Whose Fault is it Anyway? A Reflection on the Interpretation of Section 1(1)(A) of the Apportionment of Damages Act (No. 34/1956) as Amended
Tobias J.G. Louw*
Professor, Department of Philosophy, University of Fort Hare
Senior Lecturer, Department of Private Law, University of Fort Hare
This paper demonstrates the hermeneutical dimension of decision making in law with specific reference to the application of the first section of the Apportionment of Damages Act because of the divergent interpretations found in the application of this section in reported cases. These disparities raise critical questions about theoretical assumptions regarding interpretation as such. A reflection on the philosophic concept of interpretation confirms the need to critically re-examine existing interpretation practices. It is submitted that future decisions concerned with fault and the apportionment of damages take cognizance of a hermeneutical framework, which is more consistent with the Constitution of the Republic of South Africa, and thereby advance the cause of justice. The hermeneutical method of analysis is used to disclose the assumptions and meanings underlying interpretative practices in relation to the issue of the apportionment of damages, as well as to show the potential of mediating between conflicting arguments on this matter.
Restorative justice ensures that when damage has been caused by someone’s blameworthy conduct, the party at fault will be obliged to make good that loss. This legal norm is objectively expressed and codified, and therefore widely accepted as the basis of the social codes of morality, civil law, and international law. However, because these social codes protect different interests and advance disparate values, each will differ in what it regards as damaging and blameworthy conduct. Furthermore, each code will rely on different rules to determine how blame is to be imputed and which interests are to be protected. Therefore, for example, morality and the civil law will, for the above reasons, allocate culpability very differently when assessing a person’s conduct.
We contend that the effort to make sense of conflicting meanings, with a view to resolve conflicts of interpretation and understanding - whether in law, politics, morality, science, or everyday life - could be decisively assisted by insights from the theory of the rules of interpretation of meaning, or hermeneutics.1
The process of how different social codes determine fault will be explored below. It is further intended to show how important it is that the rules applied by social codes must be well defined so that they can be differentiated. This would diminish the possibility of either confusing or conflating the rules to be applied in establishing and apportioning fault when the same conduct can be evaluated by more than one code. In the law of delict, for example, no one is obliged to help a person who has been injured if there is no legal obligation to do so. However, in the context of some moral codes, a person who voluntarily assists an injured person is regarded as doing good. So, for example, in the familiar Biblical story of the Samaritan2 who helps the wounded Israelite, a member of a group he despises, is praised for being good. If, however, he had not helped the injured Israelite, his omission would have been deemed morally culpable. However, the latter judgment would be forthcoming from a Hebrew perspective, and would most likely have been viewed in a contrary way by Samaritan values, since the Jews usually had no dealings with Samaritans3.
These differences in determining fault illustrate the distinctions between morality and law, a difference that is rooted in the respective interests and values that each social code seeks to protect. It is thus of cardinal importance that in the process of adjudication, the interests and rules which apply to each code are well defined so as to avoid confusion.
The interests protected by morality and the law of delict are usually the same, but the rules used to determine fault often differ. By confusing the rules to be applied, fault can be incorrectly allocated as seen in the omission discussed above. If the rules of morality were applied in a legal claim by the Israelite instituted against the Samaritan for his failure to aid him, then it would be held that fault and liability be attributed to the Samaritan. According to the rules of delict, however, no fault would be attributed to him for this omission. This illustrates how judgments that contradict each other can be made when the rules of one social code are used to determine fault in another social code. It also shows that in order to prevent the confusion and injustice referred to above, a sound understanding of how the concept of fault is interpreted and applied in the context of each social code is needed.
Similarly, the social codes of national and international law allocate fault and liability differently. In the domain of national law, no one can be criminally charged under a law that had not been promulgated at the time when the act complained about was committed.4To punish anyone retrospectively is in conflict with the principle of legality and is based on the principle that a person must be aware of the illegality of their act at the time when the harmful act was perpetrated, and on the grounds of fairness. This equitable rule is not always upheld in the sphere of international law. After World War II some high-ranking German Wehrmachtofficers were tried and found guilty under laws which had not been part of the German legal system when the crimes they were charged with were committed. The rules applied by national law to allocate fault or guilt are clearly not the same as those applied by international law. This happens because the interests and values that these two codes uphold differ. The rules to allocate fault would thus also differ when the rules or norms of any one of these social codes are either confused or conflated when blame could easily be incorrectly allocated, thereby causing an injustice.
