legal theory: legal process: peter strauss

Peter L. Strauss, Formal and Functional Approaches to Separation-of Powers Questions

72 Cornell L. Rev. 488, 492-94 (1987)(footnotes omitted).


[As an introduction to the modern debates about constitutional separation of powers, consider this statement of the non-formalist position.]

Readers are entitled to know about their guide that he has already staked out [in prior articles here recapitulated] a position in these debates, opting for the difficult middle ground of functionalism. That analysis, which will not be repeated here, sought to show that our formal, three-branch theory of government - at least as traditionally expressed - cannot describe the government we long have had, is not required by the Constitution, and is not necessary to preserve the very real and desirable benefits of "separation of powers" that form so fundamental an element of our constitutional scheme. The Constitution does not define the administrative, as distinct from the political, organs of the federal government; it leaves that entirely to Congress. What the Constitution describes instead are three generalist national institutions (Congress, President, and Supreme Court) which, together with the states, serve as the principal heads of political and legal authority. Each of these three generalist institutions serves as the ultimate authority for a distinctive governmental authority-type (legislative, executive, or judicial). Each may be thought of as having a paradigmatic relationship, characterized by that authority-type, with the working government that Congress creates.

Although these heads of government serve distinct functions, employing distinctive procedures, the analysis argued, the same cannot be said of the administrative level of government. Virtually every part of the government Congress has created - the Department of Agriculture as well as the Securities and Exchange Commission - exercises all three of the governmental functions the Constitution so carefully allocates among Congress, President, and Court. These agencies adopt rules having the shape and impact of statutes, mold governmental policy through enforcement decisions and other initiatives, and decide cases in ways that determine the rights of private parties. If in 1787 such a merger of function was unthinkable, in 1987 it is unavoidable given Congress's need to delegate at some level the making of policy for a complex and inter-dependent economy, and the equal incapacity (and undesirability) of the courts to resolve all matters appropriately characterized as involving "adjudication." A formal theory of separation of powers that says these functions cannot be joined is unworkable; that being so, a theory that locates each agency "in" one or another of the three conventional "branches" of American government, according to its activities, fares no better. Respect for "framers' intent" is only workable in the context of the actual present, and may require some selectivity in just what it is we choose to respect - the open-ended text, the indeterminacy of governmental form, the vision of a changing future, and the general purpose to avoid tyrannical government, rather than a particular three-part model. The problem is finding a way of maintaining the connection between each of the generalist institutions and the paradigmatic function which it alone is empowered to serve, while retaining a grasp on government as a whole that respects our commitments to the control of law.

This object can be achieved conformably to the words of the Constitution, it was claimed, although at some cost to traditional understandings, by observing that the concept of a "branch," as such, is not required by the text. When the Constitution confers power, it confers power on the three generalist political heads of authority, not on branches as such. The constitutional text addresses the powers only of the elected members of Congress, of the President as an individual, and of the Supreme Court and such inferior federal courts as Congress might choose to establish. Its silence about the shape of the inevitable, actual government was a product both of drafting compromises and of the explicit purpose to leave Congress free to make whatever arrangements it deemed "necessary and proper" for the detailed pursuit of government purposes. One can easily and properly infer some relationships that the three named governmental actors must observe as among themselves and, consequently, with whatever subordinate parts of government Congress chooses to create, without having to believe that those parts must be located "here" or "there" in the government structure, or that the governmental functions they may perform are restricted by the accident of that location.

Rather than describe agencies in terms of branches, in other words, the analysis suggested one could examine their relationships with each of the three named heads of government, to see whether those relationships undermine the intended distribution of authority among those three. Just as contemporary property analysis tends to speak of the property relationship as a "bundle of rights," no one of which is essential to the characterization of a particular bundle as having the attributes of "property," this analysis of separation-of-powers issues proposes examining the quality of relationships between an agency and each of the three named heads of government. It is not necessary to insist that there be particular relationships between an agency and any of the named constitutional actors (beyond the few specified in the constitutional text) in order to require relationships of a certain overall character or quality.