The question this contribution sets out to address is whether or not law can be regarded as a science. This notion is readily accepted by many, yet it is submitted that a proper theoretical justification for such an assumption is usually missing. The traditional primary sources of law, South African case law and legislation, distinguish between legal practice and legal science, but the basis of the distinction is not clear. However, an entire body of literature in the philosophy of science has developed around the question of when a discipline will amount to science. Various demarcation criteria proposed in the philosophy of science are considered. These include that science uses the scientific method, is susceptible to falsification, is puzzle-solving within a paradigm or renders beneficial results. None of these criteria offers a satisfactory solution to the problem. The proposition by a group of philosophers including Herman Dooyeweerd, Marinus Stafleu and DFM Strauss, that the answer to the demarcation question is to be found in modal abstraction, is then considered. Modal abstraction amounts to a consideration of reality (persons, things, theories and rules) from one or more defined point(s) of entry. It is an artificial and learnt manner of thinking as it approaches reality from the perspective of one of the modalities of being. For example, juridical abstraction would mean that a cow is considered as the object of someone's proprietary rights. An abstract idea of the cow's characteristics, from a juridical point of view, is formed and the rules of property law are applied. A number of South African legal philosophers, amongst others Van Zyl, Van der Vyver and LM du Plessis, have followed this approach. The South African legislature has also attempted to define the terms "science" and "research", mainly for funding purposes. These definitions are considered and the conclusion is that they do not provide the clear-cut answers one would expect. It will be argued that the nature of activities will determine whether an endeavour is scientific or not. The conclusion is that an alignment of the demarcation criterion developed by Strauss and others and the statutory definitions can provide a workable demarcation criterion. This "test" is then applied to the activities of law students, academics, practitioners and judicial officers to determine when they will be practising "science".
KEYWORDS: Law as science; philosophy of science; demarcation criteria; scientific nature of law
