From Aristotle’s ʻArithmetic Proportion’ to Ménage-à-trois – Anglo-American Justice Theories in the Context of Norwegian Tort Law

1. Introduction¹

Ever since modern justice theorists rose up against the efficiency theorists’ attempt to conquer law – including tort law – in the 1970s, a parallel battle concerning the notions of corrective and distributive justice and the relationship between the two has taken place within the community of justice theorists itself.² Corrective justice is primarily concerned with the moral duty to repair, whereas distributive justice is concerned with distribution of resources within a community. The debate has mainly involved Anglo-American scholars, and my research question is whether and how contributions made by selected leading Anglo-American scholars can be applied in the context of Norwegian tort law.

Until the past decade, the scarce Norwegian debate on the foundations of tort law was dominated by contributions within a legal realist tradition.³ Scandinavian Legal Realism was founded by the Swedish philosopher Axel Hägerström and the Danish philosopher and lawyer Alf Ross, and arose in the early decades of the 20th century. Not all aspects of the legal movement have been embraced by the legal community, but the influence of Legal Realism on both legal reasoning and legal education has been – and still is – profound.

Within the legal realist tradition, functional arguments such as ʻrepair’ and ʻprevention’ have been key to explaining tort law. The question of the philosophical foundations of tort law has rarely been raised. The philosophical foundations of tort law are here to be understood as a set of fundamental values, concerns and/or principles that allow us to interpret the law as it stands in its best possible light (normatively speaking), to provide the best possible normative reasons for why the law ought to be (more or less) as it is, and at the same time generate suggestions for change and for solution of future (hard) cases where the law may not be clear (lacunae).⁴

Over the past decade we have seen efforts by a new generation of Nordic scholars to broaden the debate concerning the philosophical foundations of tort law by taking into account Anglo-American justice theories.⁵ Their argumentation for stepping out of the Nordic legal realist tradition has been that the functional arguments (such as ʻrepair’ and ʻprevention’) on which legal realists base their accounts do not provide as good explanatory power as justice theories based on corrective and distributive justice. However, hitherto none of them have raised the fundamental question of whether and how Anglo-American justice theories can be applied in the context of Norwegian tort law. For example, many contemporary Norwegian tort theorists seem to embrace Weinrib’s approach to corrective justice without taking into account that his approach to private law is formalistic (deeming any positive law resting upon a combination of corrective and distributive justice as incoherent).⁶ In fact, the relationship between corrective and distributive justice has hardly caught anyone’s attention. Given the basic features of a Nordic welfare state and the relationship between the National Insurance Scheme and tort law, the lack of attention given to the relationship between corrective and distributive justice can be regarded as a paradox:⁷ The presence of a distributive element is more intrusive than in any Anglo-American tort law landscape.

In this article I argue that any import of Anglo-American tort theories should be based on a holistic approach, taking into account how different justice theorists deal with the distributive elements in tort law. The Anglo-American justice theory debate is rich, and there are many justice theories to choose from. With the privilege of choice comes an obligation: if we are going to import Anglo-American justice theories in order to explain (justify) Norwegian tort law, we need to at least make sure we import a theory that explains (justifies) the main features of national tort law.⁸ In other words, we need to consider the degree to which Anglo-American theories have explanatory (justificatory) power in a Norwegian legal context. In the following analysis we shall, figuratively speaking, consider how different maps drawn up by Anglo-American justice theorists fit with a Norwegian legal landscape; see sections 2–7.

