U.S. Conflict of Laws Involving International Estates and Marital Property: A Critical Analysis of Estate of Charania v. Shulman

I.     Introduction

In the United States, people routinely move from one state to another and also own property in more than one state. They marry and, while married, acquire title to property in their individual names, or alternatively, in some form of title by which his or her spouse also obtains an ownership interest. Under the laws of some states, the spouse automatically acquires a 50% interest simply on account of being married to the individual acquiring the property.1 Even if title is taken exclusively in the individual’s name, the spouse’s legal rights under the law are not impaired. These states maintain a community-property system, as opposed to the vast majority of states that maintain a separate-property system.

But suppose the couple moves from the community-property state to a state that has a separate-property regime. Are the spouse’s previously acquired rights forfeited? Or have they been forever vested?2 Likewise, suppose the property was originally acquired by one spouse individually in a separate-property state that does not recognize community property. If the couple then moves to a community-property state, does the spouse immediately acquire rights in the property? Or does the original characterization of the property as separately owned remain in force?3

The answers to these questions are important for probate and estate administration, as well as federal estate tax law. In particular, upon the death of an individual, a critical question needs to be answered: What assets constitute the individual’s estate both for non-tax and federal estate tax purposes? More precisely, in terms of the topic this Article explores: What portion of the assets titled in the name of the decedent actually belongs to the surviving spouse on account of community-property laws? To the extent that the surviving spouse is found to own a portion of such property, that ownership will not only reduce what the devisees or heirs of the decedent receive, but it will also reduce the amount includible in the decedent’s gross estate for federal estate tax purposes, thereby reducing the decedent’s federal estate tax liability.

To determine whether the surviving spouse has an ownership interest in a particular asset titled exclusively in the decedent’s name, the law of some jurisdiction will have to be consulted. But which one? For nearly 200 years, the prevailing doctrine in the United States has been “partial mutability.” Under this conflict-of-laws rule, the right of a spouse in a movable asset acquired during marriage is determined by the law of the state in which the spouses had their marital domicile at the time of the acquisition of the asset.4 Thus, if the spouses change their marital domicile during the marriage, it is entirely possible that different movable assets will be governed by different laws. This conflict-of-laws rule is widely known as “partial mutability” because the law of the original marital domicile does not remain the governing law as to assets acquired after a change in marital domicile has taken place.5 In other words, there is “mutability.” However, it is only “partial” because with respect to rights acquired at a particular marital domicile, they are not mutable and are not lost simply by moving to a new marital domicile that does not recognize those spousal rights.6

Notice that the preceding choice-of-law rule does not apply to immovable property. For this type of property, the rights of a spouse are determined by the law of the situs jurisdiction rather than the marital domicile.7 As a result, there is a “scission” in determining marital-property rights, which elevates the significance of identifying property as movable or immovable. That task is further complicated by the fact that different jurisdictions may characterize assets differently. Thus, there is necessarily a preliminary question as to which jurisdiction’s law should be controlling on the characterization question.8

Up to this point, I have discussed cross-border marital-property rights and succession within the United States. However, many foreign countries, principally those adhering to the civil-law system, reject partial mutability and impose a marital-property rights choice-of-law rule described as “strict immutability.”9 This approach looks to the law of the spouses’ domicile at the time of their marriage as the final and immutable determinant of the spouse’s marital-property rights, regardless of the marital domicile at the time of the asset’s acquisition.10 While the spouses are typically free to mutually alter the governing law by contract, if they fail to do so, strict immutability controls.

Finally, in sharp contrast to partial mutability and strict immutability, a third choice-of-law approach is enforced by a limited number of foreign jurisdictions and is known as “total mutability.”11 Pursuant to this approach, the marital-property rights of the spouses mutate as the spouses change their marital domicile. In other words, the rights of the spouses are not finally determined until divorce or the death of a spouse, either of which terminates the marriage.12 Because this approach fosters an inherent instability in property rights, total mutability has not been widely adopted.

Inasmuch as many foreign countries do not adhere to the partial-mutability approach, but rather look to the original marital domicile under the strict-immutability approach, there is a substantial likelihood that the controlling law with respect to at least certain assets will be different, depending on which approach is applied. In contrast, when only states of the United States are involved, there is near uniformity on the application of the partial-mutability approach.13

Surprisingly, in light of the number of people with international ties, and the contentious issues that one would expect to arise, there is a dearth of federal appellate court conflict-of-laws decisions involving international marital-property rights.14 One might predict that any opinion that seeks to analyze the relevant conflict-of-laws issues in depth is likely to carry great weight and substantially influence other courts. Especially since conflict-of-laws issues involving international aspects of marital property are hardly the specialty of judges and attorneys within the United States, there may be a hesitancy and humility about criticizing any such appellate opinion. In fact, there is one such opinion by the United States Court of Appeals for the First Circuit, Estate of Charania v. Shulman, decided in 2010.15 Called upon to address a slew of international marital-property choice-of-law issues of supreme importance, the court managed to resolve virtually every one of them incorrectly. Nevertheless, the opinion is increasingly cited as authoritative.16

This Article intends to bring to light the many analytical flaws of Charania in the hope, first, of discouraging uncritical reliance on the opinion by future courts and commentators. Second, the Article’s exposure of the numerous and severe analytical flaws in the court’s application of “foreign law” is intended to serve as a cautionary lesson concerning the difficulties in getting “foreign” law right, especially when that “law” includes the foreign jurisdiction’s whole law, that is including its conflict-of-laws rules as well.

II.     The Underlying Facts of Charania

Noordin M. Charania and Roshankhanu Dhanani, both of Indian descent, were born in Uganda at a time when it was a British protectorate and administered as part of British East Africa. They married in 1967, just five years after Uganda gained its independence.17 They had both been granted United Kingdom citizenship.18

In 1972, the infamous dictator Idi Amin “ordered the expulsion of [all] Ugandans of Asian descent” and the expropriation of their property.19 Mr. Charania and his wife fled to Belgium, as he had worked for a Belgian trading company.20 The couple proceeded to settle in that country, having no property of substantial value to their names when they arrived, with just a few items of personal property.21 Indisputably, it became their marital domicile as well as the individual domicile for each of them, along with their children.22 The record does not reveal if Mr. Charania or his spouse had ever set foot in their country of citizenship, the United Kingdom, but they do not appear ever to have spent a sustained period of time there or had any other palpable ties with the United Kingdom or any of its lawmaking subdivisions.23 There is also no indication that the couple spent any time in the United States or had any ties with it or any of its lawmaking subdivisions.24

In August 1997, Mr. Charania made an investment in Citicorp, which soon became Citigroup.25 The corporation is a domestic U.S. corporation.26 On account of appreciation and stock splits, Mr. Charania’s investment blossomed over time into a valuable asset of 250,000 shares.27 On January 31, 2002, Mr. Charania died.28 The 250,000 shares had a fair market value of $11,790,000.29 At the time of his death, the shares were registered exclusively in the decedent’s name and were held by a bank in Hong Kong.30 Under § 2104(a) of the Internal Revenue Code (“IRC”), stock of a domestic corporation owned by a nonresident, noncitizen of the United States, is deemed situated in the United States31 and, on that basis, is included in the gross estate of such nonresident, not a citizen of the Unites States for purposes of the calculation of federal estate tax liability.32 Precisely because the basis for taxation of the shares of stock is their presumed situs in the United States, the fact that the decedent may never have set foot in the United States proved irrelevant.33

Although the shares of Citigroup stock were registered exclusively in Mr. Charania’s name at the time of his death, his estate filed a return reporting only one-half of the shares as being owned by him.34 In support of its position, one might have expected the estate to emphasize that the partial-mutability approach applies and that the shares were community property, having been acquired while the couple were domiciled in Belgium, a community-property jurisdiction. Belgium treats all property acquired during marriage as owned one-half by each spouse, unless the spouses elect to opt out.35 There was no evidence that Mr. Charania and Ms. Dhanani had any intent or made any effort to opt out of the default Belgian community-property law.36 Pursuant to the partial-mutability approach, the decedent’s spouse had acquired a one-half interest in the shares at the time of their acquisition. Thus, the decedent’s estate should be subject only to tax on the one-half of the shares that he owned.37

In fact, attorneys for the estate did not contend that the partial-mutability approach was determinative. Instead, they stipulated that the law of England applied and that English law applies the total-mutability approach.38 They then argued that, under the total-mutability approach, Belgian community-property law applied because the spouses had a Belgian marital domicile at Mr. Charania’s death.39 The Internal Revenue Service (“IRS”) agreed that English law controlled, but asserted that English law applies the strict-immutability approach.40 That approach, the IRS argued, required application of the law of Uganda, the original marital domicile, rather than the law of the ultimate marital domicile, Belgium.41 Since the parties stipulated that Ugandan law was the same as English law, the IRS maintained that England’s separate-property law governed and that the decedent’s surviving spouse did not own any of the shares at his death.42 Accordingly, the IRS argued, it was correct to include in the decedent’s gross estate all of the Citigroup shares.43

The First Circuit endorsed the stipulations of the parties without seriously questioning them.44 The court decided that on the basis of a single decision of the House of Lords from more than a century earlier, De Nicols v. Curlier,45 English law requires application of strict immutability. It also concluded that it was compelled to strictly apply this authority, citing two federal diversity cases requiring federal courts to abide by the holdings of the highest court of the state in which the federal court sits.46 In holding for the government, the First Circuit affirmed the decision of the Tax Court.47 The estate did not seek certiorari to the United States Supreme Court.

When one steps back and considers the outcome in Charania it seems rather surprising, if not unjust. A man of Indian descent was exiled with his spouse from Uganda, their birthplace and country where their marriage was celebrated. They fled to Belgium with no assets. They remained in Belgium for the rest of the man’s life and, while there, acquired substantial assets, notably Citigroup shares. There was only one connection with the United Kingdom: citizenship acquired as a result of their exile from the former British colony. Neither the decedent nor his spouse ever developed any palpable connections specifically with England, a political subdivision of the United Kingdom, nor with the United States. Nevertheless, the United States was able to tax the estate on the basis of English law!

III.     The Troubling Flaws in the First Circuit’s
Conflict-of-Laws Analysis

There are troubling flaws in the First Circuit’s conflict-of-laws analysis in Charania. These flaws, arguably begin with the second footnote of the opinion, in which the court stated that it “express[ed] no opinion on the appropriateness of this [principle]. Rather, we work within the framework to which the parties have agreed.”48 The “principle” referred to by the court is that “ownership of intangible personal property is controlled by the whole law of the decedent’s domicile at the time of death.”49 This ill-advised stipulation, to which the attorneys for the estate inexplicably agreed, however, does not excuse the court from the responsibility it had to apply the law properly. It should not have simply accepted the severely flawed choice-of-law analysis stipulated by the parties.50

 A.     Why the Choice of the “Whole Law” Should Not Have Been Applied by the Court

When a litigant argues before a court that the law of a particular foreign jurisdiction, rather than the law of the forum jurisdiction should apply in determining the marital-property rights of a spouse, how should the forum court respond? Any number of answers to this question have been offered. One answer is that the court should generally not concern itself with foreign law, and should apply its own local law as the default rule, perhaps with certain, very limited, exceptions.51 Others, however, reject this default to local forum law. In contrast, they propose different methods for identifying the law to be applied. One such method requires the forum court to consider a variety of factors in deciding what law to apply, such as the expectations of the parties, efficiency, and the interests of the “international system,” in order to identify the law of the jurisdiction that is deemed to have the most significant relationship to the issue.52 A third approach proffered by certain commentators and courts is to weigh the relative stake of each of the relevant jurisdictions’ governmental interests.53 A fourth alternative to the default to local law asserts that there should be a reference to a set of black-letter rules that identify the law of a jurisdiction with a particular connection.54 A fifth alternative to the default to local law ignores the need to choose any jurisdiction’s specific law and rather argues for the fashioning of a new principle of substantive law from the best elements of the conflicting laws of the involved jurisdictions.55 While hardly an exhaustive list,56 each of these five approaches have achieved some prominence, because each rests on a distinctive perspective of the role of foreign law, if any, in resolving a dispute before a forum court.

