Professor Andrew Simester is Dean of National University of Singapore’s Faculty of Law, and Amaladass Professor of Criminal Justice. A world-leading researcher in criminal law and theory, Prof Simester previously taught at King’s College, London, the University of Cambridge, and the University of Nottingham. His latest book, Fundamentals of Criminal Law (2021), has been the subject of multiple international conferences, including a forthcoming special issue of Criminal Law and Philosophy.

Dr Mark Dsouza is Associate Professor at University College London’s Faculty of Laws, and has previously taught at the Liverpool Law School. His monograph, Rationale-Based Defences in Criminal Law was published in 2017, and he won UCL’s EXALT award for Excellence in Law Teaching in 2019, and the SLS Best Paper award in 2020 for his article ‘Beyond acts and omissions: remark-able criminal conduct’’.

Mark Dsouza: Hello, Andrew. I want to start by asking a little bit about what brought you into the field that you work in: criminal law theory and doctrinal criminal law in various different jurisdictions. That's quite a wide scope of work that you do. So, what was it that sparked your interest in criminal law theory, and in doctrine across so many different jurisdictions?

Andrew Simester: It's a combination of factors. My original law degree was primarily—almost entirely—in commercial law subjects like tax and corporate finance, commercial law, and similar kinds of subjects. I had assumed, when I went into law school, that I was going to end up in practice, maybe as a barrister. But when I was studying legal theory, I had a teacher by the name of Jim Evans, who's retired now. He was my first jurisprudence teacher, and we started talking about the problems of intention and voluntariness, and I just felt they were really genuinely interesting in themselves. I thought that I might like to think further about those kinds of problems and work out what made something intentional—what was it that marked an action out as intentional. Jim suggested that I do a PhD. And then he said, ‘Well, if you're going to do a PhD, you should go to Oxford and do it under Joseph Raz’. And his point, which I've since made to other would-be Phd students was, ‘You go with the supervisor; not so much the institution.’ Joseph was of course, exceptional as a supervisor, and an exceptionally clever person. He hadn’t supervised in criminal theory areas before, or, I think, since! But it didn't matter. He was just so good that he could take students from pretty much any field. Through the sheer force of his intellect, he would be able to dissect their papers and help make them better.

So my interest in criminal theory started with Jim, and then Joseph was the one who really sharpened things up. He might say, for example, ‘just trying to analyse concepts like intention won’t do in itself. You need to think about the nature of analysis and how to give an account of something. Do you want to give an atomistic account: These are the sub-components of, say, intention’ – and I started out thinking like that. But then he said, ‘It doesn't always work like that.’ Sometimes, if you have a concept like intention, it's going to be fuzzy in the edges. And what that means is that the content of your analysis should also contain components that are fuzzy in places where the concept itself is fuzzy. Only then is it a good account. If your account gives you an artificial sharpness, it's not a good analysis. Or it's possible that the components themselves are reasonably sharp, but they're in an unusual combination, and that explains why those cases are borderline. All this is just an example. My point is that what Joseph was doing was teaching me how to think and analyse. In doing so, he took me outside the relatively narrow kind of topic that I thought I was going to research. Thus I ended up writing a thesis that was much broader. I talked about a variety of different problems in criminal law and theory – not just intention and voluntariness, but also justification, mistakes, negligence, and so on. Ultimately, the thesis ended up laying some of the groundwork for the book I published in 2021: Fundamentals of Criminal Law.

At the same time, I was always also a lawyer, and so I remained interested in criminal law as a set of doctrines. For some reason that I don’t recall, I was back visiting New Zealand at some point towards the end of my doctorate, and I realized that New Zealand still didn't have a criminal law textbook. When we studied criminal law when I was an undergraduate, we used the 5th edition of Smith and Hogan. New Zealand does have a common law background, but its criminal law is a code, the Stephens code. Smith & Hogan was therefore wildly inappropriate as a textbook. It was fine on general doctrines like causation, but it was unsuitable as a text for doctrinal New Zealand criminal law. And so I talked to my own former teacher, Warren Brookbanks. I suggested that maybe we should do something about this. And that's how I got into writing criminal law treatises. Later, having done the New Zealand book, Bob (G.R.) Sullivan and I thought we might do something similar on English law because that’s where I was working. So, this broadening out to different jurisdictions is a little bit of happenstance, reflecting where I was at the time. In turn, when I moved to Singapore, it found that it has a different code again. As you will know, it has the Macaulay code from India, which is the most common code in the world. So more recently, I’ve been thinking about that Code, and writing a bit about that. At some point, when I cease to be the Dean at NUS, one of my projects is to write a textbook for Singapore.

