Cassirer on Remand: Considering the Laws of Other Interested States

This post is by Carlos Manuel Vázquez, a professor of law at Georgetown Law School. It is cross-posted at Transnational Litigation Blog.

Claude Cassirer brought suit in federal court in California eighteen years ago against the Thyssen Bornemisza Museum of Madrid, Spain, to recover a painting by Camille Pissarro that was stolen from his grandmother by the Nazis during World War II.  After a reversal and remand from the U.S. Supreme Court last summer, the case is now before the Ninth Circuit for decision of the legal question that is likely to be decisive:  which law governs?

The district court and the court of appeals have so far framed the issue as a binary choice: the governing law on the merits is either that of Spain or that of California.   I suggest here that the issue is better framed as a choice between the law of Spain, on the one hand, and the laws of all the other states or countries with connections to the dispute, on the other.  (Disclosure: I submitted expert declarations in support of the plaintiffs on issues of public international law during earlier phases of this case.)

The U.S. Court of Appeals for the Ninth Circuit has affirmed the district court’s holding that, under the law of Spain, the plaintiff loses because the museum acquired title to the painting through adverse possession (otherwise known as acquisitive prescription).  It is equally clear that, under the law of California, the plaintiff would prevail because California does not recognize the acquisition of title to moveable property through adverse possession.  What has so far not featured prominently in the courts’ analyses of the choice-of-law issue is that the plaintiff would also prevail under the laws of all the other jurisdictions that have relevant connections to the dispute.  Under governmental interest analysis, this should be central to the analysis.

The Painting’s Journey

It is undisputed that the painting was looted from Lilly Cassirer by the Nazis.  After it was taken in Germany, the painting spent some time in California and Missouri and was subsequently sold to Baron Von Thyssen-Bornemisza by a Gallery in New York.  The painting then stayed at the Baron’s home in Switzerland for twelve years before it was loaned to the museum in 1988 and then sold to Spain in 1993.

The district court decided in this case that the Baron did not have valid title to the painting during the period in which he possessed it.  The Baron did not purchase the painting from someone with good title, and he did not obtain good title through adverse possession because he did not possess the painting in good faith, as required by Swiss law.  The court held that there were many red flags that should have alerted the Baron to the possibility that the painting had been stolen by the Nazis.

Accordingly, the museum did not acquire good title to the painting when it purchased it from the Baron in 1993.   But, the court held, the question whether the museum acquired title to the painting through adverse possession is governed by the law of Spain, and the law of Spain, unlike the law of Switzerland, allows acquisitive prescription if the painting is possessed for six years even without good faith.  The time period is longer if the possessor is an accessory to the theft, but someone who possesses the item without good faith is not for that reason alone deemed an accessory. Because the museum was not an accessory to the theft, the court held, the museum has acquired good title to the painting under the law of Spain because it had possessed it for just over six years before Claude Cassirer learned of its location and asked for it back.

California’s Approach to Choice of Law

The U.S. Supreme Court held in this case that, even in suits against foreign state instrumentalities under the Foreign Sovereign Immunities Act, a federal court must apply the choice-of-law rules of the state in which it sits.  The district court had applied California’s choice-of-law rules, but the Ninth Circuit did not review its analysis, having erroneously concuded that a federal choice-of-law rule applied. The appellate court must now review the district court’s application of California’s choice-of-law rules.

Under traditional choice-of-law rules, the issue of title to moveable property is governed by the law of the place where the property is located.  But California, like most U.S. states, long ago rejected the traditional choice-of-law approach and adopted in its place a form of governmental interest analysis.  This approach asks the courts, in cases in which the substantive laws of the relevant states differ, to determine whether the relevant states have an interest in having their laws applied.  If only one state has such an interest, then there is a false conflict, and the court applies the law of the only interested state.  If more than one state has an interest, there is a true conflict.  To resolve true conflicts, California has adopted the “comparative impairment” approach, under which the court applies the law of the state whose policies would be most impaired if not applied.

The district court in the Cassirer case focused on the interests of California and Spain.  The court first concluded that the laws of those two states differed because Spain recognizes acquisitive prescription of moveable property after six years even if the possession was not in good faith, whereas California does not recognize acquisitive prescription of moveable property.  The court then concluded that both California and Spain have an interest in having their laws applied.  Spain’s law prioritizes the interests of the possessor of the property and, more generally, the interest in certainty of title.  Spain’s interest is implicated in this case because the possessor is a Spanish entity and the painting is in Spain.  California’s law prioritizes the interest of the original owner of stolen property, and this policy is implicated in the case because the original owner’s heirs are domiciled in California.  Because both Spain and California have an interest in having their laws applied, the case presents a true conflict.

