefer not to send them back to storage as we do not care to incur any more expense.” He had withdrawn them from storage to check them over before replying to her letter.

Alphonse Mucha died in 1939, survived by his wife Marie (Maruska), his son, Jiri, and a daughter, Jaroslava. Marie was his residuary legatee and inherited “Quo Vadis” if (as we believe) Alphonse Mucha owned it when he died. Marie wrote Suster in 1940, asking him to send her husband’s paintings to New York. Imig responded, saying he had already dispatched them in response to Mucha’s earlier directions. He apparently thought she was referring to the paintings that Woods had picked up and not to the residue, apparent ly of no value, that remained in storage, having been returned there after the letter to Mrs. Friedrich was written.

The next significant document is a letter from Imig to Jiri Mucha written in 1958. Much water had flowed under the bridge since 1939. Jiri Mucha had left Czechoslovakia for France in 1939, shortly before his father’s death, had returned briefly for his father’s funeral, and had then returned to France and later joined England’s Royal Air Force. When World War II ended he returned to Czechoslovakia and moved in with his mother. He was a novelist and was not interested in art at this time. His father’s voluminous papers and correspondence, including the original letter from Dunbar in 1920 listing “Quo Vadis” among the items consigned to the Newcomb-Macklin gallery and the letters from Suster, mouldered in unopened boxes and drawers in the Mucha home. In the early 1950s Jiri Mucha was imprisoned by Czechoslovakia, allegedly for being an American spy. But despite everything, he was aware that the Newcomb-Macklin gallery might still have some of his father’s work, for in 1958 he wrote Imig that “I gather that there have been some paintings of my father, ... in storage with you for many years,” and inquiring “what the position now is.”

Imig replied that Alphonse Mucha had left paintings with Suster back in 1921, that they had been stored at Newcomb-Macklin’s expense, that most of them had been turned over to Woods, that the only paintings left were “some large murals that are still in storage and a painting called ‘The Kiss,’ ” that Newcomb-Macklin had incurred a total expense of $4,000 in storing these items, that “we don’t know what to suggest to you as we know it would be impossible to sell these murals in the United States,” but that “I feel that [‘The Kiss’] should belong to us to cover at least a part of the charges I have referred to in this letter.”

King argues that this letter again shows a repudiation of the bailment. The argument is not persuasive. Imig put forth the retention of “The Kiss” as a suggestion, and Jiri Mucha apparently so understood it, for he promptly replied: “I have to appologise for all the trouble you had with the canvasses. I think you are right suggesting that you should keep what is left in order to cover somehow your expenses. As there will be my Fathers centenary in two years and an important publicity drive is planed for that occasion not only in this country, but also abroad, especially in France, I suppose it might happen, that the murals might then fech a decent sum, should you offer them to some of the Czech comunities in the United States.”

This letter from Jiri Mucha provides the basis for the distinct argument that he abandoned all claim to the murals, including “Quo Vadis” — that he gave them to Imig in compensation for Imig’s storage costs, being embarrassed that Newcomb-Macklin had had to store these things for so many years without any recompense, because they were unsalable. One objection to this argument is that “Quo Vadis” is not a mural, that is, a wall painting. Both Imig and Jiri Mucha may have labored under a misapprehension as to what remained in storage, and Mucha could hardly have been abandoning property he didn’t know was in Imig’s possession. Abandonment is a voluntary relinquishment of rights, see, e.g., Menzel v. List, 49 Misc.2d 300, 305, 267 N.Y.S.2d 804, 809-10 (Sup.Ct.1966), aff'd as modified, 28 A.D.2d 516, 279 N.Y.S.2d 608 (1967) (per curiam), rev’d, 24 N.Y.2d 91, 298 N.Y.S.2d 979, 246 N.E.2d 742 (1969) (reinstating trial court’s award of damages), and thus implies — unless the alleged abandonment takes the form of a quitclaim deed, an issue we shall come back to in a moment — knowledge of those rights. But this point is not decisive, for in the earlier correspondence Suster and Mucha had apparently referred to “Quo Vadis” as a mural, no doubt because of its size. Dunbar in 1920 had called it a poster.