With respect to the determination of ownership of property acquired by either spouse or both spouses during marriage, jurisdictions with developed systems of law have opted for one of three black-letter rules—partial mutability, strict immutability, or total mutability.57 In so doing, they have either neglected to consider or rejected those approaches considered above that involve broader policy considerations or jurisprudential musings rather than black-letter rules.58

However, this is not the end of the story. Regardless of the approach that has been endorsed, it does not define precisely what is encompassed by the “law” to be applied. Specifically, does “law” include within its meaning the conflict-of-laws rules of the jurisdiction chosen by the particular approach? Or, does the choice of a particular jurisdiction’s law reference only its domestic law, without consideration of its conflict-of-laws rules? When the First Circuit in Charania accepted the stipulation of the parties to the “whole law” of Belgium, they were agreeing that the conflict-of-laws rules of Belgium, and not just its domestic community-property law, should be applied.59

This decision by the First Circuit is a striking departure from the prevailing rule in state and federal courts of the United States, which is to apply only the domestic law of the jurisdiction indicated under the partial-mutability approach when issues of marital-property rights are involved.60 Had the prevailing approach been applied in Charania, the determinative law would have been Belgian domestic law. Under Belgian domestic law, Ms. Dhanani, the decedent’s spouse, would have obtained a community-property interest in the Citigroup stock at the time of its acquisition. The stock would not have been the decedent’s to dispose of at the time of his death. Rather, Mr. Charania would have owned only half of the shares, and his estate’s tax liability would have been considerably less than what was ultimately imposed. Meanwhile, the court would have avoided the highly demanding and error-prone task of implementing the conflict-of-laws doctrine of a foreign country, in this case, Belgium. Indeed, that endeavor is a primary source of the court’s analytically flawed opinion.

Curiously, it does not appear that the First Circuit even realized that it was departing from a long line of precedent and commentary that explicitly rejects a reference to the “whole law.”61 In effect, the First Circuit was led astray by assuming, that since the underlying matter was the transfer of property at Mr. Charania’s death, the conflict-of-laws principles that govern succession should first be consulted.62 Thus, the court began its analysis by stating that, since the decedent died domiciled in Belgium, Belgian law governs.63 Under broadly accepted choice-of-law principles, when the right of succession to intangibles like stock is involved, rather than marital property rights in such intangibles, the law of the decedent’s domicile at death controls issues relating to their disposition.64

What the court fails to consider is that before the proper disposition of the decedent’s estate can be determined pursuant to the domicile-at-death conflicts-of-law rule in matters of succession, the property that he owned at death must be determined.65 In order to make that preliminary determination, consideration must be given to whether the decedent’s spouse, on account of community property, owns an interest in property held in the deceased spouse’s name.66 To answer that question, an inquiry into the law governing the rights of a surviving spouse in such property, must be pursued. This necessarily requires identifying the appropriate choice-of-law approach. The First Circuit failed to do this.

Specifically, the First Circuit failed to consider the proper choice-of-law for marital property rights as distinct from the choice-of-law for matters of succession. In failing to do so, it made a critical error. Relying on the agreement of the parties that the succession rule that requires reference to the decedent’s domicile at death as controlling, the First Circuit, in its opinion, simply proceeds from the statement that Belgian law applies to matters of succession, to the bare conclusion that the “whole law” of Belgium, the domicile at the death of the decedent, applies as well for purposes of determining the spouse’s marital-property rights.67 It fashions itself as sitting as a Belgian court.68 Presuming that a Belgian court would apply the “whole law” of the nationality jurisdiction of the spouses at the time of their marriage pursuant to the strict-immutability approach, the First Circuit concluded that it was required to do so as well.69 In so doing, the First Circuit made two striking analytical errors—it failed to apply the partial-mutability rule observed by American courts and it proceeded to apply the “whole law” of the foreign jurisdiction indicated rather than its domestic law.

B.     United Kingdom Law is Not English Law

As just discussed, the First Circuit erroneously presumed that it must sit as a Belgian court and that such court would apply the “whole law” of the nationality jurisdiction. The First Circuit then proceeded to make another striking analytical error. In accordance with the stipulation of the parties, the court stated that, in “applying Belgian choice-of-law rules, [it] would look to the whole law of the country of the spouses’ common nationality,”70 adding, in agreement with the parties, “that the country of the spouses’ common nationality is England.”71

The footnote associated with this language states that the court “use[s] ‘England’ as a convenient shorthand for the United Kingdom. Both the decedent and his wife were born in Uganda, at a time when that country was part of the United Kingdom. The parties have stipulated that they were citizens of the United Kingdom at all relevant times.”72

The First Circuit appears to be completely unaware that Uganda was part of the British Empire, but never a part of the United Kingdom.73 More importantly for the present analysis, the United Kingdom and England are not synonymous.74 Indeed, the United Kingdom consists of more than one territorial unit with its own distinctive laws.75 It would be as if a reference to the law of the United States were automatically equated with the law of New York.

The problem arises because, Belgium, like certain other countries, required prior to 2004 the application of the law of the spouses’ nationality in determining matters of marital-property rights.76 If the nationality is that of the United Kingdom, as with Mr. Charania and Ms. Dhanani, there is a problem because the United Kingdom has no marital-property law, just as the United States itself does not have such law.77 Reference, therefore, must be made to a lawmaking political subdivision such as England or Scotland. However, there is no direction in United Kingdom law as to which political subdivision’s law to reference.

Just as the United Kingdom has not provided a solution to this problem, Belgium did not prior to 2004. In 2004, two years after Mr. Charania’s death, the newly enacted Belgian Code of Private International Law (“PIL”) addresses some of the problems that arise when the reference to a national law is to a nation with no substantive marital-property law of its own, but has multiple legal systems that have their own marital-property laws.78 Section 2 and the subsequent flush paragraph of Article 17 of the PIL directs reliance on the channeling rules in force under the foreign national law.79 If there are no such rules, as is the case with the United Kingdom, then the law “with which the natural person has the closest connections” or, if the law is “applicable to different categories of persons . . . the legal system that has the closest connections with the legal relationship.”80

But the PIL does not address the issue confronted in Charania—with which political subdivision of the United Kingdom the decedent and his wife have the “closest connections.” The answer must be none, since they had no palpable ties with any lawmaking political subdivision of the United Kingdom. Several of the Belgian drafters recognized this gap in the Code, but simply explained it away by noting that Belgium was not the only country with this problem on account of the nationality choice of law.81

Under these circumstances, what ought to be the solution when a court, such as the First Circuit in Charania, encounters the issue?82 I concur with conflict-of-laws theorist, John Delatre Falconbridge, who, building on the work of the Italian conflicts theorist Rodolfo De Nova, concludes that the most defensible solution is to stick with the domestic law of the forum (Belgium, under the court’s flawed decision to sit as a Belgian court), as any reference to the national law “is meaningless or ineffective in the circumstances.”83Applying this approach to the facts of Charania, Belgian domestic community-property law would apply. The PIL, Article 17, with its emphasis on the law with which the individual “has the closest connections” would also point to Belgian domestic law.84

The argument could be made that there is no need to turn back to Belgian domestic law, since none of the political subdivisions of the United Kingdom enforce a community-property regime.85 However, this is far from the end of the matter. Assuming England applies a strict-immutability approach, it would not apply its domestic law, but rather that of Uganda, where Mr. Charania and his wife had their first marital domicile.86 On the other hand, Scotland, another political subdivision of the United Kingdom, appears at the relevant time to have applied and may still apply a total-mutability approach.87 That approach would require application of the law of Belgium, not Uganda.88

 C.     The Haphazard Application of the “Renvoi” Theory by the First Circuit

The First Circuit in Charania asserted that, sitting as a Belgian court, it must apply “the whole law of the country of the spouses’ common nationality,”89 which it incorrectly concludes is England. But, assuming for purposes of discussion, the First Circuit’s correlation of United Kingdom law with English law is correct, the question remains as to how confident it can be as to what a Belgian court, not an American court, would determine is the “whole law” of England and how a Belgian court would determine that an English court would thereby apply the law of Uganda. It is submitted that a court in the United States should be exceptionally wary of any attempt to determine “law” by sitting as a foreign forum (i.e., Belgium) that expresses itself in a foreign language, that operates a distinct non-common law legal system that is populated by judges for whom the common-law method of adjudication is alien, and that invests case law and scholarship with different authority than a U.S. court would. This is especially the case when the U.S. court (the first jurisdiction) must determine how a Belgian court (the second jurisdiction) would conclude, sitting as an English court (the third jurisdiction), that an appropriate court of England, with a different legal system than Belgium, would apply Ugandan law (the fourth jurisdiction). Unfortunately, there is no serious indication in the First Circuit’s opinion of an appreciation of the immensely problematic nature of the process in which it has determined to engage.

Yet, let us also assume that the proper Belgian court would approach the issue exactly as the First Circuit does. In this case, the Belgian court would conclude that English conflict of laws requires the application of the strict-immutability approach to choice of law. Further, the strict-immutability approach would itself compel reference to the law of the marital domicile at the time of marriage—Uganda. Assuming, then, that Ugandan law should apply, how then does the First Circuit, a federal court in the United States, sitting as a Belgian court, reach its holding that English marital-property law, not Ugandan law, governs? The answer reveals yet other troubling errors in the First Circuit’s analysis. These errors are rooted in the court’s statement that “the parties have stipulated that Uganda’s marital property regime corresponded, at the relevant time,90 to England’s marital property regime.”91

The referenced stipulation of the parties was a very crucial and faulty one that the First Circuit should not have relied upon. First, it is substantively incorrect. Uganda’s marital-property regime at the time of Mr. Charania’s marriage, up to the present, is not clearly the same as England’s regime.92 Second, the First Circuit does not explain why its reference to Uganda’s marital-property law does not also include a reference to Uganda’s own conflict-of-laws rules. After all, a Belgian court must apply England’s whole law, which would include its rules relating to conflict of laws. England, it turns out, applies the same “foreign court” theory, whereby it sits as a court of the foreign jurisdiction referenced93—in this case Uganda.

