In my previous post (which few people responded to*) I described Hart’s theory of law, gave some criticisms and gave my own take on things, which all things considered is still a legal positivist account of law. From this post on, I will be discussing Natural Law theories. As somewhat of a legal positivist, I will try to give criticisms of these theories where appropriate and try to show either where they go wrong or where, even if accurate, they fail to disprove legal positivism.
Today, we talk about Ronald Dworkin’s theory of adjudication. Dworkin was a student of Hart and his theory is very much a response and modification of Hart’s theory. To recap:
A law is any rule that is part of a legal system. A legal system is one that consists of the union of primary and secondary rules. Primary rules are our familiar coercive rules, the paradigmatic case being those rules found in the body of criminal law. Secondary rules are rules that pertain to primary rules either immediately or through other secondary rules. The secondary rules include private and public power conferring rules, rules of recognition and rules of succession among others. [emphasis added]
Legal Principles vs rules
Dworkin takes aim at the rules of recognition. Dworkin’s criticism of rules of recognition takes place at two levels. At one level, he doesn’t think that any general theory of law is possible. The reason for this is that he feels that in at least some jurisdictions, no single set of recognition rules is adequate to describe where all the settled law comes from. Dworkin therefore wants to eschew general theories of law like Austin’s and Hart’s. Instead, Dworkin narrows the scope of what he is doing and develops a theory of what American judges think they are doing and should be doing. Dworkin argues that when judges reach a decision for a case, they see themselves as firstly, finding the law not only from particular rules, but also from principles and secondly, reaching their decision from within the bounds of the law and not going beyond it.
Dworkin says that there is a logical/conceptual difference between rules and principles. Rules are absolute. Wherever a rule is relevant, it is decisive. So, if there is a rule about mandatory death sentences for murder, whenever a person is guilty of murder, the death penalty applies. Principles, on the other hand, are not absolute. Principles instead carry weight. So, even if there were cases where a principle was relevant, they may not necessarily be decisive.
Dworkin’s basic objection to legal positivism is that legal principles either are moral principles, originate from moral principles or have a moral character. Whereas the legal status of rules depends on its pedigree, Dworkin argues that pedigree is insufficient to make a principle a legal principle; its content plays a part as well. This, Dworkin says, is akin to moral principles whose status as moral principles is wholly determined by content. Dworkin says that even after satisfying pedigree tests, the law still needs interpretation. However, interpretation is always value laden. Therefore interpretation always involves moral values. But this is not all. Dworkin not only argues that judges resort to principles with some moral character, he says that when they do this, they do not go outside the law. If they went outside the law whenever they resorted to moral principles, then, legal positivism will still be true. However, if it is the case that judges go outside the law when they search for legal principles, then they are creating law and not discovering the law. Dworkin says that since judges do not see themselves as creating law, when they search for moral principles, they must be doing so from within the bounds of law.
So, how does this work? In Hart’s model, you take all the cases of settled law and from there derive rules of recognition that would have the greatest fit with the settled rules of law. If given the rules of recognition, there are still a few primary rules that do not fit, that provides the judge with grounds to modify those settled cases. The judge therefore goes back and forth from settled primary rules to rules of recognition until both are in reflective equilibrium with each other. For Dworkin, fit is not enough. Dworkin also envisages principles which not only have a large degree of fit, but which best justify the settled primary rules. So even if one set of recognition rules and principles had maximal fit with respect to the primary rules, if the content of the principles were insufficiently morally justified, then an alternative set of recognition rules and principles ought to be used, provided that the level of fit passed a certain threshold. Dworkin did not specify exactly what the threshold was. So, principles which had very good moral content, but did not fit very well with most of the settled law would not be part of the recognition principles.
Let’s work backwards with this shall we?
1. Dworkin’s theory doesn’t seem to be able to cope with the existence of wicked legal systems. If large portions of the primary rules are cruel, draconian and discriminatory (e.g. the American slaveholding south) then there in what sense would any set of principles which had the requisite level of fit morally justify that which cannot be morally justified?
2. So, by Dworkin’s own lights any legal principle must have institutional support. i.e. such principles are not plucked out of the air, but are often found in past decisions etc. Choice of principles is often determined by context. If Dworkin was right, then there would be cases where the context called for the use of a principle, but the principle was rejected because of its moral content. So here’s a question for the lawyers in the League: Have any legal principles fallen into disuse simply because of their moral content?
3. Even if it is the case that in the US, legal validity is in part determined by moral considerations, this need not and probably isn’t true of at least some other legal systems. Legal positivism is a claim about whether moral considerations have a necessary connection to legal validity.
4. The distinction that Dworkin draws between rules and principles is overwrought. To take an analogy from Austrian Economics, any utility function using a set of cardinal values can be transformed into a lexical ordering of outcomes with ties. Similarly, any set of principles can be transformed into some set of infinitely specifiable rules. Given finite space, infinitely specifiable rules can be roughly summarised giving them an incomplete, vague or open texture such that they function the same way as principles.
5. Even if 4) is false, it is no great modification to Hart’s theory to say that there are secondary rules and principles. All we have to do is deny that the principles have any necessary moral character (which we have done. see points 1-3)
6. Does the law always require interpretation? And does interpretation always require a value judgement? Are such values judgements always moral in nature? It seems that the answer to either question depends on what notion of interpretation one uses. If we have a thin notion of interpretation, then rather trivially, law always requires interpretation. However, with this thin notion of interpretation, interpretation need not require a value judgement or at least a moral one all the time. However, if we have thicker notions of interpretation, then it may be the case that interpretation always requires a value judgement and moral considerations may even always apply. However, under this definition, easy cases do not require interpretation while the hard cases do.
If positivism is true and the law really has this open texture, where the extant rules underspecify whether a particular extension of a rule is legally valid, then especially in hard cases, judges always go outside the law to reach their decision. This is unavoidable. Complaints about judicial over-reach are either mis-placed or have to be re-formulated. (Although they could be valid if judges blatantly ignore relevant precedence in easy cases) However, if Dworkin is right and the law is semantically complete, then accusations of judicial over-reach in hard cases will apply only if a judge completely ignores the relevant case-law such that the level of fit falls below the threshold. However this happens so rarely that most accusations of judicial over-reach are at the very least problematic, if not groundless. Here’s another question for the League: In light of Hart’s and Dworkin’s theories, how do we make sense of the idea of judicial over-reach and the ideal that judges not over-reach?
* Yes, this is me officially being huffy about so little response to a series I assumed a lot of the guys would be interested in and about a post that has taken me lots of effort to construct.