A Brief Introduction of the Puzzle of Discretion

Discretion arises when an official is empowered to exercise public authority and afforded scope to decide how that authority should be exercised in particular circumstances. At root, discretion is about power and judgment. Not surprisingly, discretion is the focus of countless law and policy initiatives designed to preserve, enhance, check, limit, shape, or eliminate discretion in a host of different settings. The articles in this special issue of CJLS/ RCDS canvass settings as diverse as prosecutors, border guards, universities, and immigration decision makers. It generates heated political debate, excited social commentary, and intensive scholarly explorations. The question of discretion is both compelling and confounding. As early as 1935, John Willis famously urged scholars to attend to "what actually happens" in the context of administrative decision making. However, studying "what actually happens" is often harder than it sounds, and theorizing discretion in ways that move beyond a binary understanding of law/discretion is a daunting challenge. We believe this challenge can and should be addressed through interdisciplinary insights and dialogue.

"Discretion, like the hole in a doughnut, does not exist except as an area left open by a surrounding belt of restriction."¹ Ronald Dworkin's oft-cited "doughnut analogy" neatly encapsulates the conventional view of discretion. Three main assumptions are embedded in this view: that law is the primary instrument of social regulation, that discretion is a residual category of law, and that this discretion is exercised by individuals who, though influenced in a wide variety of ways, are essentially autonomous. While recent scholarly analyses of discretion have begun to revisit and challenge this conventional view, its core assumptions nonetheless continue to underpin political discussions, policy debates, and judicial case law to such an extent that the conventional view of discretion sets the parameters of imagined policy and legislative reforms related to its use. To introduce this CJLS/RCDS special issue on discretion, we begin with a brief, critical review of the academic literature on discretion.² Discretion has long been the focus of scholarly and policy-related discussion and analysis in the field of criminal justice, and some of the conventional concern for discretion as a "problem" is rooted in the literature on police, prosecutorial, and judicial discretion over criminal law enforcement.

What we have termed the conventional paradigm of discretion is captured by the observation that where the law ends, discretion begins: whereas the rule of law is seen as intimately connected to notions of justice, discretion is more open-ended. As Philip Anisman observed in the first significant attempt to catalogue discretionary power in Canada, the exercise of discretion does not in and of itself suggest any particular value; depending on the context and the actions of a discretionary decision maker, discretion may mean beneficence or tyranny, justice or injustice, reasonableness or unreasonableness.³

A standard dictionary defines discretion as "the power or right to decide or act, according to one's own judgement or choice."⁴ In the context of the administration of public law and policy, the "freedom to choose" of autonomous decision makers is defined in natural opposition to the constraints imposed by legal rules. Indeed, the same dictionary defines "arbitrary" (arbitrariness being the most prevalent concern associated with discretionary decision making) as "subject only to individual will or judgement, without restriction; contingent only on one's discretion: an arbitrary decision . . . having unlimited power; uncontrolled or unrestricted by law; despotic; tyrannical." ⁵ The idea of discretion is thus heavily inflected with liberal assumptions and ideals relating to the power of law, autonomy, and freedom of choice. Whether discretion is regarded benevolently or critically, its essential and inextricable binary relationship to law is largely taken for granted.

This legal construct assumes, in its most extreme form, that law and discretion are discrete and distinct entities that are negatively correlated: more law means less discretion, and less discretion means more law. The policy implications of this view of discretion are significant. The view that the only meaningful constraint on discretion is law leads to a primary analytical focus on the "surrounding belt of restrictions" rather than on the "hole" of...