If Uganda’s conflict-of-laws rules, were to be applied, what would be the result? The answer to this question would depend on which marital-property choice-of-law rule Uganda applied at the relevant time and whether the reference to a foreign law included that foreign law’s conflict-of-laws rules. If Uganda applied the strict-immutability approach, while rejecting a reference to its conflict-of-laws rules, then Ugandan domestic law, presumed to be the same as English law, would govern the rights of the spouses. On the other hand, if Uganda applied a partial-mutability approach while rejecting a reference to the foreign jurisdiction’s conflict-of-law rules, it would result in the application of Belgian domestic community-property law, rather than England’s separate-property law. In contrast, if Uganda looks to the “whole law” of Belgium, then this would require the application of English law, which would, in turn, based on strict immutability, look to Ugandan law, and so forth, resulting in a never-ending circularity.94

What is at issue here is the applicability of renvoi, although the First Circuit failed to even mention the word, or the fact that the First Circuit was running counter to the overwhelming tendency of American precedent and commentators rejecting any role for it, especially with regard to marital-property rights.95 Indeed, renvoi has now also been rejected by Belgian scholars.96 Not the least of renvoi’s many problems is one that resonates in connection with Charania—reliance on foreign-law experts whose inevitably conflicting conclusions cannot readily be adjudicated, especially when the esoterica associated with a foreign jurisdiction’s conflict-of-laws rules are at issue.97 Of course, the fact that renvoi suffers from very limited acceptance does not mean it should not be considered. However, the First Circuit in Charania engaged in absolutely no discussion of the topic.98

“Renvoi” means a “remission” back to the forum through the application by the forum of the foreign jurisdiction’s whole law, rather than just its domestic law.99 It was originally conceived as a means of assuring a reference back to the forum’s own law, when the forum applied a nationality rule and the country of nationality applied a domicile rule.100 Thus, the forum could give maximum weight to the contact point of nationality, while still justifying the actual application of its own domestic law. Many jurisdictions opting for renvoi only recognize this sort of one-step renvoi.101 Furthermore, the renvoi is only allowed back to the domicile (forum) jurisdiction’s domestic law and not its whole law. Otherwise, the circularity problem described above would again be encountered. Additionally, one-step renvoi does not recognize a “transmission” to a third jurisdiction, which might be indicated if the nationality jurisdiction in the previous example looked to the law of the habitual residence.

In theory, however, there is no logical basis for cutting off renvoi in these ways.102If the assumption is that a reference to another jurisdiction’s law should include its whole law, then why stop at the first jurisdiction referenced? Some jurisdictions, concerned about the inefficiencies and potential for error in references to a chain of foreign laws, but not prepared to settle on a theoretically impure, one-step renvoi, limit the renvoi to two-steps, thus permitting a transmission to a third jurisdiction’s domestic law.103 However, there is no persuasive theoretical justification for a two-step limit any more than there is for a one-step limit.

The First Circuit in Charania, without explanation, applied a two-step renvoi. In doing so it did not acknowledge that it was departing from established American conflict-of-laws principles. Moreover, though purportedly sitting as a Belgian court, it offered no consideration of renvoi under Belgian law. In fact, Belgium’s approach to renvoi has oscillated considerably, including at the relevant times for determining the marital rights of Mr. Charania’s spouse. Thus, there can be no confidence that the First Circuit applied the whole law of Belgium, including renvoi, as a Belgian court would have done so.

D.     Ugandan Law is Not Identical to English Law

Assume, for purposes of discussion, that at the relevant time, Belgium applied two-step renvoi. This should result in the marital-property rights of Ms. Dhanani being determined under Ugandan domestic law, without regard to its conflict-of-laws rules. Instead of applying Uganda’s law, as discussed above, the court conflates Ugandan and English domestic marital-property law by stating that:

because the spouses were domiciled at the time of their nuptials in Uganda, and because the parties have stipulated that Uganda’s marital property regime corresponded, at the relevant time, to England’s marital property regime, the doctrine of immutability would call for application of England’s marital property regime. That is a separate property regime . . . .104

Again, the default to English law is without foundation. However, let us assume for purposes of discussion that it is correct. Apparently, because of the parties’ stipulation, the First Circuit did not feel the need to offer support for the assertion that Uganda’s law “corresponded” to English marital-property law in 1967 when the spouses married in Uganda. This is unfortunate, because it is far from clear that the stipulation is correct. Precisely what Ugandan marital-property law was at the time the Charanias married, which was after Uganda gained its independence, is exceptionally difficult to determine.105

Uganda received via the East Africa Order in Council “the substance of the common law, the doctrines of equity, and the statutes of general application in force in England on August 12, 1897.”106 English law at the time and to date recognizes separate property ownership and does not recognize community property.107 Thus, the First Circuit appears, at first glance, to have serendipitously reached the correct answer in stating that Ugandan law corresponded to English law. But this facile conclusion does not hold up upon deeper consideration. In fact, English law was residual law in Uganda, subject to the proviso that English law should be in force only insofar “as the circumstances of the Protectorate [of East Africa, including Uganda] and its inhabitants and the limits of His Majesty’s jurisdiction permit, and subject to such qualifications as local circumstances render necessary.”108

For example, religious affiliation often proves crucial in determining property rights under Ugandan law. The First Circuit was, apparently, not aware of this fact. We are not informed of the religion of the Charanias. If they were Muslim, then English common law definitely would not have applied to them, as there was and is a separate legal regime in place for Muslims.109 If they were Hindu, the law is quite murky.110 To begin with, there is not a uniform Hindu marital-property law. Not only do different sects from different regions of India observe different rules, but Hindu law, as applied to locals in the British East African protectorates of Uganda, Kenya, and Tanganyika (now Tanzania) included unique rules reflecting local conditions.111 Furthermore, there were also local statutory enactments.

Seeking a unified approach, Uganda enacted a Hindu Marriage and Divorce Act of 1961, which remains in effect to this day.112 The law seeks to regulate marriages and matrimonial causes of Hindus, Sikhs, Jains, and certain Buddhists. Specific conditions justifying divorce are addressed in the statute, but the precise issue of marital-property rights is not. Subsequent attempts to clarify the ambiguous situation with regard to marital property appear to have failed, thereby leaving the substance of the rights of Hindu couples in abeyance and unsettled.113 Indeed, as there is no statute that specifically addresses Hindu marital-property rights, the secular marital-property law arguably applies. The substance of that law is highly uncertain and no precise rules have developed to date.114 In 2013, the Ugandan Supreme Court confirmed this conclusion in Rwabinumi v. Bahimbisomwe.115 Prior to the Supreme Court’s decision, in 2008, the Uganda Court of Appeals had ruled that, since the adoption of the Uganda Constitution in 1997, all property brought into the marriage, as well as that acquired during marriage, was “joint property,” that is, each spouse had an automatic 50% interest.116 Of course, Mr. Charania acquired the Citigroup shares in 1997.117

However, Rwabinumi reversed this ruling, and held that there is no absolute right under the Constitution in support of the Court of Appeals’ position.118 Nevertheless, in its opinion the Supreme Court did not actually address what the law is. The court did agree with Muwanga v. Kintu, an unreported case decided in the same year Mr. Charania acquired the Citigroup stock, that that decision “rightly pointed out the challenges that courts will continue to face in determining what constitutes matrimonial property in Uganda.”119 In this regard, Rwabinumi implores the “[Ugandan] Parliament to enact a law that clearly defines what constitutes marital/matrimonial property as opposed to individually held property of married persons.”120

The Uganda Supreme Court’s opinion in Rwabinumi hardly establishes that Uganda had in force the categorical English rule of separate property. Rather, it demonstrates that the stipulation of the parties in Charania was a superficial statement of Ugandan law and that, whether in 1967, 1997, or in 2002, there simply may not have been a settled marital-property law in Uganda. To the extent that is the case, there was no basis for the First Circuit’s reference to English law on the basis that English law and Ugandan law correspond. Under the circumstances, it seems far more likely that a Belgian court would reference Belgian domestic law, as there was no other definitive law to reference with respect to the respective property rights of long-term Belgium-domiciled spouses.

The foregoing consideration of Ugandan domestic marital-property law was required under the assumption of a Belgian two-step renvoi. But what if the proper Belgian court would have applied total renvoi? It would then be required to consider Uganda’s marital-property choice-of-law rule. To the extent that Uganda’s choice-of-law rule corresponds to English law (which is hardly a safe assumption) the argument could be made that the rule of strict immutability applies, as per the House of Lords decision in the case of De Nicols v. Curlier.121 The problem is that De Nicols (the case the Charania court heavily relies upon to establish that the English approach is strict immutability) was decided by the House of Lords in 1900.122 This was two years before Uganda officially received English common law, but this does not mean that De Nicols became part of Ugandan common law. The reason is that the legislation receiving English common law had a cutoff date of 1897, three years before the De Nicols decision.123 Prior to De Nicols, the House of Lords had not settled on the strict-immutability approach. Indeed, common law at the time suggested a rule of total mutability rather than strict immutability. The very same House of Lords that decided De Nicols had a century earlier decided a case (Lashley v. Hog)124 that was unquestioned at the time. This case held that when an Englishman, who had acquired property while domiciled in England, died domiciled in Scotland, the property that had been separate property, became subject at his death to the Scottish communio bonorum, a loose form of community property.125 Indeed, Lord Eldon explicitly stated that marital-property rights were mutable and changed with changes in domicile.126

Furthermore, while the strict-immutability approach of De Nicols would appear to be in direct conflict with Lashley,127 this may not, in fact, be the case. The Lords deciding the matter had diverse approaches to distinguishing the two cases, as have leading commentators.128 The Lord Chancellor distinguished the two cases principally on the ground that Lashley did not involve community property at all, as Scottish law’s communio bonorum was not really community property.129 Thus, the rule of decision in Lashley was not in conflict because it was simply applying the rule that in matters of succession, as opposed to marital-property rights, the final domicile of the decedent, controls.130 This is the position that the First Circuit in Charania strenuously adopted in distinguishing Lashley.131 In so doing, it totally neglected the opinions of the other Lords in De Nicols, as well as the views of many distinguished commentators.132

In fact, at the time that De Nicols was decided, the outcome in that case was deemed by the prestigious Juridical Review as distinguishable from Lashley on the ground that there was an implied marital contract in De Nicols based on the unique French law approach at play in that case.133 Indeed, this had been the position taken by several of the Lords, other than the Lord Chancellor.134Dicey on the Conflict of Laws, the leading English treatise on conflict of laws, which had initially contended that Lashley was a succession case, had subsequently altered its position.135 The First Circuit in Charania acknowledged the change in position of Dicey.136 However, it noted a still later edition of the treatise, the 14th edition, in which Dicey shifts yet again back to the view that strict immutability is the rule in England on the authority of De Nicols and that Lashley is a succession case.137

Notwithstanding the First Circuit’s reliance on the interpretation of the cases set forth in the 14th edition of Dicey, the 14th edition was not published until 2006, four years after Mr. Charania’s death.138 More pertinent is the position of Dicey in 1967 when Mr. Charania and Ms. Dhanani married and in 1997 when the Citigroup stock was acquired by Mr. Charania.139 The most recent edition prior to the acquisition of the stock, the 8th edition, consistent with editions going back to the 5th edition in 1932, known as the Keith edition, as well as the 13th edition published in 2000 and in effect at Mr. Charania’s death in 2002, all recognized the English approach as being one of total mutability, in accord with Lashley. De Nicols was essentially cast as a narrow exception to the general rule of mutability, based on the French law of implied contract.

Of course, the preoccupation with Dicey is itself questionable and strange to lawyers in the United States, where treatises are routinely consulted, but hardly are regarded as authoritative and certainly are not deemed conclusive. Moreover, Dicey was not and is not the only respected English commentary on the case, although the First Circuit did not appear to be aware of this. While the original characterization of Lashley as a “succession case” not in conflict with De Nicols received support from several other commentators as well,140 certain commentators, including the leading competitor to Dicey, that is, Cheshire, Private International Law, endorsed the French implied-contract theory as the proper reading of De Nicols,141 thereby treating it as a narrow exception to Lashley, rather than the case that overruled it and instituted the strict-immutability approach in England. Yet, just as Dicey has equivocated, so has Cheshire. This treatise eventually abandoned its broad reading of Lashley in favor of the “succession case” reading originally propounded and then abandoned by Dicey,142 who switched, during the relevant time, to Cheshire’s original view that total mutability was the controlling English rule.

