2.1The rule as design
Adler and Borys distinguish two kinds of formal rule. Some rules enable: they help users master the work, show their reasons and leave room for repair. Others coerce: they force compliance and hide their reasons. The difference, they argue, lies in how the rules are designed (Adler and Borys 1996). The distinction sorts rules by how they are made. The Dao De Jing, read in Section 4, counts them instead, and the contrast between sorting and counting runs through this article.
Fuller makes a parallel argument from the side of law. Law has an inner morality of eight requirements, which Fuller tells through a king, Rex, who fails at each in turn. There must be rules. They must be published, prospective, clear, consistent with one another, possible to obey and not changed too often, and official action must agree with the rules as declared (Fuller 1964). Several of these requirements bear on subtraction without naming it. A rule that cannot be obeyed fails one of them; so does a rule changed too often; so does official action that departs from the declared rule.
Hart's account of a legal system belongs on the same side. A system is a union of primary rules, which impose obligations, and secondary rules about those rules: a rule of recognition, rules of change and rules of adjudication. A regime of primary rules alone would be uncertain, static and inefficient, and the rules of change, which empower people to introduce new rules and remove old ones, remedy its static quality (Hart 1961). Removal, on Hart's account, is itself governed by rules.
Grouped in this way, the three positions make a single side: the place to work on a rule is its form.
2.2The rule as residue and drift
The other side begins from the rules that already exist. Lindblom argues that policy is made by successive limited comparisons, small steps from what exists, rather than by a comprehensive choice among all options (Lindblom 1959). Applied to rules, the starting point is the rule in hand, and the next rule is a small move from it.
Streeck and Thelen describe five forms of gradual institutional change: displacement, layering, drift, conversion and exhaustion. In drift, an institution is left formally unchanged while its circumstances change, so that its effects change (Streeck and Thelen 2005). This article draws from that description the inference that, in drift, the rule's wording stays as it was while what it does changes.
Kaufman, writing of the rules and procedures of the United States federal government, describes red tape as mostly the residue of protections that people demanded. One person's red tape is another's safeguard, and so cutting it means deciding whose protection to drop (Kaufman 1977). Bozeman defines red tape as rules that remain in force and impose a burden of compliance without serving the legitimate purposes they were meant to serve. Some are dysfunctional from the start, which he calls rule-inception red tape; others become so as circumstances change, which he calls rule-evolved red tape; and whether a rule is red tape depends on whose purposes are counted (Bozeman 2000).
Grouped this way, the second side begins from rules already in place rather than from rules to be designed. Rules accumulate, and they change under their own wording.
2.3Controls that cause harm
A third set of works speaks most directly to the Dao De Jing's concern, though from a distance. Perrow argues that in systems whose parts interact in unexpected ways and are tightly coupled, accidents will come in time however well the systems are run. Buffers and redundancies have to be designed in on purpose in such systems, but safety devices and redundancy add complexity and can cause accidents themselves (Perrow 1984). Scott argues that states make society legible by simplifying it, with standard names, maps and measures, and that high-modernist schemes that impose such simplifications and override local, practical knowledge have failed, sometimes disastrously (Scott 1998).
Perrow's point concerns safety devices in tightly coupled systems, and Scott's concerns schemes imposed from above that override what people on the ground know; neither is a general claim about rules. Both nonetheless describe a harm that arrives with the thing added.
2.4The two sides together, and what neither takes as its subject
| Design side | Residue and drift side | |
|---|---|---|
| Unit of analysis | The rule to be made well (Adler and Borys; Fuller) | The rule in hand, and the next small step from it (Lindblom) |
| How rules go wrong | They coerce, hide reasons, cannot be obeyed, change too often (Adler and Borys; Fuller) | They outlive their purpose or are left unchanged while circumstances move (Bozeman; Streeck and Thelen) |
| What removal is | An act governed by rules of change (Hart) | A decision about whose protection to drop (Kaufman) |
| Where added control harms | Not its main subject | Safety devices that add complexity (Perrow); imposed simplifications (Scott) |
Read across, the table shows a gap. Both sides have a place for removing a rule; neither treats not adding, or withdrawing, as a restraint in its own right, an act that holds back what would otherwise come. Drift, as described, concerns a rule whose wording stays the same while its circumstances, and so its effects, change; it is not an account of withdrawing anything. Kaufman's cutting is a decision, but it is described by what it drops. The Dao De Jing describes not adding as a way of governing, and this article puts it into that gap.