What do we mean by discretion?

The nature of judicial discretion is something else which has occupied much academic writing. It is a word used very frequently when debating the actions of judges and is, in many senses, ubiquitous in the operation of the legal system. The problem (perhaps) with the word, however, is that it spans a number of various meanings which describe very different activities. As a consequence the use of the word discretion without further explanation has the potential to obscure rather than elucidate what is actually going on. What does it mean to say that judges are making a discretionary decision under s 33?

For Dworkin there were three different types of discretion: 21

‘First, we say that a man has discretion if his duty is defined by standards that reasonable men can interpret in different ways…Second, we say that a man has discretion if his decision is final, in the sense that no higher authority may review and set aside that decision…Third, we say a man has discretion when some set of standards which impose duties upon him do not in fact purport to impose any duty as to a particular decision…’

The first two of these Dworkin considered to be ‘weak’ discretion, the third he described as ‘strong’ discretion. 22 Dworkin was relatively uninterested in the two forms of weak discretion. Rather his concern lay in what we might call the developmental strong discretion, whereby judges are determining whether to develop the law in one direction or another. So, for example, when faced with the decision whether or not to develop the equitable notion of breach of confidence into a tort of misuse of private information, the House of Lords in Campbell v Mirror Group Newspapers Ltd 23 was, presumably, exercising the kind of strong discretion which interested (and concerned) Dworkin. We probably do not need to explore further Dworkin’s views on this type of discretion because that does not seem to be what is going on in the s 33 cases. Rather, judges deciding s 33 disputes seem to be exercising weak discretion in the first sense of Dworkin’s taxonomy. That is to say, they are applying standards which can be interpreted in different ways by reasonable people. This ‘discretion by judgment’ might be styled as a ‘someone has to decide’ discretion. This is the most commonly exercised discretion in the legal system. This is the kind of discretion that judges exercise when operating as fact finders. If the claimant says that what happened was X, while the defendant says that what happened was Y, then someone has to decide which of those two is correct. Judges are granted that role. Provided that both accounts have some element of plausibility a judge is making a discretionary choice to accept one version over another, based on a myriad of factors, including consistency (both internal and also in relation to other available evidence) but also including such highly subjective elements as the demeanour of witnesses. There is no directed outcome. If we were to put the same evidence in front of several different judges it is likely that there would be different outcomes. This type of discretion is not developmental in that it does not change the law, it merely determines the result. As a consequence it has had more limited academic attention. It is, however, not merely frequent, but also, often, determinative, because there are many cases where the dispute of fact is the only issue, and the findings of fact by the trial judge determine the outcome.

This kind of weak ‘someone has to decide’ discretion is, therefore, problematic in terms of users of the legal system because it can give rise to high levels of unpredictability of outcome. This is, however, inherent in the nature of an adjudicative system. Whenever a dispute exists on matters of fact it will always be true that someone has to decide. While s 33 disputes are not disputes of fact, they fall into a similar ‘someone has to decide’ scenario. The fact that such discretion exists in, and is integral to, the operation of the system may have relevance for the fourth measure against which we may need to judge s 33 discretion – that is certainty of outcome. While we may be concerned about the predictability deficit of s 33, we do need to bear in mind that s 33 is not turning a highly predictable system into a highly unpredictable one – it is merely adding a layer of unpredictability to a system for which some levels of unpredictability are already inherent.

This is not to say that the reduction of the predictability deficit to the lowest level achievable consistently with seeking to achieve other aims is not desirable. Satisfying this requirement while continuing to allow discretionary decisions requires a consideration of the principles and policies within which the judge is exercising this discretion. If we want to ensure that we have a system which reduces the predictability deficit as far as possible then we must consider what constraints exist on the way judges carry out this task. If there are no such constraints, or they are insufficiently directive, then s 33 decisions do risk becoming a series of separate and isolated outcomes. For the most part, the more varied and complex these constraints are, the greater the likely predictability deficit.

In relation to s 33 there are two possible sources of such restricting principles and policies. There is the statute itself and there is the interpretation of the statute imposed by the higher courts.