Basically, coming into thinking about the theory was a more orthodox path, guided by teachers who got me interested in problems and then helped teach me how to think about those problems. But the breadth, in terms of jurisdictions, reflects the fact that I've lived in different countries. One of the key things about living in more than one country is that you realize there are different ways of doing things, and that enriches the way you see what is done in the jurisdiction you live in.

M.D.: You've now had a chance to work with at least 3 jurisdictions in great detail. And you mentioned initially that the Smith & Hogan textbook is very good for the general part of the criminal law, but not so much the special part in a different jurisdiction. So, have you noticed any interesting deviations, in either the general part or the special part of the criminal laws in these different jurisdictions, over the years things since you started teaching?

A.S.: The most interesting thing about teaching in New Zealand and Singapore is that their criminal codes are very dated. The Indian Penal Code was first drafted in 1837, while the New Zealand code was based upon Fitzjames Stephens draft criminal code, which is from much later in the 19th century. That makes teaching them very interesting, particularly the Macaulay code because it's from so long ago—it's almost 200 years old in terms of its first drafting. It was drafted against the backdrop of a different set of understandings of things like responsibility that are no longer part of the common law. The common law has evolved. But the code got stuck. And that's a feature of codes. I always think it’s somewhat surprising that Europeans boast of the fact that their law is ossified in a code, with the then understandings of, for example, insanity or complicity. Our understandings of things like responsibility have matured: they are very different now from what they were in the early 19th century, when the Macaulay Code was framed. The difficulty is that, if you have a code that's framed against that backdrop, it is difficult to get out of that historical understanding. And so is becomes hard to modernize the criminal law, short of statutory reform, which is very difficult to achieve in the criminal-law context.

For reasons like this, my own teaching has been a really educational experience, making me think about the ways in which the common law has changed, and why. Not all evolutions in the common law are good, but there is a tendency for the common law to oscillate around where it should be, and over time it tends to converge on better positions. For example, common law was originally very much a ‘strict liability’ phenomenon, and then it evolved increasingly sophisticated concepts of mens rea, thanks to the great judges like Coke. And this progress, in the long run, has improved the common-law approach to crime.

M.D.: It's funny that you say that it's harder to do that when you have a concretised code, and you give examples of insanity, which the M’Naghten rules have ossified. A continental criminal lawyer might say to you that, ‘Given that we have one way in which law evolves, and we don't have the possibility of Parliament passing the buck to the courts, and courts passing the buck back to Parliament, we actually are more nimble in the way we do evolve’.

A.S.: Yes, there are strengths and weaknesses; pluses and minuses. Part of the difficulty with remaindering reform to Parliament is that criminal law is highly politicized. Politicians tend not to want to repeal crimes. As you know, there was a major initiative to codify English criminal law, and much of that initiative was admirable. In some ways it was an opportunity to modernize the criminal law. Yet they just couldn't get it enacted. There was not the political will, and in terms of the public reaction to change, it's particularly difficult to get significant revisions to the criminal law through parliamentary means. So, the fact that the judges can do it in the common law system (because their decisions create precedents) at least gives the law a form of evolution that is less politically sensitive. The case of Jogee is a good example – the case that abolished joint enterprise complicity liability in English law. That’s a good example of the courts being willing to do something that Parliament was not willing to do. The courts were responding to their sense that joint enterprise doctrine was leading to unjust outcomes, which was something they were able to do correct.

That’s a plus. At the same time, one potential minus is that they're doing it without going through a democratic process, something that raises complicated questions about justified delegation of power. The answers are not straightforward. The Swiss, for example, stage referenda to determine many important questions. Some jurists think that that is not a good idea, that we shouldn't be going back to the electorate every time, and that the electorate's proper role is more by way of supervisory overview. Of course, that itself invokes questions about what democracy requires in order to effect a just system of public decision-making. I wouldn’t want to take sides on such a large issue here. The thought I was offering was primarily that what codes tend to do is to set the structure of liability in stone. And that has strengths as well as weaknesses.

M.D.: I’d like to switch focus a bit. When you started talking to me, you said something about how dealing with problems in criminal law theory in an atomistic fashion doesn't always do the job. Things have to hang together in a certain way. So is there a Simester view of criminal law theory, a broad approach that you take to criminal law theory?