After World War II, the Allied powers implemented a “de-Nazification” programme that every German had to undergo. This was done to prevent the atrocities committed during the war being repeated and because every German was deemed responsible for the conduct of the Nazis. Prominent opponents of the Nazi regime objected to this programme. They stated that it opened a floodgate that would allow personal hatred to masquerade as civic virtue, and to hold that each individual German was guilty of the crimes committed in concentration camps and foreign lands was untrue and unjustified. At this time, Pope Pius XII had declared that “it is wrong to treat someone as guilty whose guilt cannot be proved, only because they belong to a certain community”.5This was said to prevent further injustice.
The realization that the cultural differential will unavoidably continue to effect the interpretation of surface meanings applies to all levels of social, economic and legal interaction, and even more so in today’s multicultural societies. Trompenaars6offers ample evidence of this. He shows by means of comprehensive empirical studies that even in seemingly similar cultural contexts (e.g. of the European Union), the clash between implicit core value systems are brought to the foreground whenever the interpretation of meanings on the external, explicit level (such as artefacts, products, services, rules and codes) are questioned and become the object of serious debate.
Similarly, in philosophical anthropology there are in-depth debates about the meaning of fault and how these are more than often abstracted out of rational debate. In spite of the appearance of a highly abstract debate, these core arguments provide extensive orientation and substance to more superficial discussions on all related issues. An example is provided in the bulky trilogy on the philosophy of the will by Paul Ricoeur, to rethink the existentialist concepts of choice, motive, decision, action, fault, fallibility, and evil in relation to the respective human faculties of willing, thinking, and feeling.7 This redirection of thinking about the fault, for instance, from the perspective of the hermeneutics and phenomenology of the interpretation of symbolic meaning, opens the door to the narrative redesigning8 of all the assumptions we have about the “linguisticality” of our understanding of our lifeworld. Ricoeur shows with impunity that human finitude and the tendency to err (commit faults), can never be justifiably reduced to a harsh set of parameters such as motive-decision-action, but needs to take account of the ethics of fallibility as well as the detours on which the symbolic of evil takes consciousness. He states, “Man can be evil only in accordance with the lines of force and weakness of his functions and destinations.”9
The above references make it evident that the concept of fault can be interpreted in divergent ways. This article intends to show that the courts, like the social codes discussed above, have interpreted fault in different ways when faced with disputes where fault must be apportioned between two or more parties. A careful analysis of fault, as established in the law of delict, is thus essential to gain an insight into the logic of its application. It is submitted that the disciplines of philosophy will aid the legal process to determine and apportion both fault and liability. As seen above, injustice can occur when the rules of one social code are conflated with the rules of another. These disciplines will, furthermore, aid the legal process by defining and differentiating those interests that the law protects in contrast to those upheld by other social codes. This article will critically discuss some court decisions to illustrate this and suggest that hermeneutical philosophy can alleviate many of the problems encountered when applying the section that apportions fault and liability in delictual claims for damages.
Before legislation regulated the process of apportioning fault when more than one party had contributed to the damage, which led to the dispute, the legal rule was that the party who had the last opportunity to prevent the damage forfeited their right to claim any compensation.10This rule led to inequitable results and the Apportionment of Damages Act11was promulgated to rectify this. The “last opportunity rule” was abolished in terms of section l(l)(b) of the Act. This enabled courts to determine the respective degrees of fault of the plaintiff and the defendant, and thereafter apportion liability in accordance with section l(l)(a) of the Act.12 This section would clearly only apply when more than one party contributed to the damage that caused the legal dispute. The Act also allows a court to allocate fault and liability on the grounds of vicarious liability.13
A plaintiffs claim for damages is diminished to the extent that a court deems just and equitable after considering the degree by which the claimant and defendant are held to be delictually at fault. The degree of fault of each party who contributed to the damage would be expressed as a percentage. The concept of fault, as applied within the context of the Act, will be discussed below. The question of moral blame will however not be considered, for the reason of firstly narrowing down the discussion, and secondly because the interests protected by the law are not the same as those upheld by morality.14
2 DISSIMILAR APPLICATIONS AND DIVERGENT INTERPRETATIONS OF FAULT IN RESPECT OF THE APPORTIONMENT OF DAMAGES ACT
Since the courts have not applied the guidelines for determining fault in a consistent manner, the question to be asked is whether these divergent applications constitute a conflict of interpretations. When adjudicating cases where the Act was applied to determine and apportion fault, some courts have relied on a literalist and others on a contextual interpretation to justify the decisions they made. This has caused some writers, referred to below, to comment that legal uncertainty has been created about the application of the Act and section. The uncertainty also relates to the question of whether fault includes intent in section 1 of The Act.15It is possible that the 1996 Law Commission wished to limit confusion about this concept when it recommended that the word “fault” in the context of the Act be replaced by the words “negligent conduct”.16
