be at the same time a member of more than one branch. In this way each of the branches will be a check to the others and no single group of people will be able to control the machinery of the State.
This stark, extreme doctrine we shall then label the “pure doctrine,” and other aspects of the thought of individual writers will be seen as modifications of, or deviations from, it. It is true, of course, that the doctrine has rarely been held in this extreme form, and even more rarely been put into practice, but it does represent a “bench-mark,” or an “ideal-type,” which will enable us to observe the changing development of the historical doctrine, with all its ramifications and modifications, by referring to this constant “pure doctrine.” We shall not go as far as to say that only a thinker who fully subscribes to the above formulation is a “separation of powers theorist,” for this would exclude most of those who have written on the subject and whose intentions were closely in line with the general ethos of the doctrine, but clearly all these elements must be present to some extent for a writer to be considered in this category. Many writers have of course contributed to the development of the theory by evolving one or more elements of it, without being separation of powers theorists—indeed, whilst rejecting the doctrine. Thus the idea of the functions of government has been evolved in large part by the theorists who implicitly or explicitly rejected other essential elements of the doctrine.
The first problem presented by the theory outlined above is its commitment to “political liberty,” or the exclusion of “arbitrary power.” Clearly the viability of the whole approach may turn upon the definition of liberty chosen. Thus perhaps one of the most persuasive general criticisms of the doctrine is that it has been associated with an essentially negative view of political liberty, one too concerned with the view of freedom as absence of restraint, rather than with a more positive approach to freedom. The concern to prevent the government from encroaching upon individual liberty leads to measures which weaken it to the point where it is unable to act in order to provide those prerequisites of social and economic life which are essential if an individual is to be able to make proper use of his faculties. The decline in the popularity of the doctrine in the twentieth century, both in the United States and in Britain, is closely related to the recognition of the need for “collectivist” activities on the part of government, which require a co-ordinated programme of action by all parts of the government machine. The doctrine of the separation of powers is clearly committed to a view of political liberty an essential part of which is the restraint of governmental power, and that this can best be achieved by setting up divisions within the government to prevent the concentration of such power in the hands of a single group of men. Restraints upon government are an essential part of the view of political liberty enshrined in this approach, but we shall have to consider the extent to which the proponents of the doctrine also recognized that a minimum degree of “strong government” was also necessary to political liberty, and the possible ways in which the tenets of the doctrine are compatible with the minimum needs of government action in the twentieth century. Indeed it will be assumed that the recognition of the need for government action to provide the necessary environment for individual growth and development is complementary to, not incompatible with, the view that restraints upon government are an essential part of a theory of political liberty.
The first element of the doctrine is the assertion of a division of the agencies of government into three categories: the legislature, the executive, and the judiciary. The earliest versions of the doctrine were, in fact, based upon a twofold division of government, or at any rate upon a twofold division of government functions, but since the mid eighteenth century the threefold division has been generally accepted as the basic necessity for constitutional government.5 We may not today take the scriptural authority that John Sadler in 1649 propounded as the basis for a threefold division—“And why may not the Sacred Trinity be shaddowed out in Bodies Politick, as well as in Naturall?”6—but something of a mystical quality seems still to surround this method of organizing the agencies of government. In the eighteenth century the idea of a balance or equilibrium in the system of government which depended upon the ability of any two of King, Lords, and Commons being able to prevent the third from exceeding the proper limits of its power, provided a basis for the idea, at any rate, of an odd number, rather than an even number, of governmental agencies, but today such a justification seems to have disappeared entirely,7 and in fact it is often difficult to force the manifold agencies of a modern system of government into these three categories. Nevertheless this division does reflect important, continuing elements in liberal democratic theory. The growth of three separate branches of the government system in Britain reflected in part the needs of the division of labour and specialization, and partly the demand for different sets of values to be embodied in the procedures of the different agencies, and in the representation of varying interests in the separate branches. This aspect of the doctrine, although usually assumed by political theorists rather than explicitly developed, is clearly central to the whole pattern of Western constitutionalism. The diffusion of authority among different centres of decision-making is the antithesis of totalitarianism or absolutism. Thus in the totalitarian State every aspect of the State machine is seen merely as an extension of the party apparatus, and subordinate to it. A continuous effort has to be made to prevent any division of the machine from developing its own interest, or from creating a degree of autonomy in the taking of decisions. In practice the pressures which operate against this attempt to maintain a single monolithic structure are too strong, for the price in inefficiency which has to be paid is too high, and of necessity rival centres emerge in the bureaucracy and in industry or elsewhere. But the “ideal” of the totalitarian state is that of a single all-embracing agency of government.
The “separation of agencies,” therefore, is an essential element in a theory which assumes that the government must be checked internally by the creation of autonomous centres of power that will develop an institutional interest. Without the other elements of the doctrine of the separation of powers being present we might still expect some limitation on the ability of a single group to dominate the government if separate agencies are established. Even if the personnel of the agencies overlap, powerful influences may arise to create divergences of interest within the government. Differing procedures introduce differing values and different restraints; the emergence of an “institutional interest,” the development of professionalism, the influence of colleagues and traditions, all provide the possibility, at least, of internal checks. Separate agencies, composed of distinct bodies of men even where functions are shared can be made representative of different groups in the community, and so, as with bicameral legislatures, provide the basis of a check upon the activities of each of them.
The second element in the doctrine is the assertion that there are three specific “functions” of government. Unlike the first element, which recommends that there should be three branches of government, this second part of the doctrine asserts a sociological truth or “law,” that there are in all governmental situations three necessary functions to be performed, whether or not they are in fact all performed by one person or group, or whether there is a division of these functions among two or more agencies of government. All government acts, it is claimed, can be classified as an exercise of the legislative, executive, or judicial functions. The recommendation then follows that each of these functions should be entrusted solely to the appropriate, or “proper,” branch of the government. This view of the “functions” of government is extremely abstract, and some of the attempts to justify this threefold division have reached a very high degree of abstraction indeed. It must be distinguished from the very different view of the functions of government which enumerates them as, for example, the duty of keeping the peace, of building roads, or of providing for defence. These we might label the “tasks” of government in order to distinguish them from the more abstract notion of function. In the period before Locke and Montesquieu firmly established this abstract view of the functions of government there were two main streams of thought, in one of which the word “power” was used to describe the function of legislating, or executing the law, and in the other a more practical view was taken of the multiplicity of government acts by dividing up the “attributes of sovereignty” into six, seven, or more categories, which included, as well as making laws, such tasks as the control of the coinage, or the appointment of standard weights and measures.8 The triumph of the more abstract conception of the “powers”