A.S.: Oh, I don't know! Well, I suppose it's in the opening chapter of Fundamentals. I think there are some key parts to the criminal law that demand theoretical justification, and these parts are often structural. So, consider the precise law of murder. In some jurisdictions you must have subjectively intended to kill, or in England, to inflict serious injury, when you do the act causing death. In others it’s enough that you may be intending to do something less serious, but something which creates an obvious risk of death. It might be a combination of subjective intent that falls short of what's required in England, combined with some objective components in terms of the dangerousness of what you do – there are lots of variations. I don't really have strong views about differences like that. It's certainly worth talking about, but it's not structural. What really counts for me is thinking about the basic requirements of criminal law. For example, one of the things the criminal law does when it's going well, as we know, is to convict only people who are culpable for the wrong of which they are being convicted. That creates a need to think about what makes somebody culpable, which in turn takes us into things like mens rea, but also into areas like excuses and justifications. Plus you've got related requirements, for example that the thing of which the defendant has been convicted must be a wrong, one for which the defendant is justly held responsible. Not just culpable, but also answerable: which is where doctrines like causation come in. And all of these requirements inter-relate, because the criminal law is multi-faceted. There are a lot of theorists who think the criminal law is essentially a punishment system, and that its justification is found in a justification of punishment. I simply don't think that. I see punishment in the criminal law as secondary to the conviction. The most important thing is the conviction. At the same time, I don't think uniquely about either convictions or punishment. A third thing the criminal law does is to prohibit. Before you get to a conviction, you have to have a prohibition, and the prohibition changes people's freedom. So, a just criminal law is one that not only convicts (and only convicts) culpable wrongdoers who are answerable for the wrongs of which they have been convicted, but it also convicts people only for crimes that are justly enacted.

It is crucial to see that these requirements are not independent. So, if you change the conditions of a conviction, and you say that, for example (as English law did) that objective recklessness is no longer sufficient to constitute recklessness, you change the scope of a recklessness-based prohibition. You don’t just change the range of convictions. Suppose, then, as the English courts have now ruled, that reckless criminal damage must be foreseen. If we ask, ‘When do I have to stop what I'm doing?’, the answer is no longer, when there's an obvious risk – which means I have to do a bit of a health and safety check before I throw this cricket ball (or whatever it is). Now, I'm free to act until I actually realize there's a risk, and in particular, an unreasonable risk. It is only then that I lose my freedom to do what I was about to do. My scope of freedom shifts with what looks like a culpability decision.

So, I guess to the extent that I have a distinctive theoretical approach, it lies in the fact that I see the criminal law as much more complicated than many theorists do. I think the criminal law has multiple functions which are all interrelated and not independent of each other.

M.D.: Let's focus on one of these functions to which you alluded, the criminalization function. There are various different ways in which we can decide what should be prohibited in advance. Some theorists start from existing constitutional structure, and say, ‘Well, this is what the Constitution requires of us, and then we'll follow that guidance’—that seems to be more of a civilian law approach. In the common law we tend to start with ideas that are more fundamental than that, things like harm, offence, moral wrongs. What is your approach to these questions?

A.S.: Yes, the justification of criminalization is itself a very difficult philosophical problem. My approach tends to be in line more with the common lawyers. But even within the common lawyers and philosophers from common law jurisdictions, there's a lot of variation. We can't really generalize across the common law. Some writers think that we start with morality, and that the job of the criminal law is to prohibit moral wrongs. It may not be sufficient that something is a moral wrong, but the moral wrongness of the relevant action is both a necessary condition for its justified prohibition, and also a positive reason for prohibiting it. Michael Moore is a good example of somebody who has a strong commitment to seeing moral wrongs as being the driving force behind justified prohibitions. For him, the purpose of the criminal law is to punish moral wrongs, and the only reason that the criminal law exists as an institutional form is to observe principles of legality. By contrast, I have always thought that moral wrongness is not sufficient to bring the criminal law into play. The criminal law is not a morally attractive institution. As you know, it’s nasty, brutish, and peremptory. It orders people around and treats them badly, and we hope it only does so when they deserve it. But it's not a good thing to be charged with a crime, and the criminal law is a very blunt instrument. In my view, it’s not something to be used just because we think there’s something morally wrong with a person’s conduct. That's why I end up endorsing a version of the harm principle, in as much as the criminal law should really only get involved when the relevant conduct is tends adversely to affect other people's lives: which is essentially the idea of harm. Because I think that because I'm conscious of the institutional downsides, both of having a criminal legal system and of having criminal prohibitions, I have a much less idealistic picture of the criminal law than many other legal theorists, especially those who are not lawyers.

