Lost and Found: Rethinking Governmental Interests in Nazi-Looted Art Cases
In 2024, the U.S. Court of Appeals for the Ninth Circuit ruled that Spain’s Thyssen-Bornemisza Collection (TBC) legally owns a Pissarro painting looted by the Nazis, ending decades of litigation over its rightful owner. The dispute began after Claude Cassirer, the grandson of the painting’s original Jewish owner, Lilly Cassirer, discovered the artwork in a Spanish museum. When Spain refused to return it in 2005, Cassirer — then living in California — filed suit in the U.S. District Court for the Central District of California [1].
The central issue was which law should apply: California’s, which favored the Cassirer family, or Spain’s, which favored the museum. After years of litigation, the case reached the Ninth Circuit, which applied California’s “governmental interest” test to decide [2]. Although California follows the common-law principle nemo dat quod non habet — “no one can give what they do not have” — and thus rejects any transfer of ownership in stolen property, the Court acknowledged a competing policy interest. Spain, it reasoned, had a compelling stake in preserving the stability and predictability of its property system, particularly in protecting good-faith purchasers who rely on that system’s consistency. Moreover, Spain’s interests were more directly implicated, as that the painting had never been physically located in California. The Court therefore applied Article 1955 of the Spanish Civil Code, which allows a good-faith purchaser to acquire ownership of stolen property after a prescribed period of time [3]. Accordingly, the Ninth Circuit found that the museum legally owned the painting.
Panel Judge Callahan concurred in the decision to apply Spanish law but wrote that the decision was “at odds with her moral compass, and Spain should have voluntarily relinquished the painting.” [4] The Ninth Circuit’s ruling was at odds with the public commitments of both the United States and Spain to return Nazi-looted art to its prewar owners and their heirs [5]. Earlier decisions in the Cassirer litigation — particularly those of the district court and the Ninth Circuit in prior appeals — had criticized Spain and its museum for failing to act consistently with the 1998 Washington Principles on Nazi-Confiscated Art and subsequent international declarations. Yet, both the district court and appellate courts repeatedly emphasized that they had no authority to compel Spain or its museum to follow these non-binding diplomatic commitments [6].
Although the Ninth Circuit maintained that it could not impose moral judgments on Spanish law, I argue that Spain’s public commitments to return Nazi-looted art should have been factored into its “governmental interest” analysis — particularly because the TBC is a public foundation managed by the Spanish government. My argument will unfold in two parts: first, I will examine how the Ninth Circuit defined and applied “governmental interests,” identifying internal inconsistencies and oversimplifications in its reasoning; and second, I will show that, as a state-run institution, the TBC’s actions reflect Spain’s official conduct, meaning Spain’s publicly stated commitments to restitution should have been weighed as part of its governmental interests.
California’s “Governmental Interest Approach” to Choice of Law
The Ninth Circuit ruled in favor of the defendant after applying California’s choice of law analysis — a legal process used to decide which jurisdiction’s laws should govern a case. California follows a three-step “governmental interest” approach, which asks:
- Whether the relevant laws actually differ;
- Whether there is a true conflict because each jurisdiction has a legitimate interest in applying its own law; and,
- Which jurisdiction’s interests would be more impaired if its law were not applied [7].
Note that California “employs the common law rule that thieves cannot pass good title to anyone, including a good faith purchaser” [8]. Article 1955 of Spanish Civil Code, on the other hand, says: “[o]wnership of movable property prescribes by three years of uninterrupted possession in good faith…[or] by six years of uninterrupted possession, without any other condition” [9]. First, it is clear that the relevant laws are different. In response to the second question, the Ninth Circuit found that both jurisdictions had legitimate interests. Because the painting was located in Spain, Spain had an interest in applying its own property laws, which protect purchasers who hold artwork for a certain period of time. At the same time, California had an interest in protecting its residents, like the Cassirers, through its 2010 law, §338(c)(3)(A), which reflects the state’s “strong interest in protecting the rightful owners of fine art who are dispossessed of their property” [10]. Furthermore, both California and Spain filed amicus briefs — legal documents submitted by “friends of the court” (amici curiae) who are not parties to the case but have a stake in its outcome — emphasizing their interest in having their own laws applied to the dispute [11]. As a result, the outcome of the trial rested on the third question: “which state’s interest would be more impaired if its policy were subordinated to the policy of the other state?” [12].