2. The starting points – Weinrib, Coleman and Gardner studied through a Norwegian lens, the Norwegian Victims Compensation Act

In order to examine whether and how Anglo-American tort theories can be applied in the context of Norwegian tort law, this article scrutinises selected leading Anglo-American scholars’ takes on the relationship between corrective and distributive justice within tort law through the lens of a Norwegian tort law act, the Victims Compensation Act.⁹ Since the Victims Compensation Act is based on EU legislation, the analysis may be considered relevant in a broader European context as well, even if my analysis is limited to Norwegian tort law.¹⁰

The starting point for the analysis is an evaluation of a subrogation provision in section 15 of the Victims Compensation Act (VCA). The provision states that under certain conditions a victim of a violent criminal act may be compensated by the government in place of the tortfeasor. The regulation results in a triangular relation whereby the State pays compensation to the victim in place of the tortfeasor, and then subrogates the victim’s claim for compensation. The first question is whether this Act, clearly containing a strong corrective justice basis, can be understood as a true combination of corrective and distributive justice (see section 4). A second question then arises about the degree to which the conceptions of justice developed by selected Anglo-American scholars fit this triangular relation, which is an important feature of the Norwegian tort law landscape (see sections 5, 6 and 7).

Within the framework of this article, it is not possible to take but a few Anglo-American tort theories into account, and I have chosen to base my analysis on the contributions made by three leading scholars: Coleman, Weinrib, and Gardner. I shall briefly explain my choice. Both Coleman and Weinrib are considered pioneers within modern philosophy of tort law, along with Honoré, Fletcher and Perry.¹¹ They have both made substantial contributions to the debate concerning the philosophical foundations of tort law, representing different theoretical strands. The approach of these two pioneers does not necessarily represent the cutting edge of modern tort theory, but both contributions are fundamental to the work of contemporary tort theorists, and they are of particular interest in a Norwegian context. Weinrib’s account of corrective justice seems to have a strong standing among Norwegian scholars. This does not necessarily mean that Weinrib’s formalistic approach has been adopted in all its particulars, and the question of whether his take on the relationship between corrective and distributive justice can be applied in the context of Norwegian tort law remains to be scrutinised (see section 5). As for Coleman, his approach seems to have certain common features with the Danish legal philosopher Alf Ross’s exposition of justice. From a Nordic perspective, these similarities make it interesting to propose a reading of Coleman in a Rossian light. I elaborate this in further detail in section 6.

Among the many outstanding contemporary justice theorists, Gardner represents a clear counter voice to Weinrib’s formalism, and his approach to corrective and distributive justice is quite different from Weinrib’s. Contradictions attract, and for the analysis of this article I have chosen Gardner since he often argues the opposite of Weinrib (see section 7).¹²

The scratch line for the contemporary debate is, both historically and logically, Aristotle’s theories about distributive and corrective justice.¹³ The Aristotelian concept of justice is well known to many scholars, but to new readers some introductory remarks about the Aristotelian distinction between the forms of justice are made in the following section 3.

3. Aristotle’s concept of justice – the distinction between distributive and corrective justice

In the Nicomachean Ethics, Aristotle distinguished between distributive and corrective justice.¹⁴ Distributive justice concerns distribution of ʻhonour or money or the other things that fall to be divided among those who have a share in the constitution’ (ie among members of society),¹⁵ while corrective justice ʻplays a rectifying part in transactions’ (between man and man).¹⁶

Distributive justice involves a group of individuals, which is why it can be described as social justice. For distribution of goods the principle of equality implies two things: equal individuals are to be treated equally, and just distribution must be based on merit in some sense.¹⁷ Aristotle sees distributive justice as a ʻgeometrical’ justice, and describes this as a mathematical operation, where A and B represents the parties to the distribution, and C and D represents their relative share.¹⁸ The distribution is, in other words, to be based on a proportionate equality – an equality of ratios – with comparative shares and comparative merit: The individual who deserves a lot, gets a lot, and vice versa.

Corrective justice involves individuals and implies a different form of equality, the equality of quantities.¹⁹ Relevant to tort law, if one party has done another an injustice, suffering and injustice have been unequally distributed. This inequality is the injustice that corrective justice means to restore. According to Aristotle, ʻthe law looks only to the distinctive character of the injury, and treats the parties as equal’ (independently of whether a good man has harmed a bad man, or vice versa).²⁰ Aristotle compares the original equality between the parties with a line divided into two equal parts. The inequality, or the unjust, occurs when a segment is detached from one line and added to the other. The correction consists in removing the added part and returning it to the shorter line, and thereby restoring the original equality of the lines. Aristotle connects this line metaphor to a mathematical function, the ʻarithmetic proportion’. According to Aristotle, justice that can be described as recoverable in its form ʻconsists in having an equal amount before and after the transaction’.²¹

Aristotle uses the distinction between geometric and arithmetic proportions to illustrate the distinction between corrective and distributive justice as a metaphor – it is not to be taken literally.