To this point, the district court’s analysis was sound.  The same cannot be said of its analysis of the next step—determining which state’s law would be more impaired if not applied.  The court concluded that Spain’s policies would be significantly impaired if not applied but California’s policies would be only minimally impaired.  Why?  Because California’s interest in having its law applied depended largely on the plaintiff’s fortuitous, unilateral decision to move to California in 1980, long after the painting had been stolen from his grandmother by the Nazis.

What the court overlooked, however, is that Spain’s interest in the case is equally fortuitous.  The painting was stolen in Germany and was located in California, Missouri, New York, and Switzerland before it made its way to Spain as a result of the Baron’s decision to establish a museum in Spain bearing his name. If California’s interest is to be discounted because it resulted from the plaintiff’s fortuitous decision, then Spain’s interest should similarly discounted because it resulted from the fortuitous decision of the museum’s predecessor in interest.

Spain’s Law on Acquisitive Prescription

Actually, it may not be fortuitous that stolen property will make is way to Spain, but the reason for this is one that should make a court wary to apply Spanish law.  Spain’s law of acquisitive prescription is unusually friendly to possessors of stolen property.  Common law jurisdictions generally do not recognize acquisitive prescription of moveable property.  They do not disregard the interests of possessors of property or the general interest in certainty of title, but they give effect to those interests through statutes of limitations, which limit the time the original owners have to initiate lawsuits to recover the property and in this way deter the original owners from sleeping on their rights.  But statutes of limitations often begin to run when the original owner discovers the location of the stolen property.  That is, indeed, the law in all states of the United States by virture of a federal law establishing a six-year statute of limitations for suits to recover Nazi-looted art, which begins to run upon discovery.  Other jurisdictions do recognize the acquisition of title by adverse possession, but (as discussed below) they generally require that the possessor have acquired the property in good faith, meaning without sufficient reason to believe that the property was stolen.  Jurisdictions that allow the acquisition of title by adverse possession without  good faith generally require a far longer period of possession than Spain’s six years (for example, twenty years under Italian law).

Spain’s law is unusually friendly towards possessors of stolen property in allowing the acquisition of title through bad faith adverse possession after a mere six years.  Spain is thus, relatively speaking, a haven for stolen property, and it would not be surprising to find that stolen property winds up there.  For this reason among others, scholars have advocated replacing the traditional situs rule for stolen cultural property with a lex originis rule, under which the law to be applied would presumptively be the law of the place where the property was stolen, coupled with a disciovery rule for triggering the running of the prescription period.  As noted, California has replaced the traditional rule with governmental interest analysis, but, in applying interest analysis, the same concern should lead California courts to resist applying the law of the place to which the stolen property was taken.  (Alternatively, the courts of California could refuse to apply the law the situs, if unusually friendly towards possessors of stolen property, on ground that the law contravenes California’s strong public policy.)

The museum might argue that there is no evidence that the painting was brought to Spain to take advantage of its unusually friendly law.  It may well be true that the Baron did not sell the painting to the museum in Spain in order to launder his stolen painting.  The museum’s web site indicates that, in 1988, the Baron had offers for his collection from the United Kingdom, California (Getty Foundation) and Germany, but chose to establish the museum in Spain because his fifth wife, a Spanish beauty queen, wanted to establish an art museum in her home country.  Be that as it may, it is equally true that the plaintiff’s decision to move to California was not driven by his desire to take advantage of California’s more protective law.  Indeed, when he decided to move to California, he assumed that the painting had been lost or destroyed during the war.

In sum, if the fact that the Baron’s decision to sell the painting to a museum in Spain was not taken for opportunistic reasons is not a reason to discount Spain’s interest, then the fact that Claude Cassirer’s decision to move to California was not made for opportunistic reasons is equally a reason not to discount California’s interest.  The painting’s presence in Spain, in the hands of a Spanish museum, is (at best) just as fortuitous as Claude Cassirer’s decision to move to California.

Other Interested Jurisdictions

If so, then how does one break the tie?  One answer might be to apply the law of the forum, and indeed there is California case-law placing the burden on the party arguing against applying forum law.