The Ninth Circuit’s determination that Spain had a deeper interest in upholding its laws rested on a false dichotomy. Per the directions of the California Supreme Court, the Ninth Circuit strove for the “maximum attainment of underlying purpose by all governmental entities” [13]. In the majority opinion by Judge Bea, with a concurrence by Judge Callahan, the California Supreme Court noted that Spain had “interests in providing certainty of title to entities like TBC,” and that California had an “interest in deterring theft and facilitate recovery for victims of stolen art, like the Cassirers” [14]. Although these interests are both relevant and likely true, in the next section, I will argue that they represent an oversimplification of the complex governmental interests at play. As Judge Graber explained, California law “asks whether the jurisdiction’s interests and policies as a whole would be impinged,” not just the narrow purpose of the specific statute at issue [15]. If we demonstrate, then, that Spain likewise has an interest in facilitating the recovery of stolen art — particularly art looted during the Nazi era — the foundation of the Court’s reasoning becomes considerably weaker [16].
The Washington Principles and Spain’s Interest in Restitution
Before explaining why the Ninth Circuit wrongly dismissed Spain’s interest in repatriating Nazi-looted art, I must highlight two framing principles it adopted. First, the Court must evaluate each jurisdiction’s interests “based on the circumstances of this particular dispute” [17]. Second, the analysis may not take into account “whether the Spanish rule or the California rule is the better or worthier rule” [18]. Relying on the second principle — that the court may not make independent moral judgments about Spanish law — the majority opinion dismissed the plaintiffs’ argument that Spanish law should carry little weight because it is out of step with international consensus. The opinion engaged minimally with this claim, quickly deeming it irrelevant to the legal analysis.
By dismissing this claim so quickly, however, the Court overlooked a crucial point: Spain had voluntarily endorsed international agreements affirming its commitment to the restitution of Nazi-looted art. In 1998, the Washington Principles on Nazi-Confiscated Art were established to help reunite Holocaust survivors and their heirs — such as the Cassirer family — with property unlawfully taken during World War II. The Principles call on participating nations to seek “fair and just” solutions regarding Nazi-confiscated artworks, as well as other cultural and religious property [19]. Over forty countries and several NGOs, including Spain, endorsed the agreement. Similarly, the Vilnius Forum Declaration (2000) urges all governments to make every reasonable effort to return cultural assets looted during the Holocaust era to their rightful owners or heirs [20]. It specifically encourages states to implement the Washington Principles; Spain is one of 38 countries that signed onto this document as well [21].
An internal conflict therefore arises between Spain’s general property laws and its public commitment to the restitution of Nazi-looted art. While the Ninth Circuit was correct that these international commitments are legally non-binding, it erred in excluding them entirely from its assessment of Spain’s governmental interests. There are three reasons why this matters:
First, although Spain’s property laws apply broadly to all private property, its more recent commitments — including the Washington Principles and Vilnius Forum Declaration — specifically address the restitution of Nazi-looted art. These commitments are directly relevant to both Spain’s governmental interests and “the circumstances of this particular dispute,” which the Ninth Circuit itself recognized as the proper lens of analysis. Because this aspect of Spain’s interests directly overlaps with California’s under the comparative impairment analysis, the degree of harm Spain would face from not applying its own law is correspondingly reduced [22].
Second, the Thyssen-Bornemisza Collection Foundation is a public foundation, not a private entity. The Spanish state purchased the collection from Baron Hans Heinrich Thyssen-Bornemisza in 1993 [23], and under Spanish law, a public foundation must serve the public interest [24]. This museum’s status therefore gives the Spanish government a distinct interest in ensuring that its state-managed institutions comply with the moral and cultural standards it has publicly endorsed. This same interest may not necessarily apply to a private, for-profit corporation [25].