4. Ménage-à-trois

5. Reading of Weinrib’s map in a Norwegian tort law landscape

Can Weinrib’s exposition of the relationship between corrective and distributive justice be used as a basis for explaining the main features of Norwegian tort law, which includes the Victims Compensation Act’s triangular proportion?

Weinrib deems Aristotle’s classification of corrective and distributive justice as conceptual (rather than empirical).³⁵ The difference between the two is therefore related to their structure rather than to a subject matter. This is why, according to Weinrib, neither the original notion of corrective justice nor the distributive form of justice claim a given part of the empirical world. They are both about the external relationship between individuals, and differ only in how they order the interaction between them. Thus the two forms of justice can apply to any external event.

Within a legal system, Weinrib regards corrective and distributive justice to be two modes of ordering, since they represent two different manners in which an external relationship can be coherent: ie a unified conceptual structure where the constituents express one single idea.³⁶ Weinrib asserts that because corrective and distributive justice ʻare categorically different and mutually irreducible patterns of justificatory coherence, it follows that a single external relationship cannot coherently partake in both’.³⁷ This is why, in Weinrib’s view, a positive law cannot be deemed as coherent if any given relationship rests upon a combination of corrective and distributive justifications. If the two forms of particular justice are not held apart, they will undermine each other’s justificatory force. The consequences are, according to Weinrib, extensive for the domain of private law, as the relationships within this area of law are typically bipolar and their coherence is therefore a matter of corrective justice.³⁸ The inclusion of the distributive form of justice will therefore render the relationships incoherent; as this form of justice includes considerations that are alien to the interaction between tortfeasor A and the injured B.

Aristotle’s concept of justice has been criticised for being devoid of any specific content.³⁹ To Weinrib, the formalistic approach represented by the assimilation of justice to mathematics is not a defect; in his opinion, Aristotle’s mathematical operations show that the two forms are ʻcategorically different and mutually irreducible’.⁴⁰

It seems evident that Weinrib’s formalistic approach to corrective and distributive justice is unable to justify (or otherwise explain) the triangular relation introduced by the Victims Compensation Act, since he deems any positive law where relationships rest upon a combination of both corrective and distributive justice as incoherent.⁴¹

Weinrib’s understanding of corrective and distributive justice in terms of their ʻstructure’ rather than as ʻindependent goals’ has been criticised for not being in accordance with the broader Aristotelian tradition.⁴² Aristotle himself described all things in terms of their ends, which must lead to a teleological or at least an anti-formalistic reading of his writings. And suppose it is so – suppose the correct approach to Aristotle is in fact anti-formalistic, how then are we supposed to understand Aristotle’s assimilation of justice to mathematics? In Book I of the Nicomachean Ethics, Aristotle seems to turn down a deductive method within political science, concerning such matters as ethics, politics and law.⁴³ Aristotle writes to his students: ʻit is evidently equally foolish to accept probable reasoning from mathematicians and to demand from a rhetorician scientific proofs’.⁴⁴

So if scientific proof should only be demanded from mathematicians and probable reasoning only be accepted by rhetoricians, why then does Aristotle⁴⁵ still liken corrective and distributive justice to mathematical proportions? Is there an internal tension within Aristotle’s own text? The tension is relieved if Aristotle’s likening of justice to mathematics is read as a metaphor. Aristotle’s use of this metaphor can be understood in light of the clear ideas and the persuasive force provided by mathematics and logic. It is this persuasiveness that one can suspect Weinrib of being seduced by. In any case, Weinrib’s formalism does not seem like a map by which it is possible to orient oneself in the Nordic tort landscape.