Other highly respected scholars have also expressed alternative readings of the cases and just what the English choice-of-law rule is. For example, Martin Wolff, in his thoughtful work, Private International Law, concluded that the English rule is that when the marital domicile shifts from a common-law property to community-property jurisdiction, total mutability applies, arguably, the Lashley holding.143 On the other hand, if there is a shift from a community-property law to common-law-property jurisdiction, strict immutability applies, arguably the De Nicols holding.144 Applying Wolff’s statement of English law, an English court would decide that total mutability applied in Charania, as Mr. Charania and Ms. Dhanani had a marital domicile at the time of Mr. Charania’s death in Belgium, a community-property jurisdiction.

What all this suggests is that at times relevant to the determination of Mr. Charania’s ownership interest there was considerable division as to the English marital-property choice-of-law rule. But even if the reading of De Nicols by the First Circuit was the settled English law, this would not mean that it was the controlling choice-of-law rule in Uganda. Again, Ugandan law did not receive the De Nicols decision into its common law. While Ugandan courts could choose to follow it, they would not be required to do so, as they would have been had it been in effect in 1897.145 Alternatively, they might consider themselves bound by the interpretation of Lashley as instituting total mutability, the interpretation that was widely in effect before the De Nicols decision was handed down in 1900. Admittedly, no one can be certain, as there is no record of either of these decisions having been cited or otherwise relied upon by Ugandan courts or incorporated into Ugandan legislation.

In sum, whether English or Ugandan “whole law” were applied by the First Circuit sitting as a Belgian court, it is clear that the First Circuit’s conclusion that the English domestic regime of separate property applies is not sustainable. To the extent the renvoi stops at Ugandan domestic law, it is entirely unclear what that law was or is currently. If total renvoi applies, the outcome seems to point in the direction of total mutability and Belgian domestic law of community property. Likewise, if the First Circuit had applied the standard choice-of-law rule applied typically in the United States by federal and state courts—the Restatement (Second) of Conflicts of Laws rule of partial mutability without renvoi—then Belgian domestic law of community property would apply.

E.     The Application of Strict Immutability to Exiles

The estate in Charania argued that Ugandan law (stipulated as corresponding to English law) should not apply because the decedent and his spouse had been exiled from Uganda in 1972, along with the other Ugandans of Asian descent.146 All of their property had been expropriated.147 The Citigroup stock, along with other valuable property, had been acquired after being exiled and while resident in Belgium.148

Following the overthrow of Amin in 1979, Uganda revoked the laws that had been enforced against Asians and welcomed them back.149 The decedent and his spouse did not return.150 This strongly indicates their endorsement of a newly acquired Belgian marital domicile and their governance under its property rules, as well as a decisive repudiation of ties to Uganda and a property-law system that had permitted the expropriation of all that they owned. However, under the First Circuit’s blinkered analysis, the Charanias remained tied to the property law of the offending original marital domicile.151

The First Circuit was dismissive of the Charanias’ argument that the law of the country that exiled them should not determine their marital-property rights, especially with respect to subsequently acquired property. Indeed, the First Circuit stated that “it is far from clear that an English court would necessarily view this distinction[152] either as meaningful or as cutting in favor of adopting a rule of mutability.”153 Apart from this statement, the First Circuit offered no analysis.

The argument the estate presented, however, highlights a rather disturbing weakness of the strict-immutability approach—that spouses must remain bound to the property law of the very jurisdiction that has expelled them and expropriated their property. Worse, this is the case even if the expropriation was indisputably discriminatory in blatant violation of fundamental principles of international law, as the facts of Charania reveal. Ironically, the court failed to consider how Dicey, which the court regarded as the ultimate authority when it comes to marital-property choice of law,154treats involuntary exiles. Had it done so, it would have discovered that this authority, while not addressing exile specifically in the context of marital-property rights, concludes, that in the case of involuntary exile, the exiled person should be delinked from the offending domicile.155 In reaching this conclusion, the treatise relies on case law that lends support to the conclusion that strict mutability should not be applied by an English court in situations like Charania.156

The First Circuit does respond as well by pointing out that the spouses were free under Belgian law to elect community property and failed to do so. Of course, this would have required them to anticipate the thoroughly flawed conflicts analysis of the First Circuit while they were still alive, a not insignificant challenge of foresight and predictive capability.

 Finally, the First Circuit incorrectly shaped the issue as a binary choice between strict immutability and total mutability. However, the allowance of a one-time exception based on involuntary exile would not constitute a repudiation of the strict-immutability approach and adoption of the total-mutability approach at all. By simply deeming the first marital domicile after exile and expropriation to be immutable, the preference for strict immutability could be preserved, but not at the price of an entirely mechanistic and blatantly unjust outcome. This approach would also have spared the estate the heavy tax burden it was forced to endure as a result of the First Circuit’s flawed analysis.

F.     Erie, Federal Questions, and the Proper Choice of Law by a Federal Court

As previously explained, the First Circuit never actually considered what choice-of-law rule it should apply to the question of marital-property rights. It simply concluded that since the decedent died domiciled in Belgium, Belgian conflict-of-laws rules regarding marital-property rights should apply. This led it to English law and the decision in De Nicols v. Curlier, which the court construed as imposing the strict-immutability approach. In fact, the First Circuit, first and foremost sitting as a federal court, should have recognized the Erie/Bosch issues implicated by its approach. In Erie Railroad Co. v. Tompkins,157 the Supreme Court held that a federal court with diversity jurisdiction is bound to apply the substantive law of the state in which the federal trial court sits.158 The court explained that “[e]xcept in matters governed by the Federal Constitution or by Acts of Congress, the law to be applied in any case is the law of the State.”159

Erie did not address whether the federal courts in diversity actions were also required to apply the state’s law regarding the conflict of laws. If not, could the federal court apply a distinct federal choice-of-law rule when the conflict of laws was involved? In Klaxon Co. v. Stentor Electric Manufacturing Co., the Supreme Court held that a federal court was, indeed, required to apply the conflict-of-laws rules of the state in which it sits.160 The Court’s opinion explained that if the state’s conflict-of-laws rules were not also applied, the “principle of uniformity within a state, upon which the Tompkins decision is based” would be undermined.161 For example, if a federal common law of conflicts could be applied by the federal district court, and it resulted in a different choice of governing substantive law than the conflicts rules applied by state courts, the state’s citizens might be exposed to differing substantive outcomes depending on which court within the state’s geographic bounds served as the forum. This would violate Erie’s edict regarding uniformity.162 In addition to undercutting uniformity, it would encourage forum shopping.

The principles set out in Erie and Klaxon, and their progeny, were not extended to “matters governed . . . by Acts of Congress,”163 such as the IRC. Thus, the question presents itself, what choice-of-law principles apply when a tax dispute is before a federal district court, the Court of Federal Claims, or the Tax Court, as in Charania? These three are the exclusive fora for resolving disputes involving federal taxes between a taxpayer and the federal government. Importantly, federal district courts and the Court of Federal Claims, can only hear refund claims.164 That is, the taxpayer must pay the tax assessment before commencing refund litigation against the IRS. On the other hand, the Tax Court can decide a taxpayer challenge prior to the taxpayer’s having to pay the tax assessment.165

The ownership rights of spouses can have a direct impact on their respective federal transfer tax and income tax liability. However, the law concerning the marital-property rights of spouses is not set forth in the IRC, and, thus, the substantive law must be derived from some other source. Ordinarily, federal tax law looks to state law to determine such substantive marital-property law.166 But there is astonishingly little examination of what the choice-of-law approach should be in arriving at the governing substantive law of marital-property rights.

When a federal bankruptcy matter is at issue, analogous choice-of-law questions arise. Here, too, there is an act of Congress that must be enforced, and reliance must be placed on state substantive property laws, including those relating to marital-property rights. Although there are not specialized courts like the Court of Federal Claims and the Tax Court, the federal district courts, as with federal tax controversies, do serve as a proper forum. In the federal bankruptcy setting, there has been a good deal of consideration of this conflict-of-laws question. A sharp division among the circuits exists. In particular, the Ninth Circuit has opted for application of a distinct federal common law.167 The primary justifications for this approach are that forum shopping will be discouraged and “[t]he value of national uniformity of approach need not be subordinated, therefore, to differences in state choice of law rules.”168 These values are important, as Congress has enacted a national bankruptcy law.169

Indeed, with respect to federal taxation, one can imagine a situation in which two similarly situated taxpayers are seeking a tax refund in federal district courts in different circuits. If choice-of-law rules of the states in which the taxpayers reside differ, then the taxpayers could incur different federal tax liabilities on identical facts. This would violate the overriding tax law principle that persons similarly situated ought to be taxed similarly.170

But how then do courts urging application of state choice-of-law rules justify their stance? In In re Gaston & Snow,171 a federal bankruptcy case, the Second Circuit explained:

We recognize the concerns expressed by these courts, but do not believe they implicate significant enough federal interests to justify the creation of federal common law in this case. While an interest in uniformity can justify the creation of federal common law, Klaxon rejected the need for uniformity as a justification for displacing state conflicts rules. Regarding the federal interest in avoiding forum shopping, we believe there are only a limited number of cases in which this interest is implicated. Here, for example, where the debtor was brought into bankruptcy court through an involuntary petition, it is difficult to argue that any forum shopping occurred.172

In the federal tax setting (unlike the involuntary bankruptcy setting) the taxpayer alone determines the forum, and has several choices.173 Thus, an opportunity for forum shopping is built into the system. That opportunity most typically hinges on whether the taxpayer is able and prepared to pay the tax assessment first or not. If not burdened by the need to go to Tax Court because of a lack of resources to pay the tax assessment first, the taxpayer will be free to consider which of the three fora has more favorable substantive tax opinions or more sympathetic judges, as well as whether a jury trial is desired, which is only available in the federal district courts. However, as long as there is a uniform rule regarding choice of law by federal tax fora, there will be no real opportunity to shop for a more favorable state law. Whether the forum is a federal district court, the Tax Court, or the Court of Federal Claims, the unitary conflict-of-laws rule will point each of these courts to the same controlling state law. As a result, the law determining the respective ownership rights of spouses with respect to marital property will be the same on the same facts regardless of the federal forum chosen.

On the other hand, if resort is made to the conflict-of-laws rule of the state in which the district court hearing a federal tax matter sits, the theoretical possibilities for different outcomes on the same facts increases significantly. Furthermore, if the Tax Court and the Court of Federal Claims, though national courts headquartered in the District of Columbia,174are also bound to invoke state conflict-of-laws rules, the possibilities are further magnified. Not surprisingly, these courts have tended strongly toward a federal rule based on the Restatement (Second) of Conflict of Laws.175

In re Gaston & Snow fails to explain in any meaningful way why the interest in uniformity should not, in fact, be accorded primacy. Indeed, Erie and Klaxon make clear that when an act of Congress is involved the same federalism calculus is not in play as in the diversity setting. When an act of Congress, such as the IRC, is at issue, the particular stake that a state has in uniform enforcement of its laws by courts within its geographic borders does not exist. Because state legislators and courts do not deal with federal tax matters and would not craft their choice-of-law rules to further federal interests in achieving uniformity of federal tax result, deference to state law is in most circumstances unwarranted.