A better point is that Mrs. Mucha was still alive in 1958 and was the owner of the murals — not Jiri Mucha. Imig may have had reason to think Jiri Mucha authorized to dispose of his father’s property, but a cautious dealer would demand more evidence of title. This would not matter, however, if Mucha’s statement, “keep what is left,” should be understood to mean that Mucha was giving up whatever rights he might have (or later acquire by inheritance from his mother) in whatever art works remained in Newcomb-Macklin’s custody. That would certainly bar this suit. But it is not a necessary interpretation. Bemember that Imig had not asked to be allowed to keep everything. He just wanted “The Kiss.” Mucha’s letter refers to Imig’s having suggested that he keep what is left, but that is not what Imig had suggested. Since Jiri had lived in England during World War II, his English probably was excellent in 1958 (the deposition taken for the trial of this case indicates that it is excellent today). But it was not perfect, and it is not certain that there was perfect communication between Imig and Jiri. Although the words suggest otherwise, Jiri may have intended just to allow Imig to keep whatever Imig wanted to keep — and all Imig wanted to keep was “The Kiss”— or whatever had a current market value, which again was “The Kiss.” The district judge did not have to read Jiri’s letter as a release of all claims to his father’s art works held by Newcomb-Macklin.

This conclusion is supported by events that occurred 15 years later. In the 1960s Jiri was commissioned to write a book about his father. This caused him to make a study of his father’s work and to organize exhibitions of it. Meanwhile his mother had died in 1959, devising her property in equal shares to Jiri and his sister. In 1973 Jiri was in New York for an exhibition of his father’s work. Imig called him there to say that he had found another of those works, which he offered to return to Jiri upon payment of $5,000 in storage charges. Jiri came to Chicago to inspect the work, which was the painting “Harmony,” and he agreed to take it and to pay the $5,000. While he was there Imig told him to look around the gallery and see whether any more of his father’s art works were still there; apparently they had been taken out of paid storage and put in the gallery’s storage room. Jiri looked around carefully but could find nothing.

It seems plain from this incident that Jiri had not abandoned his father’s art works in 1958. If he had, Imig would not have offered to return “Harmony” upon payment merely of the accumulated storage charges (Jiri testified that the painting was worth $50,000 at the time) and to return any other art works of Alphonse Mucha that might be lying around. The behavior of Jiri and Imig in 1973 illuminates the exchange of letters in 1958 and supports a conclusion that there had been no abandonment — no quitclaim — then. But it also supports King’s strongest argument, which is that at some time between 1958 and 1973 (if not earlier) the Newcomb-Macklin gallery surely converted “Quo Vadis,” that Jiri should have realized this, and therefore that the statute of limitations began to run more than five years before this suit was brought.

King concedes that the statute of limitations does not begin to run until the bailor discovers or should have discovered that the bailee has converted the bailed property. Maybe he should not have conceded this. The applicable statute, unlike a number of other Illinois statutes of limitations, see, e.g., Ill.Bev.Stat. ch. 110, ¶¶ 13-212, -213, -214, does not mention discovery. The Illinois courts have, it is true, applied a discovery rule to many different causes of action governed by statutes of limitations that do not mention discovery, see, e.g., Nolan v. Johns-Manville Asbestos, 85 Ill.2d 161, 166-69, 52 Ill.Dec. 1, 3-4, 421 N.E.2d 864, 866-67 (1981), but they have said that the rule will not be applied in every case: specifically, not if problems of proof created by the passage of time outweigh the hardship to a plaintiff who could not as a practical matter have sued any earlier than he did. See id. at 167-68, 52 Ill.Dec. at 4, 421 N.E.2d at 867; Gates Rubber Co. v. USM Corp., 508 F.2d 603, 606-15 (7th Cir.1975). Although the tide in Illinois is running strongly in favor of the discovery rule, see, e.g., Tucek v. Grant, 129 Ill.App.3d 236, 84 Ill.Dec. 603, 472 N.E.2d 563 (1984); Kinsey v. Scott, 124 Ill.App.3d 329, 339-40, 79 Ill.Dec. 584, 591-92, 463 N.E.2d 1359,1366-67 (1984), it must remain a matter of speculation whether an Illinois court would apply it in a case such as this, given the antiquity of the bailment and the fact that many of the principals are dead. (Imig was still alive at the time of suit but his memory had faded.) It is worth noting, however, that the Supreme Court of New Jersey, in interpreting a discovery rule apparently much like Illinois’, has held that an artist (and we suppose, the artist’s heir) is entitled to invoke the discovery rule when diligently seeking the recovery of a lost or stolen painting. O’Keefe v. Snyder, 83 N.J. 478, 497-99, 416 A.2d 862, 872-73 (1980).