M.D.: One question that that your approach seems not to address directly is, why should the State get involved, as opposed some other institution with authority, like for instance, to parents who have authority over their children? What's your view on that? Why is it the State's business that your conduct is affecting somebody else?

A.S.: Well, the “somebody else” could be a stranger, so I’m not sure that people like parents are the right people to protect the interests of strangers against the interests of their own children. That’s not going to work very well! So, the State has a higher chance of being impartial on this. And remember that the rules are going to have to be fair for everyone; they have to be even-handed across the community. So it makes sense that the state is the natural arbiter between strangers. Moreover, in principlpe, the state operates through a democratic process, which means that the creation of the laws is something that is susceptible of public review and discussion. If, instead, if you have other kinds of actors, private actors for instance, creating the rules, you don't have any of those protections. I’m not saying that the state is always impartial, or that there are no lobby groups or the like, that affect policy formation. Political parties themselves are comprised of people who tend to be self-interested. But at least it's out for public review, and the rules apply in principle to everyone, even if they benefit some people more than others. So, I tend to think the State is a problematic collective arbiter of our freedoms, but it's the least bad arbiter – a bit like Churchill's view about democracy.

M.D.: So, putting these ideas together, it sounds like the state should get involved because it happens to be the least bad agent to do these things, and it should get involved when a person's conduct interferes with someone else's life; so, the harm principle. And the rules that it sets need to be even-handed. Deciding whether subjective culpability is required for criminality, or objective culpability is required, changes what is demanded of one, ex ante. Specifically, it changes how much care we are required to take when we do things. Would that be a fair summary of your view of criminalization? Or is there something that I'm missing?

A.S.: Well, it's not just that the conduct is potentially harmful, but it should also be a wrong. That's implicit in the requirement that the state should be picking out only culpable actors. To be culpable for something, that thing has to be a wrong of some kind. I guess what's the only thing that's really distinctive there is that the wrongness can't be in the air. It's not the State's business to prohibit pure moral wrongs. Partly because doing so is really very intrusive, moral wrongs that don’t damage other people’s lives don’t seem to me to be the kind of thing that parliament should be policing.

I should add, though, there are other complexities here. One such complexity is the problem posed by remote harms. Some crimes prohibit in itself harmless activity, carrying a concealed weapon in public. Why is that a crime? Not because the carrying is in itself harmful, but because it tends to conduce to subsequent harmful activities. I think a lot of thought needs to be put into what are the appropriate restrictions on criminalization of conduct that’s not itself harmful but which has that tendency to lead to further, harmful conduct. Another challenge is what is sometimes called the ‘offence principle’. Giving offence is a very interesting case because it does involve conduct that affects other people. But it just upsets them, and doesn’t tend to cause them harm. How, then, should we respond to that? That's also a difficult question. As you know, I think we should regulate offensive conduct only if it tends to also damage people’s lives, but it’s clear that offensiveness raises additional objections to mere immorality. So, we might agree that wrongs which by their nature tend to be harmful are the right sorts of things for us to be criminalizing, but there are a lot of complicated cases beyond that which might deserve criminalization.

M.D.: And I guess this is part of the fuzziness that you were talking about earlier.

A.S.: Very much so!

M.D.: So, let's imagine now that criminalization has been sorted out. Parliament’s gotten a handle on that. And now someone has gone ahead and done something that looks like a crime, and we've got to decide: are they guilty? Are they liable to a conviction? At this point, you're looking at questions of ascriptive responsibility and culpability. What's the Simester approach to those questions?

A.S.: Reasonably mainstream. Ascriptive responsibility presents interesting questions about moral agency, especially about whether the defendant is answerable as a moral actor for what occurred. Amongst other things, we need to know whether there is a sufficient connection between the event and the defendant’s ability to engage with moral reasons. Relatedly, we need to know whether the defendant’s behaviour was voluntary. And then there are other familiar issues: if the defendant's behaviour caused the event in question by omission, for example, did the defendant have a specific duty to intervene which turned that omission into the kind of wrong that the criminal law should address? All these involve well-known doctrinal issues, yet despite being well known, they remain interesting and complex. From a structural perspective, my thinking about these problems is fairly conventional. Of course, it isn’t entirely uncontroversial. Some writers, for example, think that insanity and infancy should be excuses rather than full scale denials of responsibility. For the most part, though, academics tend to agree about where such doctrines sit within the structure of criminal liability. Mainly, the disputes lie in the details.