6. Reading of Coleman’s map in a Norwegian tort law landscape

Coleman is perhaps the tort theorist who has travelled the farthest in the landscape of corrective justice.⁴⁶ His position on the relationship between corrective and distributive justice is more ambiguous. In this regard, Coleman leaves us with a framework to be completed on ʻanother occasion’.⁴⁷

Coleman sees corrective justice as independent from distributive justice; a given right that has been infringed may very well follow from a (more or less) just distribution among citizens, but a claim in corrective justice must be based upon the fact that A has injured B, not the fact that the wrong violates the principle of distributive justice. Echoing H.L.A. Hart, Coleman claims that his approach to corrective justice is constructed from social practice, and as such not derived from abstract principle.⁴⁸ He can, in that sense, be said to have left Aristotle behind. An attempt to read Coleman in the light of Aristotle would not yield much.

It also seems futile to use Coleman’s framework concerning ʻsupervening social practice’ as a basis for explaining or justifying the main features of the Victims Compensation Act’s triangular proportion. Coleman writes:

A community might decide to allocate accident costs through the tax coffers. Under such conditions, the prevailing social practice might be reflected in remarks like “There’s a mess that you have created; WE have to do something about repairing it.” The existence of a supervening social practice could extinguish duties that would otherwise arise under corrective justice – there would be in that community no practice of corrective justice.⁴⁹

The chances are that Coleman would regard the Victims Compensation Act as such a ʻsupervening social practice’ that extinguishes duties under corrective justice. I believe my analysis of the triangular relation emanating from the Victims Compensation Act shows that this is not the case. The duties are not extinguished.

Coleman’s approach seems to have common features with the Danish legal philosopher Alf Ross’s notion of ʻjustice’.⁵⁰ According to Coleman, only ʻreal rights’ give rise to a claim within corrective justice.⁵¹ Such ʻreal rights’ are either supported (or required) by distributive justice or by a general moral principle requiring protection of rights that would not follow from the distributive form of justice. Coleman notes that such a general moral principle would have to be ʻaccepted by nearly any one’. Thus, he leaves us with a notion of relevant ʻrights’ with the qualities of a clutch, connecting some sort of moral basis (consensus) to corrective justice.

Alf Ross did not base his exposition of justice on a division between corrective and distributive justice. Significant to Ross’s exposition is the division of justice into the ʻformal’ and ʻmaterial’ demands of equality.⁵² Formal equality implies that equal instances are treated equally.⁵³ To Ross, justice as formal equality does not tell us much, as its practical content is dependent on conditions outside the principle itself. Thus the ideal of equality as such does not imply anything but a correct application of a general rule, and can therefore be expressed as a demand for rationality – meaning that any treatment of individuals shall be predetermined through objective criteria.⁵⁴

Coleman’s exposition of corrective justice is similar to Ross’s ʻformal equality’ in the following respect: Coleman’s notion of corrective justice implies that if A injures B, then corrective justice has predetermined the effects, which are that A has a duty to pay compensation and B has a right to be compensated.

If we now look at Ross, the next question concerns material equality, that is what is to be deemed as equal, or which criteria ought to be decisive of what is deemed equal. Ross perceived material equality as a subjective and emotional expression for the resistance against any given scheme:⁵⁵

To invoke justice is the same thing as banging on the table: an emotional expression, which turns one’s demands into an absolute postulate. That is no proper way to mutual understanding. It is impossible to have a rational discussion with a man who mobilizes ʻjustice’, because he says nothing that can be argued for or against. His words are persuasion, not argument ….⁵⁶

Coleman’s approach to ʻreal rights’ (giving rise to a claim within corrective justice) can be regarded as having the following common features with Ross’s exposition of material equality: Coleman does not develop in any way what moral foundation might be sufficient for justifying ʻreal rights’; he just stipulates that such moral foundation has to be ʻaccepted by nearly anyone’. Coleman acts exactly as Ross would require a legal scholar to do: supplying a conceptual framework for discussion, but being barred from material discussion based on reason. By stating that the moral foundation of ʻreal rights’ must be accepted by ʻnearly anyone’, Coleman brings in a democratic problem which Ross, on his side, solved by stating that assessments of what is just must be made by legislators or judges (based on what Ross called the judge ideology).⁵⁷