In practice, federal courts and state courts have not been at odds when it comes to choice of law. Federal courts have generally looked to the Restatement (Second) of Conflict of Laws. In the context of marital-property rights, the Restatement (Second) rules as to choice of law are largely derived from consistently enforced state court decisions.176 Whether federal common law or state choice-of-law rules are applied, the partial-mutability approach reigns.177 As this Article demonstrates, the First Circuit in Charania, without obvious awareness that it was doing so, adopted an approach urged by the litigants that implicitly rejected this dominant approach. By opening up property-law determinations that underlie tax determinations to the diverse choice-of-law rules of foreign jurisdictions, it has effectively rejected a uniform choice-of-law rule for federal courts in federal tax matters.

G.     The Significance of Commissioner v. Estate of Bosch

The preceding consideration of Erie and its progeny and more generally, the interaction between conflict-of-laws and federal statutes, is not the end of the matter. The First Circuit essentially deduced that, since it was sitting as a Belgian court, it owed deference to Belgian conflict-of-laws rules because in a diversity matter it would owe deference to the conflict-of-laws rules of the state in which it sits. As support, the First Circuit’s opinion actually cites two federal court decisions for the proposition “that, when called upon to apply state law, a federal court should normally ‘take state law as it finds it.’”178 The two cases cited, A.W. Chesterton Co. v. Chesterton and Kassel v. Gannett Co., however, were diversity cases and not a matter governed by an act of Congress.179 They were also cases in which the federal court could look to a prior determination of the highest court of the state with respect to the issue being litigated.180 In Erie, the Supreme Court indicated that federal courts are bound by the acts of the legislature and the “highest court” of the state.181 The “proper regard” to be shown lower courts, as well as the standard for determining whether the highest court had decided the actual issue before a federal court, and whether the deference owed to state courts applies to courts of foreign countries, were not addressed.182 These questions are important in evaluating the Charania opinion, inasmuch as the First Circuit never addressed them.

Another very troubling omission by the court is its utter disregard of the most critical precedent bearing on the questions posed in the preceding paragraph, that is, the United States Supreme Court’s decision in Commissioner v. Estate of Bosch.183 In that case, which goes unmentioned in Charania, the Supreme Court confronted the specific question whether the decrees of lower courts, in Bosch a probate court, are “binding on a federal court in subsequent litigation involving federal revenue laws.”184 The Court “h[e]ld that where the federal estate tax liability turns upon the character of a property interest held and transferred by the decedent under state law, federal authorities are not bound by the determination made of such property interest by a state trial court.”185 Endorsing language in King v. Order of United Commercial Travelers of America (a post-Erie opinion), the Supreme Court in Bosch states: “Moreover, even in diversity cases this Court has further held that while the decrees of ‘lower state courts’ should be ‘attributed some weight . . . the decision [is] not controlling . . .’ where the highest court of the State has not spoken on the point.”186 Furthermore, “[i]f there be no decision by that court then federal authorities must apply what they find to be the state law after giving ‘proper regard’ to relevant rulings of other courts of the State.”187

Since Bosch, lower federal courts have not been especially concerned with the “proper regard” requirement. In an empirical study of cases through the end of 1992, Paul L. Caron determined that virtually all federal courts either engage in de novo review of state laws, simply ignore state law, or pay mere “lip service” to state law.188 While highly critical of this “subversion,” even Professor Caron recognizes the primacy of the revenue concern, at least when the taxpayer has not been a party in a lower state court decision subsequently offered as binding precedent in the federal tax proceeding.189 Following up on Professor Caron’s study, I have examined decisions since 1992 reported by Westlaw and have found that lower federal courts are still following one of the three approaches to “proper regard” that were identified by Professor Caron. Thus, regardless of the precise meaning given to “proper regard” by federal courts in tax matters, it is clearly not the “strict” regard that the First Circuit in Charania concluded was necessary.190

Moreover, Bosch involved state trial and probate court decisions. At stake was the proper respect the federal system must show lower state courts when federal courts are required to reference state substantive law in order to determine federal tax consequences. In Charania, the First Circuit seemed entirely unaware that the federalism concerns at stake in Bosch, as well as Erie and Klaxon, are not at stake when the law of a foreign country, which is not part of the federal system, is being referenced. There is no reason for the same degree of solicitude to be shown. In light of this, the Charania court’s obeisance to the 1900 House of Lords decision in De Nicols v. Curlier makes little sense.191

Indeed, the First Circuit concluded that it should put itself in the position of a Belgian court.192For such a court, U.S. federalism concerns and the emanations of Erie would not be pertinent. Moreover, under the Belgian legal system, as with most civil-law legal systems, precedent does not carry the same weight as under a common-law system.193Often, scholarly authority is more significant than lower court decisions.194 With time, prior decisions may be deemed to have become antiquated and not deserving of continuing respect.195

Even assuming that U.S. federalism concerns do apply, and on those grounds, respect must be shown by a U.S. federal court deciding U.S. federal tax matters, to the law of Belgium, and its conflict-of-laws rules, there is simply no way to know how the highest court of Belgium would construe De Nicols v. Curlier—a conflicts-of-law decision by the United Kingdom’s House of Lords from more than a century ago. The rationale of the decision, as well as the opinion’s continuing authority, have been strenuously debated by commentators in England and elsewhere for more than a hundred years.196

In any event, the fact that De Nicols was decided by the highest court of the United Kingdom at the time, hardly demands slavish adherence to it even if the federal courts were required to show the same deference to foreign high courts as to state high courts under Bosch. Significantly, one of the litigants in Charania, the IRS, had already taken a position in several administrative rulings opting for an extremely narrow reading. Indeed, the IRS has maintained that it is not necessarily bound to adhere to the decisions of the highest court of a U.S. state. The Service has reasoned that Bosch:

did not say that the decision of the highest court of state is always binding; seemingly in a federal tax case, such decisions remain vulnerable to the charge that they emerged from a non-adversary proceeding. . . . A second reason for giving no weight . . . is found in Lanigan v. Commissioner, 45 T.C. 247 (1965). . . . The tax court in a well-reasoned opinion based on substantial authority, held that the practice of following a state court adjudication of a property interest does not apply when the decree has no effect under state law as a determination of such an interest.197

This same reasoning has been echoed in other administrative determinations, including Revenue Ruling 69-285, in which the Service stated that “[a] state court decree is considered to be conclusive in the determination of the Federal tax liability of an estate only to the extent that it determines property rights, and if the issuing court is the highest court in the state.”198 The requirement that property rights must have been adjudicated is based on language in Bosch wherein the Supreme Court states that the question before it is the effect of “a state trial court adjudication of property rights or characterization of property interests.”199 The emphasis on what was decided by a state high court was explained further in General Counsel Memorandum 39,183,200 which detailed the reason behind the Service’s narrow reading of Bosch. It stated that:

[S]ince there is no federal law of property . . . the question of the extent of the decedent’s interest in property is a matter of state law. If a state decision does not determine the extent of the decedent’s property interest, such decision is simply not relevant for federal estate tax purposes.201

De Nicols, the principal case on which Charania is grounded, did not decide the extent of property rights. Rather, it decided the conflict-of-laws question, that is, what law should determine the extent of property rights.202 There is a serious question whether the IRS’s administrative authorities should have been brought to the attention of the First Circuit, as they clearly are in conflict with the view that a federal court is bound by the high court of a “state.” Again, this further assumes the word “state” includes the foreign country of Belgium.

Apart from the tax administrative authorities cited, federal courts jurisprudence makes clear that a federal court need not be a “ventriloquist’s dummy” in applying the law of a state, even if there is a state high court decision.203 In Bernhardt v. Polygraphic Co. of America, the Supreme Court required application of the decision of the Vermont Supreme Court precisely because “there appears to be no confusion in the Vermont decisions, no developing line of authorities that casts a shadow over the established ones, no dicta, doubts or ambiguities in the opinions of Vermont judges on the question, no legislative development that promises to undermine the judicial rule.”204 In other words, “Bernhardt implicitly recognized that federal courts should be sensitive to possible changes in state law that they are required to apply in deference to the Erie principle.”205 When considered in light of Bernhardt and subsequent authority following it, De Nicols appears as a perfect candidate for reconsideration by federal courts, and not one requiring slavish adherence, as per the First Circuit’s conclusion that it was “bound to adhere to the rule of De Nicols absent a compelling showing that the English courts would scuttle that rule.”206

IV.     Conclusion

This Article’s primary objective in considering the Charania decision is to expose the many analytical errors the First Circuit committed as it engaged in the choice-of-law process. The sheer quantity and severity of the errors offer striking evidence of the risk of a seriously flawed outcome that stems from a court unnecessarily—and in opposition to existing precedent—engaging incautiously in the conflict-of-laws process, especially when international elements dominate. For those with cross-border wealth, Charania leaves matters very much unsettled—it shows that normal expectations of spouses as to ownership and tax liability may not be proven accurate and outcomes may simply not be determinable in advance, a highly inefficient result.

Going forward, Charania’s flawed analysis must not be permitted to gain precedential value. If that occurs, the longstanding, workable, and predictable conflict-of-laws rules that have been almost uniformly applied to date by courts in the United States with regard to conflict of laws involving marital property would be jeopardized. Most notably, the impulse to sit as a foreign court needs to be curbed, especially to the extent it entails application of renvoi. The partial-mutability approach, coupled with a reference only to the foreign domestic law, achieves this goal. Likewise, a uniform set of federal choice-of-law principles, such as one based on the Restatement (Second) of Conflict of Laws, should be preferred in matters of property ownership, especially affecting federal taxation. The acknowledgement of the inapplicability of Erie, Klaxon, and Bosch is also critical so as to avoid slavish adherence to often poorly understood foreign law in situations that do not present the domestic concerns central to those decisions. On facts similar to Charania, courts should favor the single remission to the well understood domestic law of a country with a highly developed legal system, like Belgium. It is much more likely to yield a fair and predictable result, and one that constitutes a more accurate determination of foreign law, than serial applications of the whole law of a variety of jurisdictions.

In light of the complexities and pitfalls associated with U.S. courts looking abroad for controlling law, a skeptical eye must be brought into play whenever a departure from forum law and application of a foreign law is urged. When reference is to be made to a foreign law at a national level and no such law exists, as is true of the United Kingdom, the reference to the foreign law is even less defensible.

Lastly—at least in the context of marital-property law—the venerable doctrine of partial mutability, coupled with rejection of renvoi, does appear to yield a result that comports with spousal expectations. Certainly, this is true in the involuntary exile setting, as in Charania, where the shortcomings of strict immutability are especially striking. Particularly, when the vital interests of efficient and consistent tax assessment are at stake as well, as in Charania, the case for the partial-mutability rule, without renvoi, is all the more compelling.


  1. [1]. These would be the community-property states: Arizona, California, Idaho, Louisiana, Nevada, New Mexico, Texas, Washington, and Wisconsin. See Robert H. Sitkoff & Jesse Dukeminier, Wills, Trusts, and Estates 520 n.1 (10th ed. 2017). “Alaska, South Dakota, and Tennessee offer a community-property option.” Id.

  2. [2]. For consideration of this topic, see 1 Jeffrey A. Schoenblum, Multistate and Multinational Estate Planning § 10.21[C]–[D] (2010).

  3. [3]. Note that even if spouses are moving from one community-property jurisdiction to another, those jurisdictions may have different community-property rules, so that it will still be necessary to determine which jurisdiction’s law is controlling in determining what rights the spouse has. Id. § 10.21[D].