But, on closer inspection, the relevant interests are not in equipoise.  California and Spain are not the only jurisdictions with connections to this dispute.  Both the painting and Lilly Cassirer were initially located in Germany.  Germany’s law allows acquisitive prescription in ten years, but only if the property was possessed in good faith.  (A statute of limitations cuts off the original owner’s power to bring an action to recover the property after thirty years, but it does not vest title in the possessor.) As the district court held in this case, the Baron did not acquire title to the property under Swiss law of acquisitive prescription by virtue of his possession of the painting because he did not possess the painting in good faith. Application of the German law of acquisitive prescription leads to the same conclusion.  The court did not address whether the museum possessed the painting in good faith because that issue was not relevant under Spanish law.  But surely the Baron’s lack of good faith should be attributed to the museum that he co-founded and bears his name.  In any event, as the district court found, the red flags that alerted the Baron to the possibility that the painting was stolen by the Nazis were equally apparent to the museum.

As noted, the painting later spent time in California, Missouri, and New York. The laws of Missouri and New York on acquisitive prescription are in all relevant respects the same as California’s.  The painting then spent some time in Switzerland, and, as we have seen, the plaintiff should prevail under Swiss law as well.

As for Lilly Cassirer, after escaping from Germany, she lived for some time in England.  English law, like the law of California, does not technically recognize acquisitive prescription, but its statute of limitations limits the time in which to bring an action for conversion.  The limitations period has the same effect as acquisitive prescription because § 3(2) of the Limitations Act provides that, after the expiry of the limitations period for bringing an action for conversion, the original owner’s title to the movable property is extinguished.  The limitations period is generally six years, but in the case of theft, the limitations period begins to run from the date of the first “innocent” conversion. “As regards the original thief, or . . . any party acquiring the movable from him who is not in good faith,” Faber & Lurger note, “it would appear that there is no limitation period for the bringing of an action in coversion.”

From England, Lilly moved to Ohio, which has the same law regarding adverse possession as California.  Neither Lilly’s moves to England and Ohio nor Claude’s move to California were driven by a desire to take advantage of those states’ protective law of acquisitive prescription.  Indeed, if Lilly had wanted to take advantage of a jurisdiction’s law of acquisitive prescription, she could have moved to practically any jurisdiction other than Spain.  As we have seen, Spain’s law of acquisitive prescription (as interpreted by the district court and court of appeals in this case) is an outlier in recognizing a change of title as a result of possession of stolen property without good faith in a mere six years.

Should the court broaden its focus and consider the laws and interests of these other jurisdictions?  The district court’s own analysis suggests so.  After all, if the interest of the plaintiff’s current place of domicile is discounted because it resulted from his fortuitous decision, then surely the law and interest of the place from which he moved should be considered instead.  Courts that discount a party’s domicile if acquired after the start of the dispute generally consider instead the interest of the jurisdiction from which the party moved.  And if the interest of the place to which the stolen painting was taken is discounted because it resulted from the fortuitous (or non-fortuitous) decision of the possessor’s predecessor, then surely the interest of the place from which painting was taken should be considered instead. The district court additionally discounted California’s interest because the original taking did not occur in California and because the Baron did not purchase the painting in California.  These reasons for discounting California’s interest suggest that the court should consider instead the laws of the place where the original taking occurred (Germany) and the place where the Baron bought the painting (New York).

There is, indeed, substantial authority for the proposition that the interests of jurisdictions with connections to the dispute should be aggregated when these laws have the same content.  The Restatement (Second) of Conflict of Laws makes this point explicitly.  A comment to § 145 on torts explains that “when certain contacts involving a tort are located in two or more states with identical local law rules on the issue in question, the case will be treated for choice-of-law purposes as if these contacts were grouped in a single state.”  The same comment appears in numerous other provisions of the Second Restatement, including the provision on real property (§ 222) and the provision on chattels (§ 244).  The laws of the jurisdictions discussed above are not identical in all respects, but they are identical in the relevant respect: under each of these laws, the plaintiff should prevail.

Although California has not adopted the Restatement (Second) as its choice-of-law rule, the Restatement’s approach to aggregation is in principle equally relevant to governmental interest analysis in general.  A contrary rule would allow circumvention of the relevant states’ interests in a dispute through a divide-and-conquer strategy.  The district court in this case appears to have fallen into this trap.


On remand from the Supreme Court, the Ninth Circuit certified the choice-of-law question under California law to the California Supreme Court, but that court denied the request. It is now up to the Ninth Circuit to review and correct the district court’s application of California’s choice-of-law rules.  In doing so, the court of appeals should consider not just the interests of California and Spain but also those of Germany, New York, Missouri, Switzerland, England, and Ohio. The fact that all those jurisdictions would reach the same result as California is a strong reason to rule in favor of the plaintiff in this case.