INTRODUCTION
In chapter one of this book, I identified three assumptions that underpin the default vertical approach towards constitutional rights. In chapters two and three, I argued that existing judicial (and scholarly) models that attempt to depart from default verticality nonetheless remain tied to one or more of its underlying assumptions.
This is because, in different ways, these models (state action and indirect horizontality) continue to treat the state as the primary normative site for applying the constitutional rights framework. This led to the question - asked at the end of chapter three - whether the problems with default verticality could be resolved by jettisoning the vertical/horizontal divide altogether, and applying constitutional rights in the ‘private sphere' in exactly the same manner as we apply them against the state. Unbounded direct horizontal- ity, however, brings problems of its own. The structure of the rights framework resists any direct transplantation from state to non-state entities. The attempt to do so raises issues that are both conceptual (what do we do when the subject of horizontal rights obligations is also a rights bearer?) and pragmatic (the collapsing of all existing private law domains into the constitutional rights framework). Consequently, any model that advocates direct horizontality would necessarily have to be a bounded, or a limited model: that is, it would have to set principled constraints upon direct horizontality, in terms of both which rights apply horizontally and which ‘private' relationships they apply to. In the jurisdictions that I have surveyed in this book - including jurisdictions where the constitution does make provision for direct horizontal application - courts have (so far) been hesitant to develop a principled account of bounded direct horizontality. In this chapter, therefore, I will turn to contemporary scholarly accounts. In particular, I will study the models developed in Johan van der Walt's The Horizontal Effect Revolution and the Question of Sovereignty (2014) and Jean Thomas's Public Rights, Private Relations (2015).This chapter considers the bounded horizontality models of Johan van der Walt and Jean Thomas. I examine these two contemporary authors because, like the arguments developed in this book, they too reject both default vertical- ity as well as the existing models of departure from it. Furthermore, they too articulate a comprehensive model of bounded horizontality that attempts to delink itself from some or all of the assumptions underlying default verticality. This chapter, thus, aims to set out their respective accounts, and examine whether and to what extent they are successful in setting out a plausible model of bounded horizontality on the lines indicated above. This, in turn, will set the stage for chapter five, where I will articulate my own institutional approach towards horizontality that - I will argue - incorporates the insights of both Van der Walt and Thomas, while avoiding some of the weaknesses of their respective models.
This chapter will begin with an account of Van der Walt and Thomas's respective models. It will not, however, be a simple summary. Rather, I will attempt to reconstruct their arguments from the perspective of the three assumptions of default verticality discussed in chapter one and examine the approach that these scholars take towards those assumptions (section II).
Next, I will examine the strengths and weaknesses of each model. I will argue that Van der Walt correctly recognises that the key to the application of the horizontal rights framework lies in how certain private relationships are nested within institutional structures that lend them peculiar salience. Van der Walt, however, insists on limiting these institutional structures to the interactions between ‘social majorities and minorities' - a limitation that (I will contend) is misplaced.
On the other hand, Thomas correctly argues that the horizontality enquiry ought to focus on questions of power and vulnerability, and how these attributes frame certain private relationships. Thomas, however, goes on to argue that horizontal rights in the ‘private domain' must retain the structure of established public law rights, and that therefore the (private) subject of horizontal obligations must exercise state-like power (for example, by holding a monopoly over access to a basic good). This, I will suggest, is unduly limiting. In many cases that appear intuitively fit for applying the rights framework, it is not that a private actor individually exercises a monopoly (such as the racist house owner or landlord in a racist society), but that the background social institutions (in this case, racism) accord to their individual action a salience that goes beyond the concrete power that they exercise (section III).
Combining the core insights of Van der Walt (relevance of background institutions) and Thomas (focus on power and vulnerability), however, yields the basis for the institutional model. I will conclude this chapter by briefly flagging this point, before articulating and developing the model in chapter five (section IV).
II.