So, where does this leave us, and our triangular relation? The map Coleman provides us with may very well fit with a Norwegian legal landscape, but it seems that his exposition of justice does not bring us much further than Ross’s ideas. From a Nordic perspective, it seems, then, that Coleman’s theory does not bring any progress if we compare it with what we already have (Ross).

7. Reading of Gardner’s map in a Norwegian tort law landscape

When considering the place of distributive justice in tort law, Gardner describes his stand as ʻalmost diametrically’ opposite to Weinrib’s; Weinrib considers distributive justice as being ʻextinct’ or ʻalien’ to the law of torts.⁵⁸ According to Gardner certain questions of distributive justice are central to the law of torts, and cannot but be faced by those who administer and develop it, precisely because the law of torts is a site of corrective justice.⁵⁹

Gardner emphasises that there are ʻrights to and duties of corrective justice that exist independently of the law, and independently of any other kind of use, observance, recognition, or adoption by anyone’.⁶⁰ He refers to the process of making these rights and duties part of the law – either intentionally or accidentally – as their ʻinstitutionalisation’. One of the sub-questions that arise when asking whether the institutionalisation has been done well is whether it has been done ʻjustly’.⁶¹ In tort law, classifying some wrong as a tort entails a legal right to corrective justice. To Gardner, the question of whether tort law should give this recognition – ʻcomplete with generous terms for power-sharing and cost-sharing as between the aggrieved party and the legal system’ – is a question of distributive justice.⁶² By recognising a wrong as a tort, some people are given the privilege of official support in their personal affairs (access to justice). However, as stressed by Gardner, ʻwhat is distributed remains something irreducibly corrective’.⁶³ In Gardner’s opinion this lends explanatory priority to corrective over distributive justice ʻin what Weinrib might call the “immanent rationality” of tort law’.⁶⁴

Gardner holds that his continuity thesis helps us understand why we want to do corrective justice and see corrective justice done. Another question is when and how, if at all, we should support the doing of corrective justice. Gardner sees the law as a method of support. The chances are that he would see the Victims Compensation Act as such a method, enforcing corrective justice and containing ʻ“localized” distributive justice’.⁶⁵ Gardner opposes Weinrib’s idea that corrective justice is the only true function of tort law, since ʻlocalized’ distributive justice also has ʻa key role to play’ doing justice among parties in tort law.⁶⁶ To me, this seems to be the kind of role that the subrogation provision has been assigned for in the Victims Compensation Act.

8. Concluding remarks

The Anglo-American debate on the philosophical foundations of tort law has inspired Nordic tort scholars to scrutinise tort law in new ways during the past decade, bringing in innovative perspectives. Still, when Nordic scholars have sought to broaden the philosophical foundations of tort law by basing their expositions on Anglo-American justice theories, no one has hitherto raised the fundamental question of whether and how these theories can be applied in the context of Norwegian tort law. The main argument for stepping out of the Nordic legal realist tradition has been that a function-based theory does not entail the same explanatory power as justice theories based on ideas of corrective and distributive justice. I believe my analysis shows that this may not always be the case, in particular with regard to Weinrib’s and Coleman’s expositions.⁶⁷ In my opinion, the quest for new theories with stronger explanatory and justificatory force is important in order to move legal doctrine forward. However, theories that have such explanatory and justificatory force within one national legal order (the US or the UK) may not necessarily represent real progress in new legal surroundings. In a Norwegian context, we need to take into account whether and how these theories also contain the power to explain and justify the main features of the Norwegian law, which we presuppose them to explain and justify. As the analysis of the Victims Compensation Act shows, this may not always be the case when theory is imported from Anglo-American legal scholarship.