This project description focuses on a specific occurrence of normative conflicts. It addresses the need of deciding the applicable law when conflicting pieces of legislation coming from different legal systems have to be merged. I will check whether and how the argumentation method could help to deal with these cases of normative interactions. Actually, from the logical point of view, the situation described poses some challenging issues, first of all that of taking into consideration contextual reasoning. Furthermore, the task of merging of normative provisions from different legal systems is itself far from being just an automatic activity: merging is not obvious and when it is required in a concrete case, the point is that in principle the systematic character of law cannot allow the legal operator to mechanically combine X from system x and Y from system y. The argumentation method can show its efficacy when reasoning in such situations is often characterized by interpretive uncertainty. The methodology I will follow consists of two main steps: firstly, I outline the legal case study drawn from international taxation law, i.e. juridical double imposition; then, I propose some argument schemes that can exemplify the legal reasoning and the inference chain that lead the legal operator when facing such situations.
This paper aims to shortly address a specific legal issue among those that in many
subtle ways characterize the fast-growing phenomenon of interactions among
different legal systems. The law, far from just being an ordered, hierarchical, formalized
and isolated system [
If this is the broad legal framework, even though necessarily sketched in a general
fashion, what I would like to consider in the next few paragraphs is the specific
situation when conflicting pieces of legislation coming from different legal systems need
to be merged in order to decide the applicable law. As I will explain further down, I
take as a case study the hypothesis of juridical double taxation that can occur in
international tax law. The perspective I assume is that of argumentation methods applied
to the legal argumentation required in such cases [
Governments make use of several criteria to delimit the range of their tax jurisdiction and of their income taxation. Two of the most implemented principles are that of world-wide taxation for residents and that of territoriality for nonresidents holding some income in the territory of the state. Then, governments usually provide their citizens for offsets as regards extraterritorial income. Analogous solutions are adopted for companies. Nevertheless, exactly the fact that often the States opt for competing criteria highly increases the risk of conflicts among laws and especially of international double imposition, i.e. where the same income is taxed by two countries in the same period and by means of the same tax.
Juridical double taxation is currently a matter of high interest, as it is confirmed by the attention periodically paid to the issue by the European Commission (EC). The EC actually considers it a fundamental part of its strategy of addressing the crossborder tax problems within the internal market. Its interventions are intended to limit situations of conflicts between the European Treaties and the bilateral double taxation treaties that Member States have concluded with each other and with third countries.1
Theoretically, the legislator of a Member State can proceed in many ways in order to draft its tax law provisions in cases characterized by international elements. The legislator, in the full exercise of its own sovereignty, can consider one’s own provisions exclusively, fueling the high risk of double taxation (as a matter of fact, this is just a theoretical hypothesis, since the ever-growing international relationships force the State to come to an agreement with other States for the benefit of their citizens and companies). The legislator can provide for unilateral measures to apply in such cases (e.g. possibility for the taxpayers to deduce what they have already paid abroad). 1 http://ec.europa.eu/taxation_customs/taxation/company_tax/double_taxation_conventions/index_en.htm (accessed on 13/09/2014). The legislator can sign bilateral double taxation treaties (often, shaped on the OECD model tax convention), that usually provide for preferences among different criteria of international taxation in name of the prohibition to tax twice the same income.
Diversified actions are required because “no uniform or harmonization measure designed to eliminate double taxation has as yet been adopted at Community law level”, as the European Court of Justice (ECJ) openly recognized not long ago.2 This is also true, more generally speaking, at the international level. The main reason is the (political and normative) sovereignty each State is still capable to exercise in the fiscal area as regards its own territory.
However, the ECJ has changed its opinion on this matter over time. It has progressively taken into consideration that, as a European judicial body and in the light of the current state of EU law, it has no such a power to impose its own fiscal regulation on the Member States. So, from the tendency to recognize the existence of a prohibition of double imposition (appealing to the general principle of nondiscrimination)3 it has ended up accepting juridical double imposition as a still unavoidable restriction to the fundamental liberties (that thwarts the full success of the internal market).4 Nevertheless, a recent opinion5 seems to bring us back to the first rulings, since the Court has resolved the case referring to the principle of free movement of capitals and the impossibility to limit that fundamental liberty through double imposition. 3
This being the juridical context in the EU, what makes the double imposition issue interesting and challenging from the perspective of legal reasoning is the necessity to deal with the systematic and closed character of two legal systems whose norms are conflicting.