The rules used to determine and apportion the different degrees of fault of each party who contributed to the damage were laid down by Thompson J.A. in South British InsuranceCompany Ltd v Smit17.Here the court held that, having determined the degree by which a plaintiffs conduct deviated from that of a reasonable person, their fault would be expressed as a percentage. It was further held that the figure arrived at after subtracting the percentage of the plaintiffs fault from 100% would be the percentage of the plaintiffs fault, because the two percentages added together would amount to 100%. A simple arithmetical calculation would thus determine the defendant’s degree of fault. This method for apportioning fault was again applied and followed in the later Appeal Court decision of AA Mutual Insurance Association Ltd v Nomeka18.This raises the simple but critical question of how a defendant’s delictual fault can be determined by using an arithmetic calculation and not applying the rules of the law of delict to establish the defendant’s delictual fault. However, in the Appeal Court decision of Kruger v Coetzee19the degrees of fault of both plaintiff and defendant, who had both contributed to the damage, were determined by applying the rules of delict to each party separately. This judgment was given four years after the decision of South British Insurance Company Ltdv Smit,20thereby creating a new method to apportion fault in cases falling within the ambit of the section. By following the arithmetical method to establish the defendant’s degree of fault, the Nomeka21judgment ignored the rules of judicial precedent when the court did not follow the method to determine fault as laid down in Kruger v Coetzee22. This was done without giving reasons. It is submitted that to use delictual rules to establish a defendant’s fault is accepted legal practice and, with respect, the honourable court erred by applying the arithmetic method to establish a defendant’s fault. Because a person’s delictual fault and liability is, with few exceptions, based on their negligence or intent, it is evident that by evaluating their conduct, and not another’s, that their fault is determined. It is therefore submitted that to determine a defendant’s fault by relying on the plaintiffs conduct is not only unfair but contrary to the basic principle that each person is personally liable for their own wrongful and blameworthy conduct.
The authors Van der Walt and Midgley support this mechanical application of the section under discussion. According to their interpretation, the section does not stipulate that a comparison be made between the degrees of fault of the plaintiff and the defendant or any other parties. They also state that this section does not lay down that the conduct of the defendant and other parties be separately evaluated according to delictual principles to apportion their respective degrees of fault. The learned authors consequently maintain that a plaintiffs claim will be reduced by the percentage that the plaintiff was at fault23, and that the degree of delictual fault of all other parties would be determined by the degree that the plaintiff was at fault. As stated above, to attribute fault to a defendant by only using the degree of the plaintiffs fault, is to hold the defendant liable on the basis of conduct which was not the defendant’s, and over which he had no control. This is in conflict with delictual principles and is neither fair nor equitable. For these reasons, it is submitted that as a legal method to establish fault and apportion liability it is seriously flawed and should not be applied.
In the case of Jones NO v Santamversekeringsmaatskappy Bpk-the Appeal Court was again faced with the question of how fault was to be apportioned when the plaintiff and the defendant had both contributed to the damage.24 The honourable court first evaluated the conduct of the plaintiff and defendant separately within the context of delictual principles. The ratio between the respective degrees of fault of each party was then calculated in order to establish the percentage of fault that could be attributed to the defendant. This percentage would also be the percentage of the damages the defendant would be obliged to compensate the plaintiff. It is respectfully submitted that apportioning fault in this way is more equitable because each party is imputed with the degree of fault that is directly related to their own conduct. The method of apportionment, as applied by the Appeal Court, effectively abandoned the mathematical approach because the court interpreted the section differently in order to justify its method of apportionment. This method of establishing and apportioning fault was also supported by many writers because it was seen as advancing a more equitable apportionment of both fault and liability.25
Yet another interpretation of this section is found in the judgment of General Accident Versekeringsmaatskappy SA Bpk v Uijs NO26Here the factors of fairness and equity, referred to in the section, and the boni moreswere taken into account by the court when the respective degrees of fault of the parties were apportioned. In this case, the plaintiff had not fastened his safety belt while travelling as a passenger in a motor car. The driver of the motor car had insured his vehicle with the defendant, and his reckless driving caused the accident that injured the plaintiff severely. The plaintiff and the defendant were each held to have been 50% at fault. The court, however, refined this finding after evaluating the conduct of each party in the light of the factors mentioned above. It was finally decided that the plaintiffs failure to fasten his safety belt was not as reprehensible as the defendant’s reckless driving and it would thus not be fair to hold that both parties were equally at fault. The court also considered the community’s strong condemnation of reckless driving when determining the respective degrees of fault of the two parties. On the basis of fairness and equity, it was decided that the defendant was 66% at fault and the plaintiff was 33% at fault. The defendant was thus obliged to pay the plaintiff 66% of the damages claimed.
