The presumption of innocence (PoI) is considered to be a fundamental principle of criminal law. Over the past decades, however, the emphasis on the rights of suspects and defendants has given way to a more instrumental view of criminal law as a means to reduce risk and attain safety. One can think, for example, of recent Acts on Terrorism that do not require ‘suspicion’ but merely ‘indications’ of a terrorist crime, thereby lowering the level of suspicion required for investigative activities in the pre-trial phase; of plea-bargaining that has been introduced in various legal systems; of the verdict of the European Court of Human Rights (ECtHR) in Salabiaku v. France where the Court allows for presumptions of fact or of law to operate in the trial phase; of the possibility of review after wrongful acquittals and of convicts who face special measures after serving their sentence in the post-trial phase, to name just a few examples. Such an instrumental approach puts fundamental principles such as nulla poena, ne bis in idem, nemo tenetur, in dubio pro reo, nullum crimen sine culpa, as well as the PoI under pressure.

These developments make reflection on the PoI an urgent affair. Reflection is even more important because the principle seems to have such an elusive nature. What exactly is the PoI? Notice how wide-ranging the above examples are. Are they all connected to the, or rather a, PoI? Or should we restrict application of the PoI to the defendant in the trial phase and let other principles do the job in pre- and post-trial phases, where citizen and suspect, and acquitted and ex-offender, are the key players? In other words: is a broad interpretation of the PoI that has a bearing on all these examples desirable? And how is the PoI actually interpreted and applied in legal practice? These questions are at the heart of this special issue.

Antony Duff, a leading expert in criminal law theory and the philosophy of punishment, provides the main contribution to this issue. A central theme within Duff’s philosophy is the idea that trial and punishment should be understood communicatively. When the prosecution has a prima facie case that someone did what he is accused of having done, the defendant should answer by providing counter-evidence or a justification or an excuse; if he cannot provide a rebuttal, he should accept that punishment is his due. Therefore the defendant is required to appear in person and expected, though not forced, to enter a plea. The punishment, in turn, communicates that what the offender did is wrong and it asks him to repent, or at least to pay his debt, for his criminal actions.

Duff utilizes the idea of communication to account for some salient features of the practice of criminal law. For example, trial and punishment only make sense in relation to a reason-responsive agent; the accused should thus be fit to plead. The communicative approach also makes clear the importance of the principle of proportionality. If punishment is to be an expression of the wrongness of the offender’s actions, it has to be proportionate to the seriousness of the crime and responsibility for that crime. Finally, the PoI too makes sense within the idea of communication. To ask the defendant to respond to allegations is only fitting if the prosecution has first shown that there is a reasonable suspicion that he indeed did what the prosecution claims he has done.

In his present contribution, Duff is concerned with the question whether the (or a) PoI has a role to play beyond the confines of the criminal trial. He answers the question in the affirmative. Although Duff explicitly states that his account flows from common law systems, he is not so much interested in the meaning of the PoI in particular legal systems, but rather provides us with an analysis that purports to show that the PoI, or rather many presumptions, can be discerned in a wide range of contexts. The PoI plays a role not only during the trial, but also prior to the trial as well as once a sentence has been served or where a defendant is acquitted. According to Duff, a modest kind of civic trust that entails that it is only appropriate for a citizen to be burdened with the role of defendant, i.e., to be put on trial and asked to answer, if there is sufficient evidence of his guilt, is basic to citizenship. Duff utilizes a broad interpretation of the PoI, combined with the idea of civic trust, to criticize, inter alia, selectively applied precautions, pre-trial detention and post-punishment preventive measures.

The five other articles in this special issue respond in different ways to Duff’s analysis.

Thomas Weigend and Magnus Ulväng offer an analysis of the nature of the PoI and subsequently question the idea that a broad PoI is warranted and necessary. Weigend’s main concern is that a broad PoI lacks a precise meaning. The function of the strict PoI is to restrict the powers of the state. Postulating a broad principle of civic trust seems to be empirically unwarranted, because many people actually break the law. Moreover, such a principle is redundant as civil liberties can do the job that Duff intends a broad conception of PoI to fulfil. Magnus Ulväng first analyses the nature of the PoI and argues that it has less weight than Duff seems to assume. It is not a rule of obligation or a principle, but rather a rule of thumb. Subsequently, Ulväng too argues that we do not need a broad PoI; rather, many of the purposes of the PoI can be served by a principle of fairness that already applies within substantive criminal law.

Both Geert Knigge and Lonneke Stevens criticize Duff, arguing that his analysis is not consistent with legal practice, thus challenging Duff’s claim that his account flows from positive law. Knigge agrees with Duff that many different presumptions with different rationales underlie the verdicts of the ECtHR, but then criticizes Duff’s claim that a principle of civic trust unifies them. According to Knigge, if anything, respect for dignity rather than a principle of civic trust underlies the verdicts of the ECtHR. Lonneke Stevens discusses the presumption of innocence in relation to the Dutch practice of pre-trial detention. Empirical research shows that judges use pre-trial detention as an instrument to realize incapacitation and pre-punishment. Thus, as far as pre-trial detention is concerned, Duff’s PoI is hardly more than an ideal that has only limited normative effect in legal practice.

Alwin van Dijk is the only author who adopts and even broadens Duff’s already broad interpretation of the PoI. According to Van Dijk, the important discussion is that which questions whether interferences with the PoI are justifiable and, if so, on what grounds. Subsequently, Van Dijk argues that although it is often thought that most PoI interferences are justified on consequentialist grounds, the retributivist anti-PoI duty to punish the guilty might be the worst enemy of innocents.

In his reply to his critics, Duff further elucidates his reasons for talking of different – moral rather than legal – presumptions of innocence outside the criminal trial. He argues that even if our aim is only to understand the PoI as it figures in our current legal practice best rationalized we cannot restrict ourselves to legal theory. We must take political theory into account and deal with the proper aims of the state and the relationship between state and citizen. Central to this relationship are the questions of when and how the role-bearer’s innocence is challenged or denied.

At the moment there is a broad interest in the PoI, and rightfully so. We are therefore very pleased to present a challenging interpretation of the PoI and five critical reactions to it. We hope that the articles in this issue can function as a thought-provoking contribution to the continuing debate about one of the most fundamental principles of our criminal law systems.