The task of merging of normative provisions from different legal systems is far from being just an automatic activity. In principle the systematic character of law cannot allow the legal operator first to mechanically recognize other State’s legislation and then to combine X from system x and Y from system y, most of all if in system X some term tx is systematically defined in other provisions that are not directly involved in the merging operation.
When it comes to solve conflicts of laws within just one legal system, several
approaches, dating back to mid-Nineties, have turned out useful [
All the methods meant to work in just one system are somehow defective once they are shifted to conflicts occurring across states: they do not give adequate tools to comprehend in a unique vision two different styles of interpretation. Indeed, an interpretive principle that can be decisive in one system according to its own legislation may not be accepted in the other.
So, in cases of conflicts of norms provoking double imposition cases, I can identify some characteristics that prevent the use of above mentioned approaches: it is necessary to take into account at the same time pieces of legislation coming from different legal systems: each legal system is basically a closed system and the recognition of the foreign norm is not automatic; those provisions could be conflicting with one another and the conflict can be genuine or merely apparent, as I will show below; the norms need to pass through the interpretation procedure: they need to be interpreted both in the light of international agreements signed by the parties, if existing, and in a systematic way, that in turn may reveal conflicting interpretation of a same term; this means that a synthesis of the two systems is somehow required; contextual information plays an essential role: only when aware of many factors featuring the concrete situation, it is possible to evaluate the strength of opposite arguments; all this contributes to bring about uncertainty and explicit or implicit incoherence.
Even though it has been not yet fully explored, the topic is not new to AI and law
scholars. Besnard et al. [
Starting from a similar case study, I will adopt a different perspective. As I will show in the next paragraphs, I will apply the argumentation method, in particular considering the possibility of merging different systems and interpretations through meta argumentation schemes in order to find a mechanism able to better deal with the just sketched problems. 6 http://www.impots.gouv.fr/portal/dgi/public/documentation.impot?pageId=docu_international&espId=1&sfid=440 (accessed on 16/09/2014).
The first step when dealing with normative conflicts is to determine whether they are genuine or simply apparent. 4.1
Let us consider the following case of juridical double imposition7 before the Italian Supreme Court. In the later years, C., resident in Italy, has been receiving from Luxembourg a retirement pension for working activity as an employee in the private sector there. According to the Luxembourg legislation, C. has to pay taxes on those pensions to the local social security body. In addition to this imposition, regularly paid, for four years C. has been also subject to the Italian personal income tax (IRPEF) for the same income item. Therefore C. decides to resort to the taxation judicial authority in Italy for the reimbursement of the amounts that he assumes to have wrongfully paid to the Italian Tax Agency (Agenzia delle Entrate). He bases his lawsuit on the Bilateral Convention existing between Italy and Luxembourg.8
At the closure of the appeal, the judicial Commission (Commissione Regionale Tributaria dell’Umbria) recognizes that C. is entitled to obtain the full restitution of the amounts paid to the Italian government until then. Actually, according to the interpretation given by the judges, those payments would not be justified in the light of the Bilateral Convention, since pensions would be included in art. 18 par. 2 that provides for as follows: “Notwithstanding the provisions of paragraph 1, pensions and other social security allowances paid by a Contracting State under social security law may be taxed in that State.” Therefore, Luxembourg would be the only State entitled to impose the taxation on C.’s pensions.
The Italian Tax Agency files an appeal to the Supreme Court, mainly asserting a wrong interpretation of the convention: the pensions would be encompassed in the par. 1 of the same provision that provides for the taxation in the State of residence. In the alternative, however, also applying par. 2, the pension should be taxed in both States and the State of residence should then apply the mechanism of deduction of what the taxpayer has already paid.
The Supreme Court substantially agrees with the judges of the appeal that the pensions are included in par. 2, though, unlike the previous opinion, the Court holds that also the State of residence is entitled to exercise its fiscal power, since in the conventional provision there is no explicit reference to the adverb “only”, as the one in the first paragraph. So, the income item may actually be taxed twice, without any violation of the existing Convention and provided that the State of residence makes available to the taxpayers ways to deduce the amount they have paid to the other contracting state (art. 24). On the basis of these remarks, the Supreme Court annuls the ruling and forwards the case to the court of appeal again.