In the recently published book, The Law of Delict in South Africa27, another interpretation of the section is noted. The authors suggest that a party’s fault is determined, not by establishing the fault by applying delictual rules, but, by relying on the court’s “gut feeling” of what a fair allocation would be. This approach is open to criticism, firstly on the grounds that fault would be established on the basis of the subjective feelings of a court. Secondly, the Appeal Court has laid down that negligence is determined by using the objective standard of the reasonable person.28 Thirdly, a person found to have been at fault may lodge an appeal because a subjective factor, like a judge’s “gut feeling”, is not one of the elements that legally define or determine fault. Their reference is not supported by any legal authority and is further undermined by their discussion of negligence, when in the same book it is stated that negligence is objectively determined29. It is common cause that “fault” is a wider concept than “negligence”. Despite the focus on the latter, assumptions about the former continue to inform both meanings. The Appeal Court has clearly laid down that fault must be objectively determined30. For these reasons it is submitted that the subjective (“judge’s gut-feel”) approach to apportionment should not be followed.
In the case of Harrington NO v Transnet Ltd31,two security guards were struck by a Metrorail train that was travelling at an unscheduled time. As a result, they suffered serious injuries. They instituted action against Transnet, the holding company of Metrorail, and the train driver to recover the damages they had suffered. They argued that because the train driver had not given them sufficient warning of the train’s approach, the driver and his employer, on the basis of vicarious liability, should be held liable for the damage they suffered. To guide him in the process of apportioning fault between the various parties, Blignault J. took note of how the Australian, Canadian, and United Kingdom courts had established the degrees of fault of several parties under similar circumstances32. The method of apportionment applied in these jurisdictions was followed by Blignault J. This meant that the court evaluated the respective degrees of fault of the plaintiffs and defendants separately. The plaintiffs were held to be jointly 33% at fault because they had not kept a proper lookout while walking next to the railway line. The two defendants were, jointly and severally, found to be at fault for the remaining 66% and were thus also jointly and severally liable to pay the plaintiffs two thirds of the damages claimed. By relying on extra-legal texts, such as the decisions of courts outside the South African jurisdiction, the court illustrated how the Constitution and extra legal texts play an important role in the process of interpretation.
In the case of Vorster v Santamversekeringsmaatskappy Bpk33,the two defendants were participants in a drag race, and the plaintiff was a passenger in one of the vehicles. He had not secured his safety belt. One of the drivers had insured his car with the first defendant, and when the vehicles collided, the plaintiff was badly injured. The court proceeded to evaluate the conduct of each of the three parties separately so as to apportion the degrees of fault fairly.34On appeal, the finding of the trial court, that the plaintiff had been 30% at fault and the two defendants had been 70% at fault and were thus jointly and severally liable for 70% of the damages, was confirmed35. In the Harringtoncase, discussed above, the defendants, Metrorail and the train driver were treated as a unit for purposes of determining their degrees of fault, whereas the defendants in the Vorstercase were separately evaluated so that, in the words of the court, a fair apportionment could be made. The failure of the court in the Harringtoncase to follow the method of apportionment that was applied and confirmed by the Appeal Court in the Vorstercase36was criticized by two authors.37They stated that the degrees of fault of the driver and his employer should not have been evaluated as a unit but should have been separately evaluated and apportioned, as was done in the Vorstercase. Because the apportionment of fault effectively determines the percentage of the damages that the defendant is obliged to pay the plaintiff, it could create financial problems for a defendant if this was not fairly done. So, the train driver could have been financially crippled because of the apportionment applied in the Harringtoncase. In the jurisdictions of Austria, Belgium, Germany, Switzerland, and U.S.A., the degree of fault of each party who contributed to the damage is separately evaluated.38 This, it is submitted, is the most equitable way to apportion fault in terms of the section because the personal conduct of each person would be judged by applying the delictual rules for determining fault to their conduct. Such an interpretation and application of the section would be more contextual since it examines the situation pertaining to each person’s conduct.
