The risk of parasitism as communities grow brings us to our fourth and perhaps most important modeling choice. We considered that the implementation would only be done through a decentralized enforcement mechanism. Environments where the application is decentralized are not difficult to find, especially before the emergence of the nation-state (Hadfield & Weingast 2011a). But the problem of parasitism may well be a major reason for consolidating state enforcement into a centrally controlled authority with a monopoly on legitimate coercive power. Here, however, our model suggests a fascinating hypothesis. We have shown a correlation between normative characteristics often associated with the desirable attributes of governance by law, not by men, and the problem of coordination and incentives for collective participation in punishment. An institution that hopes to achieve an effective legal order in this environment must ensure that its system is general, open, stable, impersonal, etc. This suggests the possibility that a regime relying on centralized coercive violence may not be restricted in the same way. We ask ourselves: does the move to centralised enforcement go hand in hand with a derogation from the rule of law? In other words, can any system that relies solely on centralized enforcement be classified as a legal system? Or does it tend (or necessarily) to turn into a Tyrranian or dictatorial order? We suspect that any order we wish to identify as legal must be based on decentralized enforcement, at least to some extent and perhaps to a large extent. This could be true because a regime that depends solely on centralized punishment must spend exponentially growing resources to administer a system of detection and punishment (exponentially, because delegating these tasks to state employees requires the application of the rules for those performers) or rely on extraordinary and disproportionate penalties to compensate only for probabilistic recognition (Becker 1968). Our model suggests an additional reason to expect that the use of exclusively centralized enforcement may be incompatible with the legal system: by relaxing the incentive restriction, a centralized law enforcement system is free to apply rules indistinguishable from dictatorial fiat.
A second difference between our approach and the existing literature is that, with the exception of Basu (2000) and McAdams (2005), the existing coordination reports of the law focus on the coordination problem faced by officers with primary behavior: the choice of the side of the road they want to drive on, If they claim disputed property or apply a conventional interpretation of a law, for example.15 In our model, on the other hand, the problem of coordination is a problem faced by agents who can punish primary behavior: responding to those who drive on the wrong side, taking what is not theirs, or adopting unconventional legal readings. That is to say, we focus on how the characteristics of the legal rules that govern primary conduct affect the coordination problem faced by the authorities responsible for enforcing those rules. It also makes our approach much more general than existing accounts. As McAdams (2000) notes carefully, the expressive representation of law (Sunstein 1996) is a partial representation of law that applies only to attitudes in which primary behavior is characterized by an overall incentive for coordination; This is an environment where no sanctions are required to enforce a legal standard. In our presentation, we start from the much more ordinary environment in which a legal regulation punishes a certain behavior and on this basis channels the behavior towards compliance. Our emphasis on decentralized collective application also highlights a feature of the legal system that is not so often emphasized, namely the role of an authoritative steward of a common logic that coordinates punishment by providing a single classification system. We show that these features of the legal system can serve to address the two main problems faced by a community seeking to deter misconduct through decentralized collective punishment: they help coordinate punishment decisions and create the incentive to bear the personal costs associated with the punishment that benefits the larger group. Our positive analysis thus adds a new dimension to our understanding of the normative characteristics of the legal system.
Most of the existing literature in legal theory deals with normative representations of these normative characteristics. Public courts, impersonal thinking, and universality, for example, are often understood as moral or political boundaries imposed by moral or political theory on the exercise of power by (especially democratic) governments.