Be all this as it may, King argues only that Jiri Mucha should have suspected conversion (or some equivalent misconduct entitling the bailor to demand immediate possession, cf. Refrigeration Sales Co. v. Mitchell-Jackson, Inc., 770 F.2d 98 (7th Cir.1985)) more than five years before he sued. An argument can be made that by 1973 Jiri, who by 1966 knew his father had painted a large painting known as “Quo Vadis” (see Mucha, Alphonse Mucha: His Life and Art 327 (1966)) and should have suspected that it had been consigned to Newcomb-Macklin which had then lost or otherwise disposed of it, was under a duty if he wanted to preserve his rights to the painting to inquire more closely into its peregrinations since 1920. But we do not think that the district judge so clearly erred in holding that Jiri Mucha did not have such a duty in the circumstances that we can reverse his determination though we might disagree with it as an original matter.

The rest of the story is swiftly told. In 1979 Imig, by now an old man, decided to close down Newcomb-Macklin and liquidate its contents. Rupprecht, of Great Lakes Hot Tubs, Inc., bought a fan and an ice box from Imig for $65. On his way out of the gallery Rupprecht asked Imig whether he could have some rolled-up paintings he had seen in the basement. Imig said he could. A year later Rupprecht sold “Quo Vadis,” which was one of the paintings, for $150 to a Chicago art dealer named Tome, who did business under the name “Fly-by-Nite Galleries.” Now enters the defendant, King, a dealer in antiquities who has, however, no place of business, but works out of a shopping bag. He had done business before with Tome and wanted to buy “Quo Vadis.” A price of $35,000 was agreed on, which King paid in cash, silver candlesticks, and other silver objects over a period of a year, completing the transaction in 1981.

King does not argue that he can prevail against Mucha by virtue of the doctrine that allows a merchant entrusted with goods to transfer the entruster’s title to a buyer in the ordinary course of business. See UCC § 2-403(2). This doctrine would have empowered Imig to transfer good title to “Quo Vadis” to a buyer in the ordinary course, because Mucha had entrusted the painting to Newcomb-Macklin, a dealer in paintings. But Rupprecht was not (1) a buyer (2) in the ordinary course of business. Hence he did not obtain Mucha’s title to the painting, and could not convey title to Fly-by-Nite and thence to King, even if King was a buyer in the ordinary course from Fly-by-Nite (which we need not decide). See 3 Anderson on the Uniform Commercial Code § 2-403:54, at p. 599 (3d ed. 1983). King might have a claim against Tome for breach of an express or implied warranty of title, see UCC § 2-312; cf. Note, Uniform Commercial Code Warranty Solutions to Art Fraud and Forgery, 14 Wm. & Mary L.Rev. 409 (1972), but no such claim is made in this lawsuit, to which Tome is not even a party.

In 1982 a friend of King’s named Rapparlie, apparently acting on King’s behalf, wrote Jiri Mucha, seeking information about the painting (e.g., its symbolism). Although the letter does not state that Rapparlie had seen the painting, the nature of the questions indicates that he must have; he also said that he “represented]” the painting, whatever that means. Jiri smelled a rat. As the letter had been sent from Chicago, it was apparent that New- comb-Maeklin had sold or given away the painting. Investigation led to King, and this suit was filed in 1983.