Further evidence of how the legal discourse within the academic and judicial spheres can refine the interpretation of the section is seen in the decision of the High Court in the application of Right v Medi-Clinic Ltd39.The applicant, an obstetrician, who with the respondent, a hospital, were found to have been equally at fault for the injuries suffered by a child at childbirth. The obstetrician approached the High Court with the request that the degree of fault attributed to him be diminished because the hospital, he alleged, had been more negligent than he had. His application was based on section 2(8)(a)(iii) of the Act which empowers a court to vary the degree of fault that attributed to a party. This section is similar to section l(l)(a) and serves a similar purpose. When the degree of fault attributed to a person is not supported by evidence or is excessive, an application under section 2(8)(a)(iii) can be made to vary the degree. Such an application could also vary an apportionment if a party was prejudiced because their personal conduct had not been evaluated but, because it had been so closely linked to another party’s, that it was treated as a unit for purposes of apportionment. This application would preclude the respondent and the applicant being seen as a unit because they are on opposing sides in court. In the Rightapplication, the court held that the hospital was at fault to a greater degree than the obstetrician was. This is, with respect, an equitable method of apportioning degrees of fault in a case where two or more parties might appear to be acting as a unit. The powers given to the court by this application procedure make it clear that the degrees of fault of each party who contributed to the damage will each be separately determined according to rules of delict.
Some writers have, because of the different ways that the concept fault has been interpreted, concluded that legal uncertainty has been created.40Other writers, on the other hand, have maintained that justice will be promoted when factors outside the written text, but within the confines of the law of delict, are used to interpret and clarify the meaning of fault.41It is clear from the cases discussed above that there are divergent ways that fault has been interpreted and applied. Broadly speaking, a tension exists between a literal and a contextual interpretation. For example, one writer42 has called for a realism that takes into account other legal texts and relevant factors in the process of interpretation. Others prefer to rely on a literalist interpretation of a text because it ensures greater legal certainty in contrast to a more open and contextual interpretation. The literalist approach was introduced into South Africa by De Villiers CJ.43Because of the tension that has arisen between the literal and the contextual interpretations of the section, a reflection on the process of interpretation is called for.
Those who hold that the meaning of a legal text is to be found in the intention of the legislature would apply a formal and literalist method of interpretation to establish the meaning of a text. The method of interpretation effects the way a text is applied. Judges who use a literalist method would play a stenographic role because to discern the intention of legislators in their absence in order to establish the meaning of a text would be a very abstract process. When judges interpret a text, they are also bound to follow the meaning given to the text in earlier decisions. The constructs of another judge could restrict the process further. This formal and literalist approach is both abstract and restrictive and although it ensures legal certainty this is gained at the price of neither considering the context of the conduct complained of not taking other relevant factors into account. This raises the question of the methodology of interpretation.
The assumption that there is legal certainty about an interpretation and that the subjective intention of an author can be objectively known has been criticized by both legal authors and philosophers.44For most of the twentieth century the main philosophical schools such as linguistic analysis, hermeneutics, structuralism, deconstructionism, phenomenology, and Neo-Marxism have all separated the author’s intention from the text. All maintain that there is a separate “autonomy of the text/literary work” from the author. Thus Palmer states, “The author’s intentions ... are held rigidly separate from the work; the work is a ‘being’ in itself, a being with its own powers and dynamics.”45The examples previously discussed showed how the courts practically calculated the apportionment of damages between the plaintiff and defendant by means of reducing damages in a just and equitable manner. It also showed how respective courts have interpreted this section of the Act differently, and raises the question of whether a narrow and literalist interpretation can justify its claim to be the exclusive and correct interpretation of this section. Dissimilar applications are accounted for by the varied interpretations of this section. This underlines the view that interpretation means, in effect, that to determine the value of an item, act, or event in relation to other items, acts or events, one must ask whether legal certainty can be achieved so simply, especially when the interpretation process requires a selective prejudgment of what is to be interpreted.46It also becomes clear that the claim of “legal certainty” as found in a literal interpretation cannot be justified because its method of interpretation proceeds from a misapprehension of the general process of interpretation. Such a restrictive interpretation does not accord with the Constitution or with the Interpretation of Legislation Bill, 2006.47