In this opinion, the Supreme Court seems to accept the fact that double taxation may happens in the broad context of internationalization, where two or more sovereign States may have the competing legitimate powers to impose their taxation laws. Still, the analysis of the case shows how the conflict here is just apparent. The Bilateral Convention includes a sort of closing rule, art. 24 (“Provisions for the elimination of double taxation”) and it identifies mechanisms to avoid in such cases double imposition that ends up weighing on the shoulders of the taxpayer. Accordingly, Italy, as the state of residence, should allow the taxpayers to detract what they paid abroad. 4.2
Let us now examine Kerckhaert and Morres case, since precisely on that occasion the ECJ recognized the impossibility to generally avoid juridical double imposition, considering that no uniform law on the matter exists either at the communitarian or international level.
In 1995 and 1996 Mr. and Mrs. Kerckhaert-Morres, Belgian residents, received dividends from a company established in France. In France the gross dividends were made subject to a levy of 15%, deducted at source by way of tax on income. In Belgium, in their tax return, the spouses applied to take advantage of the tax benefit provided for in Article 19. A (1) of the France-Belgium Convention corresponding to the French tax at source. That tax benefit had been withdrawn by the Belgian legislature and therefore their application was rejected. In short, in Belgium the dividends received from a company established abroad are taxed as if they were received by a national company, i.e. at 25% tax rate, without any chance to detract what they have possibly already paid abroad. Belgium definitely applies the world-wide taxation principle to tax its residents and this causes a genuine conflict of norms, even if acceptable in the light of EU law.
The case is paradigmatic under many facets: international elements, exercise in parallel by two Member States of their fiscal sovereignty based on competing principles, no “measures necessary to prevent situations such as that at issue in the main proceedings by applying, in particular, the apportionment criteria followed in international tax practice”. In detail, the country of residence adopts the world-wide taxation principle for its residents, so that all their income has to be subject to the national tax law; whereas the other country considers the territoriality as the leading criterion in order to tax nonresidents holding some income in the territory of the state.
In absence of further specifications in either legislation (e.g. explicit exceptions or
unless clauses [
In the framework and in absence of explicit exceptions (e.g., deduction measure) in either legislation, the conflict proves to be genuine: people are required to pay taxes in country 1, because of their residence there, and also in country 2 because they have been receiving dividends from a company established in the territory of country 2. 5
The European judges have de facto returned the hot issue to the senders: first, to the Member States in charge of finding normative solutions to avoid cases of double imposition, secondly, to the national judicial authorities, often the last resort to preserve decent standards of justice. The task of merging pieces of legislation crossing different legal orders actually challenges the national judges because citizens usually look to them in order to see the application of criteria of fairness and equity in the tax field.
Merging is far from being an automatic activity. Indeed, it is not obvious because the law is systematic in nature. In principle, this peculiar feature of the law does not allow the national judge to mechanically combine the norm X taken from system x and the norm Y taken from system y or to know if in X some term tx is systematically defined in other provisions that are not directly involved in the merging operation.
As regards the possible emerging conflicts, they do not simply end in conflicts between norms, but they can actually pertain to the realm of interpretation. It is not rare that two legal systems deal with the same normative provision though different modes of interpretation. So the judge faces one of the two circumstances: 1) conflicts among two interpretations that, considered together, make the conflict actual; 2) conflicts among two interpretations that, on the contrary, help to dissolve the conflict.
The scenario uncovers a set of problems to tackle: a) two opposite arguments, each relying on a rule coming from a different legal system; b) need to analyze both legislations and their interpretive backgrounds; c) need to see whether there is a way to reason with these arguments (meta-argumentation) and try to solve the conflict through the appeal to a general principle. 6
Going back to Kerckhaert and Morres case, according to Belgium, legal system 1 (S1), the spouses have to pay taxes there, country 1, because of the place of their residence. On the other hand, on the basis of French legal system (S2), they are subject to taxes in France, country 2, because it is where they received the dividends.
Let us try to outline the situation in Table 1.