Finally, reasons One and Two make the circumstances of this case extraordinarily narrow. It is unreasonable to suggest that requiring a state foundation — created to serve the public interest — to return a painting that the government itself has acknowledged should be restituted on moral grounds would destabilize private actors’ legitimate expectations of property ownership under Spanish law [26]. Spain’s comparative impairment is further reduced.
Conclusion
It is not my claim that considering Spain’s commitment to restituting Nazi-looted art would necessarily have changed the Ninth Circuit’s ultimate decision. The Court noted that California’s connection to the case was relatively limited — the Cassirer family’s residency being its primary tie — and Spain’s tangible relationship to the painting itself remains stronger. Moreover, Spain’s repeated refusal to return the artwork despite its public commitments complicates any argument that those commitments meaningfully shape its governmental interests.
In September 2024, Governor Gavin Newsom signed a law mandating that California substantive law apply in all actions by California residents seeking recovery of fine art or culturally significant objects, including those covered by the Holocaust Expropriated Art Recovery Act of 2016, Cal. Civ. Code §338(c)(4). The statute explicitly references the Cassirer litigation and seeks to deter art theft while establishing clearer restitution standards. Although the Supreme Court has since vacated the Ninth Circuit’s ruling, this legislative fix addresses only the outcome of the case, not the underlying analytical flaws which produced it.
Ultimately, the Court’s decision to exclude Spain’s restitution pledges from its “governmental interest” analysis represents a missed opportunity to fully engage with the nuances of international and domestic policy. The Cassirer ruling highlights the broader need for courts to adopt a more holistic framework for evaluating governmental interests: one capable of reconciling tensions between formal laws and a state’s broader policy commitments. Such an approach might enable judicial reasoning to align more closely with evolving international norms of restitution and justice, while still preserving the integrity of domestic legal systems.
Footnotes
[1] Cassirer v. Thyssen-Bornemisza Collection Foundation, No. 19-55616, slip op. at 2-3 (9th Cir. Jan. 9, 2024). https://cdn.ca9.uscourts.gov/datastore/opinions/2024/01/09/19-55616.pdf. [2] While I examine the specific workings of this test in the body of the article, California’s “governmental interest” approach resolves choice-of-law conflicts by comparing the policy objectives of the jurisdictions involved. The Ninth Circuit ultimately applies the law of the state whose interests would be most seriously impaired if its policy were subordinated. [3] Spanish Civil Code, 1955 (2013 translation), art. 1955, https://data.globalcit.eu/NationalDB/docs/spanish-civil-code-ENG%202013.pdf. [4] Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 3. [5] Elena Baylis, “Introductory Note to *Cassirer v. Thyssen-Bornemisza Collection Foundation *(9th Cir.),” International Legal Materials 63 (2024): 1080, https://scholarship.law.pitt.edu/fac_articles/609. [6] Baylis, Introductory Note to Cassirer v. Thyssen-Bornemisza Collection Foundation, 1080. [7] William S. Dodge, “Ninth Circuit Decides Cassirer in Favor of Spain,” Transnational Litigation Blog, January 10, 2024, https://tlblog.org/ninth-circuit-decides-cassirer-in-favor-of-spain/. [8] Baylis, Introductory Note to Cassirer v. Thyssen-Bornemisza Collection Foundation, 1080. [9] Spanish Civil Code (2013 trans.), art. 1955. [10] Cassirer v. Thyssen-Bornemisza Collection Foundation, No. 19-55616, slip op. at 7 (9th Cir. July 9, 2024), https://cdn.ca9.uscourts.gov/datastore/opinions/2024/07/09/19-55616.pdf. [11] Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 24. [12] Dodge, “Ninth Circuit Decides Cassirer in Favor of Spain.” [13] Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 7. The phrase “maximum attainment of underlying purpose by all governmental entities” reflects the idea that courts should interpret and apply laws to preserve each jurisdiction’s policy goals as much as possible. Under California’s governmental interest test, judges must first determine to what extent the laws and underlying interests of the jurisdictions actually differ, since any overlap means there is no true conflict on those points. Once the differences are identified, the Court seeks to apply the law that minimizes interference with each government’s policy objectives. In this case, I argue that the majority treated the interests as wholly opposed, and in doing so, overlooked areas where California’s and Spain’s policies could have been