Having said that, the disputes themselves can be enormously complex. At the moment, one of the most interesting areas I’ve been thinking about is causation. It is tremendously difficult to come up with a credible set of principles that can account for the myriad ways in which forces in the world interact with human behaviour and ultimately lead to outcomes such as harms to other people. We can – and do – have arguments about when causal responsibility should be ascribed to the defendant in situations where the harm has occurred through, or in conjunction with, the agency of other people, or unexpected events, and so on. There is no simple answer to such questions. That's why Hart and Honore wrote a whole book!

M.D.: On the culpability question: in the common law, we tend to draw a fairly sharp distinction between subjective mens rea (where the defendant actually foresees or intends the relevant outcome) and objective mens rea (where they did not, and instead acted, say, negligenty). Doubtless that’s all very intuitive and plausible. But what is it about subjective fault that gives us an especially good account of why somebody deserves a conviction, and which isn't there in objective fault?

A.S.: Different jurisdictions draw such mens rea distinctions in different ways, but the moral pressure points are always in the same places. In the Indian Penal Code, for example, there's a crime of causing death by a ‘rash’ act. That presents the same problem: should ‘rash’ be interpreted as an objective or subjective mens rea standard?

Just as a matter of background, it seems to me that capability can lie in a case where the defendant didn't subjectively foresee a harm that he or she caused. Some people deny that. I wouldn’t want to deny that. But what I would accept is that culpability doesn't lie for the same reasons. So, if somebody foresaw their wrongful act – they were aware that they were, or may be, inflicting a significant harm in a situation where there was no justification for doing so – we can say, when they nonetheless go ahead, that they had chosen to do so: and in so choosing, they manifested a preference for bad values. We might say that they have aligned themselves with wrongdoing, and that gives us a much more direct insight into their values, into what kind of person they are. We can describe them as a person who prefers to do the wrong thing, who prefers to do something they shouldn’t be doing. But we can't say any of that when the actor didn’t think of the risk. It may be that, if they’d thought of the risk, they would have been horrified. My point is that, in cases of inadvertent wrongdoing, we can't make the same kind of direct inference about the actor’s values. That doesn’t mean we can’t blame them for failing to think about a risk. I think that, at least sometimes, we can. The fact that they failed to think about a risk can tell us something about them too. Maybe they didn't care enough; maybe their empathy for the people around them is deficient in some way. But even if we grant that, it's clearly not the same explanation of culpability as it is for somebody who has actually perceived the risks and deliberately taken gone ahead. Ultimately, if both objective and subjective forms of mens rea are to be genuine grounds of blame, we need it to be the case that both of them reflect badly upon the defendant’s moral values. But the way in which they do so is different. The objective account of culpability faces more hurdles in making that link to the defendant’s moral values.

M.D.: Okay, so let's move further along the trial process. You mentioned insanity and involuntary actions earlier. Beyond those exculpatory pleas, there are other defensive pleas like self-defence, necessity, and duress. What explains why they exculpate?

A.S.: I have a broad trichotomy of defences, or what are typically called defences in the common law. Some defences essentially deny moral agency, like infancy, insanity, and involuntary action. This event didn't belong to the defendant as a moral agent, as somebody who is sufficiently capable of reasoning morally about what they do. The thing about those defences is that if somebody has a full-scale irresponsibility defence, then it doesn't matter what they did. There's no evaluation of the rights or wrongs of what they do. We just don't get to that evaluation. By contrast, defences such as self-defence, duress, and the like (and I set aside procedural defences like abuse of process here), are either justifications or excuses. They're different from irresponsibility defences, in as much as they involve the court or the jury in making an evaluation of the defendant’s rationale for acting. When a defendant acts with an excuse or with a justification, they choose to do the actus reus; they act with mens rea. But they can offer a further explanation of why they chose to do the actus reus. Unlike the irresponsibility defences, we then have to make an evaluation of their reasons for doing the actus reus.