  4. [4]. See Saul v. His Creditors, 5 Mart. (n.s.) 569, 603–08 (La. 1827). The holding was endorsed by Joseph Story, who stated the rule as follows: “[W]here there is no such express nuptial contract, the law of the matrimonial domicil is to prevail, as to the antecedent property; but the property acquired after the removal is to be governed by the law of the actual domicil.” Joseph Story, Commentaries on the Conflict of Laws, Foreign and Domestic, in Regard to Contracts, Rights, and Remedies, and Especially in Regard to Marriages, Divorces, Wills, Successions, and Judgments 284–85 (4th ed. 1852); see Harold Marsh, Jr., Marital Property in Conflict of Laws 108–09 (1952).

  5. [5]. See, e.g., Marsh, supra note 4, at 108 (explaining the shift in which law governs depending on when and where assets are acquired); Schoenblum, supra note 2, § 10.21[C] (explaining how different laws may govern different assets depending on time and place of acquisition).

  6. [6]. Marsh, supra note 4, at 108–10.

  7. [7]. Schoenblum, supra note 2, § 10.21[C].

  8. [8]. Id. “Characterization” has been criticized as an escape device from rigid choice-of-law rules. See, e.g., Lea Brilmayer & Raechel Anglin, Choice of Law Theory and the Metaphysics of the Stand-Alone Trigger, 95 Iowa L. Rev. 1125, 1134–35 (2010) (describing how courts used characterization in suits to avoid otherwise strictly observed choice-of-law black-letter rules). But see James Y. Stern, Property, Exclusivity, and Jurisdiction, 100 Va. L. Rev. 111, 161–63 (2014) (endorsing the situs rule and arguing that the breadth with which the situs rule is applied to real property tends to reduce characterization problems).

  9. [9]. See Walter Pintens & Jens M. Scherpe, Matrimonial Property, in Encyclopedia of Private International Law ch. M-7, § 3(b)(3) (Jűrgen Basedow et al. eds., 2017).

  10. [10]. Marsh, supra note 4, at 106–08.

  11. [11]. Switzerland is a leading example of a foreign jurisdiction that applies total mutability. See Philippe Pulfer & Azadeh Djalili, Acquiring Domicile in Switzerland: Consequences of the Matrimonial Property Regime, 17 Tr. & Trustees 323, 325–26 (2011). The principle is also known as “conflit mobile.” See generally François Knoepfler et al., Droit International Privé Suisse 115–18 (2004).

  12. [12]. Marsh, supra note 4, at 104–06.

  13. [13]. See Schoenblum, supra note 2, § 10.24[A]; see also Friedrich K. Juenger, Marital Property and the Conflict of Laws: A Tale of Two Countries, 81 Colum. L. Rev. 1061, 1071 (1981) (discussing why the United States adopted the partial-mutability approach); Robert A. Leflar, Community Property and Conflict of Laws, 21 Calif. L. Rev. 221, 229 (1933) (“The ‘source doctrine’ under the law of any given state is equally applicable to property acquired with other property just brought into the state, and to property acquired with other property which has been in the state all along.”).

  14. [14]. There are numerous decisions and administrative rulings that involve income tax issues. See, e.g., Crawford v. United States, 4 Cl. Ct. 699 (1984); Santiago v. Comm’r, 61 T.C. 53 (1973), aff’d, 510 F.2d 223 (D.C. Cir. 1975) (per curiam); Rev. Rul. 59-199, 1959-1 C.B. 182. Although lingering in the background, there is little, if any, discussion of the varied conflict-of-laws approaches to marital-property rights and which is controlling.

  15. [15]. Estate of Charania v. Shulman, 608 F.3d 67, 68–69 (1st Cir. 2010). The First Circuit affirmed with respect to the substantive issues, but reversed the Tax Court with respect to the imposition of certain penalties. Id. at 77.

  16. [16]. For example, the opinion relied heavily on the English conflict-of-laws treatise, 2 Dicey et al., The Conflict of Laws (14th ed. 2006) [hereinafter Dicey 14th Edition]; see Estate of Charania, 608 F.3d at 72 n.4, 73–75. The 15th edition now cites the First Circuit decision in support of the stance it takes that English law follows strict immutability. 2 Dicey et al., The Conflict of Laws 1482 n.131 (15th ed. 2012) [hereinafter Dicey 15th Edition]; see also infra note 98 and accompanying text.

  17. [17]. Estate of Charania, 608 F.3d at 69; Uganda Independence Act, 1962, 10 & 11 Eliz. 2, ch. 57.

  18. [18]. Estate of Charania, 608 F.3d at 69.

  19. [19]. Id.

  20. [20]. Id.

  21. [21]. Id.

  22. [22]. See id.

  23. [23]. See id. None of the available briefs, nor the court’s opinion, mention any presence, at any time, of either spouse in England or any other part of the United Kingdom.

  24. [24]. See id.

  25. [25]. Id.

  26. [26]. Id.

  27. [27]. Id.

  28. [28]. Id.

  29. [29]. Id.

  30. [30]. Estate of Charania v. Comm’r, 133 T.C. 122, 124 (2009) aff’d in part, rev’d in part,
    603 F.3d 67 (1st Cir. 2010).

  31. [31]. I.R.C. § 2104(a) (2012). If, indeed, a situs rule justifies federal estate taxation of stock of a nonresident, not a citizen of the United States, then might it be argued that the same situs rule should be applied when determining marital property rights, even though stock is movable property? This would unify the tax and marital-property rights treatment. On the other hand, it would result in spouses having potentially different property rights, depending whether or not marital property rights were being determined for purposes of the federal estate tax with respect to the same asset. As previously discussed, when real property is involved, the situs law determines marital-property rights rather than the law of the marital domicile for both tax and non-tax matters. See supra text accompanying note 7.

  32. [32]. I.R.C. §§ 2033, 2101(a); Estate of Charania, 608 F.3d at 71. While a substantial credit against estate tax was, and continues to be, afforded citizens and residents, see I.R.C. § 2010(c), the same is not true for nonresidents, not citizens of the United States, who are granted a strikingly smaller credit. See I.R.C. § 2102(b)(1). Moreover, while the estate of a nonresident, not a citizen of the United States may claim a marital deduction for property passing to a surviving spouse who is a citizen, no such deduction is available if the surviving spouse, as in the case of Ms. Dhanani, is not a citizen of the United States. See I.R.C. § 2056(d).

  33. [33]. Estate of Charania, 608 F.3d at 71.

  34. [34]. Id. at 70.

  35. [35]. Id. at 71–72.

  36. [36]. Id. at 75.

  37. [37]. Id. at 70.

  38. [38]. Id. at 71.

  39. [39]. Id. at 71–72.

  40. [40]. Id. at 72.

  41. [41]. Id.

  42. [42]. See id. at 70, 72.

  43. [43]. Id. at 76.

  44. [44]. See, e.g., id. at 72 (accepting, without further inquiry, the parties’ stipulation that Uganda’s marital-property regime at the time of marriage was equivalent to that of England’s marital-property regime); see also infra note 50 and accompanying text.

  45. [45]. De Nicols v. Curlier [1900] A.C. 21 (HL) (appeal taken from Eng.).

  46. [46]. Estate of Charania, 608 F.3d at 74; see also infra text accompanying notes 96–102 (discussing the doctrine of renvoi).

  47. [47]. Estate of Charania, 608 F.3d at 77. Certain other aspects of the 2009 Tax Court opinion, Estate of Charania v. Comm’r, involving issues other than the conflict of laws and marital property, were reversed by the First Circuit. Id. at 76–77.

  48. [48]. Id. at 71 n.2 (citing Borden v. Paul Revere Life Ins. Co., 935 F.2d 370, 375 (1st Cir. 1991)).

  49. [49]. Id. at 71.

  50. [50]. A court is not bound by the parties with respect to a question of law. See 83 C.J.S. Stipulations § 28 (2017) (collecting cases). With respect to tax matters, courts have held that they are not bound by a stipulation by the parties as to conclusions of law. See, e.g., Saviano v. Comm’r, 765 F.2d 643, 645 (7th Cir. 1985) (“[W]hile the parties are free to stipulate to the factual elements of the transactions, the court is not bound by the legal conclusions implied by the terminology utilized.”). More generally, the Federal Rules of Civil Procedure provide that “[i]n determining foreign law, the court may consider any relevant material or source, ... whether or not submitted by a party.” Fed. R. Civ. P. 44.1. The accompanying Advisory Committee Notes state: “There is no requirement that the court give formal notice to the parties of its intention to engage in its own research on an issue of foreign law which has been raised by them ... .” Id. advisory committee’s note. Furthermore, “the court’s determination of an issue of foreign law is to be treated as a ruling on a question of ‘law,’ not ‘fact,’ so that appellate review will not be narrowly confined by the ‘clearly erroneous’ standard of Rule 52(a).” Id.

  51. [51]. See generally Albert A. Ehrenzweig, The Lex Fori—Basic Rule in the Conflict of Laws, 58 Mich. L. Rev. 637 (1960) (describing how various courts should select choice of law).

  52. [52]. See Restatement (Second) of Conflict of Laws § 6 (Am. Law Inst. 1971). The Restatement (Second) approach was foreshadowed in an article by its Reporter. See Willis L.M. Reese, Conflict of Laws and the Restatement Second, 28 Law & Contemp. Probs. 679, 692–99 (1963). However, the significant-relationship approach in practice often leads to the application of black-letter rules in various areas of law. In the context of marital-property rights, partial mutability is applied except in rare circumstances. See Restatement (Second) of Conflict of Laws § 258 cmt. a. As such, the Restatement (Second) approach is in accord with overwhelming American case law for nearly two hundred years. See supra text accompanying note 4.

  53. [53]. See, e.g., Brainerd Currie, Selected Essays on the Conflict of Laws 192–93 (1963).

  54. [54]. This is the position of the Restatement (First) of Conflict of Laws as further developed by its Reporter, Professor Joseph Beale in A Treatise on the Conflict of Laws. Restatement (First) of Conflict of Laws § 77 (Am. Law Inst. 1934); 1 Joseph H. Beale, A Treatise on the Conflict of Laws § 7.1, at 55 (1935). With regard to marital-property rights, the Restatement (First) § 290 categorically applies the partial-mutability approach. Restatement (First) of Conflict of Laws § 290.

  55. [55]. See, e.g., Arthur Taylor von Mehren, Special Substantive Rules for Multistate Problems: Their Role and Significance in Contemporary Choice of Law Methodology, 88 Harv. L. Rev. 347, 367–69 (1974) (discussing Neumeier v. Kuehner, 286 N.E.2d 454 (N.Y. 1972)).

  56. [56]. For a summary of several of these approaches, see generally Symeon C. Symeonides, American Choice of Law at the Dawn of the 21st Century, 37 Willamette L. Rev. 1 (2001).

  57. [57]. See supra notes 5–11 and accompanying text.

  58. [58]. The black-letter partial-mutability approach is emphasized explicitly by the Restatement (First) of Conflict of Laws § 290 and in actual application by the Restatement (Second) of Conflict of Laws § 259. This partial-mutability rule had previously been endorsed by Joseph Story. See Restatement (Second) of Conflict of Laws § 259 (Am. Law Inst. 1971); Restatement (First) of Conflict of Laws § 290; supra note 4 and accompanying text.

  59. [59]. See Estate of Charania v. Shulman, 608 F.3d 67, 71–72 (1st Cir. 2010).

  60. [60]. See id. at 72 (“In the United States, courts have tended to favor the doctrine of mutability.”).