The letter from Rapparlie put Jiri on notice that he was the probable victim of a conversion by his bailee, but he acted promptly and sued within the five-year period. If as we have concluded the district judge was entitled to find that Jiri had never abandoned his rights to the painting and that the statute of limitations for conversion had not begun to run before 1981 (or at the very earliest 1979, when the conversion took place though Jiri knew nothing of it till sometime after he received the letter from Rapparlie in 1981), then Jiri owns the painting and is entitled to get it back from someone who does not claim to be a buyer in the ordinary course of business. It is true that without Tome’s and King’s efforts Jiri might never have learned of the whereabouts of a painting that may be worth hundreds of thousands of dollars to him, but they do not claim that Jiri owes them any finder’s or salvor’s fee beyond the money King has spent on restoration, which the court ordered Mucha to repay. See 2 Palmer, The Law of Restitution § 10.3, at p. 371 (1978); Seavey & Scott, Notes on Certain Important Sections of Restatement of Restitution § 117 (1937).

The last issue in this appeal arises from the fact that Jiri owns only a one-half interest in the painting. King argues that Jiri’s sister, Jaroslava, who owns the other half, is barred by the statute of limitations from reclaiming her interest, making King (at worst) the co-owner of the painting, with Jiri. But quite apart from evidence presented by Jiri that his sister was mentally ill at all times relevant to this case, which would toll the statute of limitations if the mental illness were severe enough to create a legal disability, see Ill.Rev.Stat. ch. 110, 1113-211, we do not see how the statute of limitations could be thought to have run against her. There is no indication that Jiri ever told her about the letter from Rapparlie. We do not accuse Jiri of bad faith, because, although he is not his sister’s legal guardian, he may for all we know be appearing in a representative capacity in this suit so far as her share of the painting is concerned.

Since it is not certain that Jiri Mucha is suing on behalf of his sister as well as himself, we must consider King’s alternative argument, which is that Jiri, owning as he does only half the painting, can win only half the painting, leaving the other half with King unless and until Jaroslava appears. This argument if accepted would require that the painting be sold and the proceeds divided between Jiri and King, leaving Jaroslava to sue King on her own. This seems an unreasonable result on at least two grounds. First, it gives the converter possession of half the proceeds of the converted property, as against the lawful co-owner of the property. Second, it gives the converter the power to force the sale of the property, though the lawful owners might prefer not to sell it.

It would be surprising if the law gave a converter such rights, and as a matter of fact it does not. What is true is that the general rule, which we may assume without having to decide King might have invoked (but did not), is that one co-owner cannot maintain an action for replevin, even against a converter, without all the other co-owners joining in the action. See, e.g., Harris v. LeMasters, 215 Ill.App. 282, 284-85 (1919); Mayo v. Jones, 8 Wash.App. 140, 142, 505 P.2d 157, 158 (1972); but cf. Kresha v. Kresha, 211 Neb. 92, 317 N.W.2d 776 (1982) (cotenants of real property). See generally Annot., 110 A.L.R. 353 (1937). All this means, however, is that Jaroslova may have been an indispensable party plaintiff to this suit, see Fed.R.Civ.P. 19(a)(2); and the defense of failure to join an indispensable party is waived if not made in the trial court, see Fed.R.Civ.P. 12(h)(2), and it was not. For pertinent illustrations of the principle see Cromwell v. Ward, 192 Neb. 178, 181-82, 219 N.W.2d 446, 449 (1974); Heinen v. Home Mutual Casualty Co., 5 Wis.2d 282, 292-93, 92 N.W.2d 836, 841 (1958).

So King’s last defense fails. But we want to be sure that Jaroslava’s interests are protected. The district judge should therefore retain jurisdiction of this case for the limited purpose of ensuring that Jaroslava or her legal guardian is aware of the recovery of the painting, so that her rights in it can be enforced. We therefore modify the judgment by directing the district court, before allowing Jiri Mucha to take possession of the painting, to require him to file with the court an acknowledgment signed by Jaroslava (unless she has a legal guardian, in which event he must sign the acknowledgment) that she is aware of the recovery of this valuable painting of which she is a 50 percent owner. Subject to this minor modification, the judgment is

Affirmed.