Suppose, then, that we are in the realm of justifications and excuses. We are evaluating the defendant’s reasons for acting. Now we have a dichotomy. From the point of view of exculpation, the defendant’s reasons can do one of two different things. They could be sufficient. We might say in such cases, ‘Normally, this would be a bad thing to do, but on this occasion it was morally permissible. It was justified.’ Self-defence is of that kind. If you're attacking me with a with a sword, and I pull out my gun and shoot you in self-defence, the conclusion of the law is that this is a very unfortunate thing, but it was permissible in that situation for me to do it. Excuses aren't like that. But they still hold the defendant up to a standard of reasonableness. In duress, we ask, ‘Was the threat sufficient to make it understandable that I acted as I did?’ Again, this is a question about the reasonableness of my response. Did I live up to the standard of a reasonable, decent person when I succumbed to the threat? If it was reasonable to succumb to the threat, or understandable – if an ordinary decent person would have succumbed to the threat – then I shouldn't be blamed for doing so, because I've lived up to the standard that can reasonably be expected of me. That it doesn't make what I did okay. I may have committed a tort, and maybe have to pay compensation for my wrong. So justifications and excuses come apart on question of permissibility. But they come together on the question of ‘Were your reasons adequate to either make it permissible or otherwise exculpate?’.

M.D.: From your catalogue of writing, which is vast, I notice that you say relatively little about punishment. Firstly, why? And secondly, what is your thinking on punishment?

A.S.: There's a big literature on punishment theory, and up to a point, I'm not sure that I have a great deal to add to what has already been written. I am a retributivist about punishing people: the fundamental justification for punishing people is that they deserve it. That in turn implies that they’ve done a culpable wrong. (Alternatively, it may be that they are answerable for someone else’s culpable wrong, but then they could only be answerable through having done something wrong themselves.) It’s is a fairly mainstream position. To my mind, Andreas von Hirsch articulated a persuasive version of desert-based punishment. Basically, my feeling is that the literature has covered what I have to say in this area already.

M.D.: The reason I ask is that for some theorists, their view on the nature and purpose of punishment dictates the rest of their thinking about the criminal law. It seems to me that is not your way of approaching these questions. So, if that's the case, does your view of punishment at least influence your thinking on other matters of substantive criminal law very much?

A.S.: Not hugely. I don't see punishment as a distinctive or defining feature of the criminal law. The law doesn't have a monopoly on punishment. Parents punish their children. A red card in football is a form of punishment. Punishment is a society-wide feature. It is, of course, something that the criminal law happens to do a lot. The criminal law is, amongst other things, a punitive institution, and that generates some constraints on it. But for the most part, the constraints are already there in the conviction. Conviction publicly labels a person as a certain kind of culpable wrongdoer. That label needs to be justified. I don’t see the additional infliction of punishment as really changing that, although it obviously adds to the case for setting a criminal legal system up so that it only convicts culpable wrongdoers. But punishment is not even uniquely criminal even within the legal system. Tort law has punitive damages. There are other areas of law that punish, and sometimes you can get a conviction without punishment. So, I tend to think that substantive criminal law already contains the basic need to get to a justification of punishment. Admittedly, there are some interesting further questions, such as about the quantum of punishment. Again, though I don't have anything in particular to say about that, except insofar as I tend to think that luck should not affect the quantum of punishment. That's a controversial view, but I think that whether you succeed or fail in an attempt to kill someone should affect the conviction, i.e. what offence you are convicted of, but should not necessarily change the punishment. But beyond some fairly simple, abstract thoughts like that, I don’t have too much more to say.

 M.D.: So, I just want to ask, as a final question, about your academic or your legal theory heroes. You mentioned Joseph Raz. You mentioned a few other names: Jim Evans, you mentioned Andreas von Hirsch. Are there any others that are Simester’s go-to philosophers of criminal law? Others that influenced you a lot?

A.S.: I did my DPhil at the same time as John Gardner, and we spent quite a lot of time walking around Christ Church Meadow, talking through criminal theory problems. John was certainly somebody who influenced my thinking. He also had an influence on the way in which I wrote. I learned a lot from reading his articles. John was also a fellow student of Joseph's. His approach to criminal theory was a little different to mine, but I gained a great deal from talking to him and getting a fellow student’s perspective on the kinds of things Joseph was trying to teach us. John was somebody with a significant, and in many ways enduring, influence on how I think about criminal law problems. That would be the other person I would want to point to.

M.D.: Thank you, Andrew. Thanks very much for your time. It's been really interesting for me to get know you a bit better.

A.S.: Thank you, Mark.