  61. [61]. Id. at 71.

  62. [62]. See id. at 71.

  63. [63]. Id.

  64. [64]. See, e.g., Howard v. Reynolds, 283 N.E.2d 629, 630 (Ohio 1972) (illustrating that Ohio uses the decedent’s domicile at death as the controlling law); Restatement (Second) of Conflict of Laws § 260 (Am. Law Inst. 1971); Schoenblum, supra note 2.

  65. [65]. See Schoenblum, supra note 2, § 11.02.

  66. [66]. See id.

  67. [67]. Estate of Charania, 608 F.3d at 71. The First Circuit appears to be taking the position that it must decide marital property rights by sitting as a court at the domicile of the individual decedent at death. No authority, however, is cited in support of this proposition. See generally id.

  68. [68]. See id.

  69. [69]. Id.

  70. [70]. Id.

  71. [71]. Id.

  72. [72]. Id. at 71 n.3.

  73. [73]. See Antony Allott, Essays in African Law 3–4 (1960).

  74. [74]. The problems arising from a reference to the national law of a country where the law is made at the level of political subdivisions is analyzed in considerably greater depth in John Delatre Falconbridge, Essays on the Conflict of Laws 202–16 (2d ed. 1954); see also 1 Dicey 15th Edition, supra note 16, at 93 (“If, on the other hand, the English court allows the foreign expert witness to assume that the national law of a British citizen is English law, as has been done in other cases, it is basing its decision on a manifestly false premise.”).

  75. [75]. Pippa Rogerson, Collier’s Conflict of Laws 4 (4th ed. 2013).

  76. [76]. The principle is reflected, for example, in the judicial decision reported in Antwerpen, Oct. 26, 1999, Revue Général de Droit Civil Belge-Tijdschrift voor Belgisch Burgerlijk Recht [TBBR-RGDC] 2001, 33 (Belg.); see also Alain Verbeke, Belgisch Erfrecht in Kort Bestek: Met IPR-Aspecten en Praktische Tips Voor Nederbelgen [Basic Principles of Belgian Marital Property and Inheritance Law: With Conflict of Laws Rules and Belgo-Dutch Cross Boarder Issues] 75 (2003) (unpublished manuscript), Following the enactment of the Belgian Code of Private International Law in 2004, Article 51 changed the rule to the habitual residence of the spouses. The Belgian nationality rule was longstanding. See, e.g., 1 Francis Wharton, A Treatise on the Conflict of Laws or Private International Law 435 (George H. Parmele ed., 3d ed. 1905) (noting how “several European states” emphasized that some principles of law are national or local and should not “be enforced on subjects who are temporarily residing within the bounds of a state”).

  77. [77]. See Schoenblum, supra note 2, § 9.09.

  78. [78]. Code de Droit International Privé [hereinafter CDIP] [Code of Private International Law] (Belg.).

  79. [79]. See id. art. 17, § 2.

  80. [80]. Id.

  81. [81]. See Gonzalo Parra-Aranguren, General Provisions on the Conflict of Laws in the Belgian Code of Private International Law (2004), in Liber Memorialis Petar Šarčević: Universalism, Tradition, and the Individual 111, 120–26 (J. Erauw et al. eds., 2006).

  82. [82]. In fact, the 2004 PIL does largely abandon nationality in favor of habitual residence, with only a few exceptions. See, e.g., CDIP art. 51 (relating to the matrimonial-property regime).

  83. [83]. Falconbridge, supra note 74, at 208.

  84. [84]. CDIP art. 17. Admittedly, in other cases, it would not automatically default to the forum’s own law as Falconbridge urged. Falconbridge, supra note 74, at 208.

  85. [85]. But see Chris Clarkson et al., Commission Européenne, JAI/A3/2001/03, Study on Matrimonial Property Regimes and the Property of Unmarried Couples in Private International Law and Internal Law 5–8 (2003) (UK), http://www.pedz.uni-mannheim.
    de/daten/edz-k/gdj/03/england_report_en.pdf. (“[W]hile the act of marriage has no effect upon the ownership of property [in the United Kingdom] and the rules concerning ownership (legal and equitable) of the family home are the same irrespective of whether the parties are married or not, there are nevertheless numerous statutory provisions dealing with the property relations between married persons ... .”).

  86. [86]. See supra text accompanying note 45.

  87. [87]. See infra text accompanying notes 125–29.

  88. [88]. See Marsh, supra note 4, at 108 (discussing Lashley v. Hog (1804) 47 Eng. Rep. 1243, 1243); see infra text accompanying note 127–32 (discussing the House of Lords decision on appeal from the Scottish Court of Session in Lashley, 47 Eng. Rep. at 1243).

  89. [89]. Estate of Charania v. Shulman, 608 F.3d 67, 71 (1st Cir. 2010) (emphasis added).

  90. [90]. What if the country changes its property law over time? This question is suggested by the First Circuit but not addressed. The issue might also have affected the outcome in Charania, but is not considered at all by the First Circuit. Id. at 73.

  91. [91]. Id. at 72.

  92. [92]. See infra text accompanying note 104.

  93. [93]. See Falconbridge, supra note 74, at 172.

  94. [94]. This is also known as “the ping-pong theory of renvoi.” See id. at 187.

  95. [95]. See Restatement (Second) of Conflict of Laws § 258(1) (Am. Law Inst. 1971) (“The interest of a spouse in a movable acquired by the other spouse during the marriage is determined by the local law of the state which, with respect to the particular issue, has the most significant relationship to the spouses and the movable under the principles stated in § 6.” (emphasis added)). Comment (b) to the section amplifies upon the foregoing rule: “Except in rare circumstances, this state will be the state where the spouses were domiciled at the time the movable was acquired... . [T]he local law of the state where the spouses were domiciled at the time the movable was acquired will usually be applied to determine marital property interests therein in the absence of an effective choice of law by the parties.” Id. § 258 cmt. b.

  96. [96]. The 2004 PIL eliminates renvoi, subject to a limited number of exceptions. CDIP art. 16.

  97. [97]. In this regard, see one of several classic critiques of renvoi by Ernest G. Lorenzen, The Qualification, Classification, or Characterization Problem in the Conflict of Laws, 50 Yale L.J. 743 (1941), wherein Professor Lorenzen states:

    Personally, I cannot approve a doctrine which is workable only if the other country rejects it. Apart from that, I do not favor handing over our Conflicts problems to so-called experts on foreign private international law. It is difficult enough to get accurate expert testimony with respect to foreign municipal law, but such testimony is much more unreliable with respect to foreign Conflict of Laws. For these reasons I should still regard the general acceptance of the renvoi doctrine in our law as most unfortunate.

    Id. at 753–54.

  98. [98]. Significantly, Dicey’s 15th edition of The Conflict of Laws recommends against application of renvoi. Dicey 15th Edition, supra note 16, at 95. This is crucial in that the First Circuit in Charania relies on Dicey’s reading of De Nicols as authority for the foreign-court theory, whereby an English court should sit as a foreign court under the choice of law and apply its whole law. Estate of Charania v. Shulman, 608 F.3d 67, 73–74 (1st Cir. 2010). In other words, Dicey is credited by the court as authoritative when it supports the court’s position, but when it would yield a different outcome on account of the rejection of renvoi, it is ignored by the court.

  99. [99]. See Schoenblum, supra note 2, § 9.10.

  100. [100]. See Erwin N. Griswold, Renvoi Revisited, 51 Harv. L. Rev. 1165, 1169–70 (1938).

  101. [101]. See Marsh, supra note 4, at 112–13.

  102. [102]. See Griswold, supra note 100, at 1177.

  103. [103]. See, e.g., Falconbridge, supra note 74, at 183.

  104. [104]. Estate of Charania v. Shulman, 608 F.3d 67, 72 (1st Cir. 2010).

  105. [105]. See Allott, supra note 73, at 3–51. Technically, Uganda did preserve the prior common law at the time of its independence via the Judicature Act 1966, § 14. However, what that common law was is quite uncertain, as even prior to independence the question was fraught with difficulty. While English common law was generally received in Uganda, the reception was not wholesale and involved many twists and turns as the courts responded to local circumstances and needs. See id.

  106. [106]. East Africa Order in Council, 1902, Stat. R. & O. No. 2, as amended by the East Africa Order in Council of 1911, art. 15.

  107. [107]. See Clarkson et al., supra note 85, at 5.

  108. [108]. East Africa Order in Council, 1902, Stat. R. & O. No. 2, as amended by the East Africa Order in Council of 1911, art. 15.

  109. [109]. J.N.D. Anderson, Colonial Law in Tropical Africa: The Conflict Between English, Islamic and Customary Law, 35 Ind. L.J. 433, 437 (1960).

  110. [110]. See J. Duncan M. Derrett, East Africa: Recent Legislation for Hindus, 11 Am. J. Comp. L. 396, 397 (1962). Unlike Muslims, Hindus were not exempted from the Uganda Succession Ordinance, 1906 (Cap. 34), s. 50. See Derrett, supra, at 403 n.49. However, there is a Ugandan statute indicating distinctive treatment for Hindus in the context of marriage, but not necessarily with respect to marital-property rights. See infra text accompanying notes 113–15.

  111. [111]. The vast majority of Indians, both Hindu and Muslim, in Uganda were from Gujarat. See Palash Ghosh, Uganda: The Legacy of Idi Amin’s Expulsion of Asians in 1972, Int’l Bus. Times (Mar. 13, 2012, 3:05 PM),

  112. [112]. See Derrett, supra note 110, at 397–98; There Is Need to Review the Law on Marriage and Divorce, New Vision (Sept. 1, 2011, 3:00 AM),

  113. [113]. This failure is reflected in unsuccessful efforts, generally, to clarify the marital-property rules. See Anthony Luyirika Kafumbe, Women’s Rights to Property in Marriage, Divorce, and Widowhood in Uganda: The Problematic Aspects, 11 Hum. Rts. Rev. 199, 199–201 (2010).

  114. [114]. Informally, separate property ownership is probably widespread. This is primarily due to the weaker position of women in Uganda and their difficulty in asserting rights that they may otherwise have. See id. at 203–04.

  115. [115]. Julius Rwabinumi v. Hope Bahimbisomwe, (2013), Civ. App. No. 10 (2009), U.G.S.C. 5 (Uganda),

  116. [116]. Julius Rwabinumi v. Hope Bahimbisomwe, (2008), Civ. App. No. 30 (2007),
    U.G.C.A. 19 (Uganda),

  117. [117]. Estate of Charania v. Shulman, 608 F.3d 67, 69 (1st Cir. 2010).

  118. [118]. Rwabinumi, (2013) U.G.S.C. 5.

  119. [119]. Id. ¶108.

  120. [120]. Id. ¶115.

  121. [121]. De Nicols v. Curlier [1900] A.C. 21 (HL) (appeal taken from Eng.).

  122. [122]. Id.

  123. [123]. See supra text accompanying note 106.

  124. [124]. Lashley v. Hog (1804) 47 Eng. Rep. 1243, 1243.

  125. [125]. Id. at 1263–65.

  126. [126]. Id. at 1261; see also Darrell E. Burns, Jr., De Nichols [sic] v. Curlier: Revisited, 14 Osgoode Hall L.J. 797, 799 (1976) (“Lord Eldon stated that the spouses’ matrimonial property rights were mutable with changes in their domicile.”).

  127. [127]. Marsh, supra note 4, at 107 (“Obviously, they are prima facie in conflict.”).