I could further synthetize as follows: S1 = R1; S2 = R2; R1 is incompatible with R2. They are incompatible because of the undesirability of double taxation in the light of the principles of fairness, ability-to-pay and equity shaping tax law. The conflict between R1 and R2 seems to be unsolvable, since neither rule has the power to defeat the other one. The national judges cannot appeal to lex specialis, lex posterior or lex superior principles because those are effective within the same system.
Things thus standing, the judges can proceed exploring the following possibilities: remaining in their own legal system, they can look for a superior normative provision (e.g., a constitutional norm, or a norm drawn from bilateral conventions, as in the Italian case law above considered) that can solve the conflicts; in absence of superior norms, they can identify the interpretation that better suits the case and ends up untangling the normative mish-mash (e.g., equity principle: the principles of equity, of ability to pay and of progressive contribution to the public expenses are quite common in EU countries’ Constitutions);9 they can consider different ways of interpretation, even coming from different legal systems, and pick the one solving the conflict; considering double taxation as a violation of the principle of non-discrimination, they can resort to the European Court of Human Rights (ECtHR), a superior judicial body that enjoys the so called margin of appreciation when interpreting the national laws in the light of the European Convention of Human Rights and can make use of comparative argumentation.
Interpretation gains an essential role and I think that the general structure for
interpretive arguments as identified in Sartor et al. (2014) [
See for example art. 53 of Italian Constitution, normally read in combination with art. 3 (principle of non-discrimination).
Translated to my case study, E can be R1 stating the world-wide taxation principle for residents and occurring in D, i.e. the tax law of S1. The setting S can refer to the general principle of equity in taxation according to which if an income item, e.g. the dividends received abroad, has already been taxed in another country where it has been produced, it is not subject to the tax law of the country of residence. This setting S exerts relevance on the interpretation I, so that E has to be interpreted accordingly and therefore does not have to be applied. Therefore, following this argumentation scheme (AS from here on) that gives space to the equity principle, the conflict can be solved.
This reasoning, based on meta-argumentation, can be performed by the superior judicial body considering both argumentation schemes as follows: R1 and R2 are incompatible. S1 = AS X, where R1 and R2 are compatible. S2 = AS Y, where R1 and R2 remains incompatible.
These remarks are still introductory and try to identify possible directions of investigation, such as that of meta-argumentation. What I reckon as essential is the role assigned to the national judge who effectively has to find a way out of the juridical deadlock created by the conflicting rules and interpretations. The judges can actually decide to build their own argumentative path starting from two conflicting arguments and assigning strength to each of them according to superior principles or reasons that are substantive to the legal order, this considered in a broad way. But they can also recognize a role to a superior judge entitled to perform some kind of comparative argumentation, i.e. the ECtHR. 7
So far, I have outlined my main research question: which kind of argumentation scheme can help in figuring out a way to compound two or more diverse legal systems and reach a suitable legal solution? Is it conceivable a mechanism able to solve the conflicts occurring also at the interpretive level?
The first phase of my research has been committed to the identification of the substantial research problems. Indeed, I have observed the legal case study of juridical double imposition, where the problem of merging pieces of legislations coming from different legal systems is mostly clear. As a second level of analysis, I have begun to address the theory of legal reasoning regarding the problem considered. I have tried to informally identify the argument schemes in order to later develop a descriptive and prescriptive model of reasoning, exploring the chance to make use of metaargumentation. The argumentation schemes will be then tested at the theoretical level. 8
The present work is still provisional and tries to address the question about how to find a solution to the conflicts existing among legal provisions coming from different legal systems using the tools of argumentation theory and meta-argumentation in particular. In doing so, I have considered a case study drawn from international taxation law trying to figure out how the normative conflict can be translated in a conflict of interpretations among which the national judge or a superior judge can choose to reach an adequate legal solution.
My research methodology starts from the empirical observation of the legal problem followed by the description of a related case study. Afterwards, I apply to it argumentation theory tools in order to test their efficacy in dealing with conflicts of laws and conflicts of interpretations.
I have not yet achieved proper research results. Though, I have defined the boundaries of my research, I have chosen a precise research methodology and tried to figure out how interpretive argumentation schemes can contribute to this research field in an innovative way. Future work will include the formalization of the identified arguments through argumentation logics and the investigation on whether and how nonmonotonic logic, such as default logic, could provide useful insights and solutions to the problem. As a plausible endpoint, I envision the construction of an argument framework.