reconciled. [14] Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 21-22. [15] Ibid., at 22. This framework is reflected in the doctrine that “if a particular statute is infrequently enforced or interpreted even within its own jurisdiction, it has limited application in a conflict-of-laws case.” If the governmental interest analysis were purely about the text or intent of the law itself, the archaism would be largely irrelevant. Instead, the focus is on how that law fits into and/or reflects the state’s overall interests. [16] See footnote 13, which explains that when the policy goals of two jurisdictions overlap, the degree of actual conflict between their laws diminishes. If Spain’s interests also include facilitating the recovery of Nazi-looted art, this shared objective would narrow the conflict identified by the Court and weaken its justification for applying Spanish law. [17] Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 8. [18] Ibid. [19] U.S. Department of State, “Best Practices for the Washington Conference Principles on Nazi-Confiscated Art,” March 5, 2024, https://www.state.gov/office-of-the-special-envoy-for-holocaust-issues/best-practices-for-the-washington-conference-principles-on-nazi-confiscated-art. [20] “Vilnius Forum Declaration,” pt. 1, Looted Art Commission, accessed October 13, 2025, https://www.lootedartcommission.com/vilnius-forum. [21] “Vilnius International Forum on Holocaust-Era Looted Cultural Assets: Vilnius Forum Declaration,” Commission for Looted Art in Europe, accessed October 13, 2025, https://www.lootedart.com/MG8D3S66604. Although the Washington Principles (1998) and the Terezin Declaration (2009) are not legally binding and there is no court or arbitration body that enforces them, signatory nations, including Spain, have practical reasons to uphold them. As Judge Graber noted, if these treaties had legal force, Spanish law would already mirror California’s and no conflict analysis would be needed (see also: Cassirer v. Thyssen-Bornemisza Collection Foundation, slip op. at 17–18). Still, the agreements carry real weight: they shape international expectations, influence domestic legislation including the U.S. Holocaust Expropriated Art Recovery Act of 2016, and affect a nation’s cultural reputation and diplomatic credibility. Thus, even absent formal penalties, Spain has a strong interest in honoring the moral and political commitments embodied in these declarations. [22] In the comparative impairment analysis, the “degree of harm” refers to how much a jurisdiction’s policy goals would be undermined if its law were not applied. The greater the setback to a government’s objectives, such as maintaining predictable property rules or fulfilling restitution commitments, the greater the harm. Conversely, when both jurisdictions share similar policy aims, the harm to either from applying the other’s law is reduced, since their core interests are partly aligned. [23] Museo Nacional Thyssen-Bornemisza, “The Thyssen-Bornemisza Collection Foundation Is a Public Foundation,” Museo Thyssen, accessed October 13, 2025, https://www.museothyssen.org/en/transparency-site/institutional-organisational-information/organisational/structure. [24] Lourdes Márquez de la Calleja, Legal Environment for Philanthropy in Europe: Spain 2020 (Dafne-EFC), accessed October 13, 2025, https://www.transnationalgiving.eu/wp-content/uploads/2022/09/Spain_2020LegalEnvironmentPhilanthropy.pdf. [25] A government has a distinct interest in the conduct of its public institutions because they embody the nation’s values and policy commitments. That same interest does not necessarily extend to a private, for-profit corporation, whose actions are guided primarily by commercial objectives rather than by public accountability or moral responsibility. The Restatement (Second) of Conflict of Laws § 6 (1971) supports this distinction by instructing courts to weigh the nature and purpose of each jurisdiction’s interests; it recognizes that public, governmental interests differ from private ones. Likewise, in Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 423 (1964), the Supreme Court acknowledged that when a state-controlled entity acts, it does so as an instrument of national policy, thus reinforcing the point that Spain’s moral and cultural commitments are directly implicated through its ownership of the Thyssen-Bornemisza Collection Foundation. [26] Note the prior discussion of the Vilnius Forum Declaration (2000), which urges governments to make “every reasonable effort” to return Nazi-looted art to its rightful owners or heirs. Simply returning the Cassirer painting would likely fall within that commitment. Thus, complying with these principles would not undermine Spain’s property regime; rather, it would fulfill the moral obligations Spain has already endorsed.