  128. [128]. See id. at 107–08.

  129. [129]. Burns, supra note 126, at 801.

  130. [130]. Id. The characterization of Scottish marital-property rights at the time has been severely criticized. See id. at 803–04.

  131. [131]. Estate of Charania v. Shulman, 608 F.3d 67, 73 n.6 (1st Cir. 2010). The First Circuit cites to Dicey 14th Edition, supra note 16, at 1299, for the proposition that there is no English authority for the mutability theory. See Estate of Charania, 608 F.3d at 73–74. In fact, in the most recent edition of the treatise, the authors state: “The exact limits of the principle established in De Nicols v. Curlier have never been settled.” 2 Dicey 15th Edition, supra note 16, at 1482.

  132. [132]. Estate of Charania, 608 F.3d at 73–74.

  133. [133]. Notes on Decided Cases, Effect of Change of Matrimonial Domicile on Community of Goods or Marriage-Contract, 12 Jurid. Rev. 213, 213–14 (1900).

  134. [134]. Id. at 214. Nevertheless, the First Circuit claims that “the text of De Nicols belies this reading,” failing to note the several Lords’ opinions at the time endorsing this very reading. Estate of Charania, 608 F.3d at 74. Compare Dicey & Morris, On the Conflict of Laws 1068 (11th ed. 1987) (offering a far narrower reading of the significance of the De Nicols decision and recognizing the differences among the Lords), with 2 Dicey 14th Edition, supra note 16, at 1299. It is noteworthy that, contrary to the First Circuit reading, a leading American treatise on conflict of laws, Eugene F. Scoles et al., Conflict of Laws 611 (4th ed. 2004), endorses the narrow reading of De Nicols as turning on an implied contract under French law.

  135. [135]. See Estate of Charania, 608 F.3d at 73–74.

  136. [136]. Id.

  137. [137]. Id.

  138. [138]. See id. at 69, 72.

  139. [139]. See id. at 69.

  140. [140]. See John Westlake, A Treatise on Private International Law, with Principal Reference to its Practice in England 79–80 (5th ed. 1912); see also Falconbridge, supra note 74, at 106–07 n.(h) (listing authorities supporting and contradicting the idea that De Nicols was about succession).

  141. [141]. G.C. Cheshire, Private International Law 494–98 (2d ed. 1938).

  142. [142]. G.C. Cheshire, Private International Law 499–502 (4th ed. 1952); see Marsh, supra note 4, at 107–08.

  143. [143]. Martin Wolff, Private International Law 365­–69 (1945). It is not clear whether Wolff would apply the “foreign court” theory as well in conjunction with the total-immutability approach, which could alter the outcome.

  144. [144]. Id.

  145. [145]. See Allott, supra note 73, at 32–33.

  146. [146]. Estate of Charania v. Shulman, 608 F.3d 67, 69 (1st Cir. 2010).

  147. [147]. Id.

  148. [148]. Id.

  149. [149]. See Ghosh, supra note 111.

  150. [150]. Estate of Charania, 608 F.3d at 69.

  151. [151]. Id. at 77.

  152. [152]. That is, the difference between voluntarily departure and involuntary exile.

  153. [153]. Estate of Charania, 608 F.3d at 74.

  154. [154]. See supra note 131.

  155. [155]. See Dicey 15th Edition, supra note 16, at 156–57.

  156. [156]. See May v. May [1943] 2 All ER 146, 148–49 (Eng.).

  157. [157]. Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938).

  158. [158]. Id. at 78.

  159. [159]. Id.

  160. [160]. Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496 (1941).

  161. [161]. Id.

  162. [162]. The Klaxon court did not actually address whether conflict-of-laws rules are part of the substantive law of a state. See id. at 487. This is actually a complex issue with many ramifications.

  163. [163]. Erie, 304 U.S. at 78.

  164. [164]. 28 U.S.C. § 1346(a)(1) (2012).

  165. [165]. I.R.C. § 6213(a) (2012).

  166. [166]. See Morgan v. Comm’r, 309 U.S. 78, 80 (1940) (“State law creates legal interests and rights. The federal revenue acts designate what interests or rights, so created, shall be taxed.”).

  167. [167]. See In re Lindsay, 59 F.3d 942, 948 (9th Cir. 1995) (“In federal question cases with exclusive jurisdiction in federal court, such as bankruptcy, the court should apply federal, not forum state, choice of law rules.”).

  168. [168]. Id.; see also In re E. Livestock Co., 547 B.R. 277, 282–85 (Bankr. S.D. Ind. 2016) (noting federal interests as a variable in choice-of-law determinations). See generally Kimberly J. Winbush, Annotation, Determination Whether Bankruptcy Court Should Apply Choice-of-Law Rules of Forum State in Which It Sits or Choice-of-Law Rules of Federal Common Law, 21 A.L.R. Fed. 3d Art. 2 (2017) (discussing decisions that have identified federal interests when making choice-of-law determinations).

  169. [169]. See In re E. Livestock Co., 547 B.R. at 282–83, 285.

  170. [170]. This principle is commonly described as horizontal equity in the income tax context. See, e.g., Charles E. McLure, Jr., Integration of the Personal and Corporate Income Taxes: The Missing Element in Recent Tax Reform Proposals, 88 Harv. L. Rev. 532, 535 (1975). This does not mean all uniformity can be achieved. See Randolph E. Paul, Selected Studies in Federal Taxation 5 (2d series 1938) (“We shall get nowhere rapidly with the problem of simplification until we recognize that what we bravely call uniformity on a national scale is a myth.”).

  171. [171]. In re Gaston & Snow, 243 F.3d 599, 606 (2d Cir. 2001); see also In re Merritt Dredging Co., 839 F.2d 203, 206 (4th Cir. 1988) (stating that the Klaxon rule should be applied when a bankruptcy court is seeking to determine the property interest of the debtor).

  172. [172]. In re Gaston & Snow, 243 F.3d at 606 (citations omitted).

  173. [173]. See supra text accompanying notes 165–68.

  174. [174]. Although headquartered in the District of Columbia, the judges of the Tax Court travel, hearing cases throughout the country. Arguably, then, the Tax Court might apply the choice-of-law rule of the state in which it is sitting on a particular tax case. Indeed, this was the approach in Sartori v. Comm’r, 66 T.C. 680, 689 (1976) (relying on Pennsylvania choice-of-law rule because court was trying case in Pittsburgh).

  175. [175]. The Restatement (Second) is not the only alternative for a unitary rule. See, e.g., Edmond N. Cahn, Local Law in Federal Taxation, 52 Yale L.J. 799, 820–23 (1943) (predicting the development of unique conflict-of-laws rules emphasizing economic situs, although built on otherwise applicable conflict-of-laws principles).

  176. [176]. See cases cited in Restatement (Second) of Conflict of Laws § 258 (Am. Law Inst. 1971) and Restatement (First) of Conflict of Laws § 290 (Am. Law Inst. 1934).

  177. [177]. Arguably, the total-mutability approach, by contrast, would violate the due process clause as “a disturbance of a vested right.” In re Estate of Thornton, 33 P.2d 1, 3 (Cal. 1934). The adoption of a conflict-of-laws rule does not violate the due process clause if not arbitrary or fundamentally unfair, that is, the state having its law apply “must have a significant contact or significant aggregation of contacts.” Allstate Ins. Co. v. Hague, 449 U.S. 302, 313 (1981). Presumably, a noncitizen, not resident in the United States would have no such rights. Cf. Kerry v. Din, 135 S. Ct. 2128, 2131 (2015) (finding that it was not a violation of due process rights of wife legally in United States to deny visa to nonresident, noncitizen husband without explanation because he has no right to immigrate).

  178. [178]. Estate of Charania v. Shulman, 608 F.3d 67, 74 (1st Cir. 2010) (citing A.W. Chesterton Co. v. Chesterton, 128 F.3d 1, 7 (1st Cir. 1997); Kassel v. Gannett Co., 875 F.2d 935, 950 (1st Cir. 1989)).

  179. [179]. See A.W. Chesterton Co., 128 F.3d at 7; Kassel, 875 F.2d at 949–50.

  180. [180]. A.W. Chesterton Co., 128 F.3d at 5; Kassel, 875 F.2d at 941.

  181. [181]. Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938). Likewise, the federal courts are bound by the law declared by the state legislature. Id.

  182. [182]. Id.

  183. [183]. Comm’r v. Estate of Bosch, 387 U.S. 456, 463–66 (1967).

  184. [184]. Id. at 457 (internal quotation marks omitted).

  185. [185]. Id.

  186. [186]. Id. at 465 (alterations in original) (quoting King v. Order of United Commercial Travelers of Am., 333 U.S. 153, 160–61 (1948)).

  187. [187]. Id.

  188. [188]. Paul L. Caron, The Federal Courts of Appeals’ Use of State Court Decisions in Tax Cases: “Proper Regard” Means “No Regard,46 Okla. L. Rev. 443, 486 (1993).

  189. [189]. Id. at 486–87.

  190. [190]. Estate of Charania v. Shulman, 608 F.3d 67, 75 (1st Cir. 2010).

  191. [191]. De Nicols v. Curlier [1900] A.C. 21 (HL) (appeal taken from Eng.). With regard to the sharply differing considerations when U.S. domestic versus international conflict of laws is involved, see Douglas Laycock, Equal Citizens of Equal and Territorial States: The Constitutional Foundations of Choice of Law, 92 Colum. L. Rev. 249, 259 (1992) (“It is a serious mistake to discuss domestic and international choice-of-law cases interchangeably, even though that practice is nearly universal in the conflicts literature.”).

  192. [192]. Estate of Charania, 608 F.3d at 71.

  193. [193]. See Mark Van Hoecke, Legal Culture and Legal Transplants, in Law, Society, and Community: Socio-Legal Essays in Honour of Roger Cotterrell 273, 278–80 (Richard Nobles & David Schiff eds., 2014); Belgium, European E-Justice, https://e-justice.europa.
    eu/ (last updated July 18, 2017).

  194. [194]. Van Hoecke, supra note 193, at 279–80.

  195. [195]. See id. at 279.

  196. [196]. See supra text accompanying notes 133–45.

  197. [197]. I.R.S. Tech. Adv. Mem. 83-39-004 (June 14, 1983) (emphasis in original).

  198. [198]. Rev. Rul. 69-285, 1969-1 C.B. 222 (emphasis added) (citing Comm’r v. Estate of Bosch, 387 U.S. 456 (1967)).

  199. [199]. Estate of Bosch, 387 U.S. at 456–57.

  200. [200]. I.R.S. Gen. Couns. Mem. 39,183 (Mar. 6, 1984).

  201. [201]. Id. at *7; see also I.R.S. Gen. Couns. Mem. 33,934 (Sept. 27, 1968), withdrawn, I.R.S. Gen. Couns. Mem. 35,073 (Oct. 10, 1972).

  202. [202]. Estate of Charania v. Shulman, 608 F.3d 67, 73 (1st Cir. 2010).

  203. [203]. 19 Charles Alan Wright et al., Federal Practice and Procedure § 4507 (3d ed.), Westlaw (database updated Apr. 2017) (quoting Richardson v. Comm’r, 126 F.2d 562, 567
    (2d Cir. 1942)).

  204. [204]. Bernhardt v. Polygraphic Co. of Am., 350 U.S. 198, 205 (1956).

  205. [205]. Wright et al., supra note 203, § 4507.

  206. [206]. Estate of Charania, 608 F.3d at 75.


Centennial Professor of Law, Vanderbilt University Law School.