Solum on the Normative Arguments for Originalism (Part Two)

(Part One)

This is the second part in a brief series of posts looking at Lawrence Solum’s normative arguments for originalism. Originalism is a theory of constitutional interpretation which holds that a written constitution (could be any, but the US constitution usually) ought to be interpreted in accordance with the semantic content that its clauses had at the time of ratification. Roughly speaking, there are two ways in which to cash out what that semantic content is: (i) intentionalism, which holds that the content is fixed by the intentions of the framers; and (ii) conventionalism, which holds that the content is fixed by the public meaning of the clauses at the time of ratification.

In an earlier post, I outlined some of Solum’s arguments in favour conventionalism. In this two-part series, I’m changing tack and looking at Solum’s normative arguments in favour of originalism. The arguments are taken from the book Constitutional Originalism. These normative arguments play a crucial role in the debate over originalism. After all, even if Solum and other originalists are correct in saying that the semantic content of the constitution really is fixed at the time of ratification, it does not follow that judges and other officials ought to respect it. There may be moral reasons for overruling or occasionally disregarding it; for changing and amending it; or for tearing it up and drafting a new one.

In part one, we looked at one normative argument in favour of originalism: the argument from popular sovereignty. As we saw, there are a number of problems with this argument, at least in the form adopted by Solum. Today we’ll look at another argument, one that happens to feature quite prominently in the debate about originalism: the rule of law argument. According to this argument, originalism can be endorsed thanks to the positive role it plays in promoting the rule of law. To flesh out this argument, we’ll first need to consider the virtues of the rule of law. Then, we’ll need to explain how originalism contributes to it. Finally, we’ll need address the criticisms of the rule of law argument.

Let’s get to it.


1. The Rule of Law and its Virtues
It’s almost a cliche to say this, but here goes: the rule of law is a somewhat elusive concept. The basic idea is readily stated: people should be governed by laws, not by arbitrary exercises of power. To put it another way, and to quote from John Adams, there should be a “government of laws, not of men”. That’s not to say there won’t be men (and, of course, women) doing the governing, it’s that they will not be above the law. They, like everyone else, will be subject to it and accountable under it.

That much is straightforward. The devil is in the details. That is, in precise mechanisms through which this preferred state of governance is to be realised. Some complain that the “rule of law” is an empty piece of rhetoric, a placeholder for all the positive aspects of a political system. But this isn’t entirely fair. There is some reasonable guidance out there. For instance, the American legal philosopher Lon Fuller famously identified eight features of good law-making. Among these features were publicity, prospectivity, certainty, and consistency. These contributed, in his words, to the “inner morality of the law”, which is presumably what the rule of law consists in. More recently, the English jurist, Lord Justice Bingham, identified another set of eight principles — among them, again, accessibility and predictability, but also lack of discretion, equality and basic humans rights protection — which he felt constituted the rule of law. Fuller and Bingham seem, by my lights, to be on the right track.

Ignoring the mechanisms for the moment, let’s turn to the moral question: why should we prefer the rule of law? On the face of it, there doesn’t appear to be any necessary connection between the rule of law and the morally preferred state of existence. It is surely possible for there to be a benevolent dictatorship or a wicked set of accessible, predictable and consistent laws? Maybe; maybe not. There are some who argue that the kinds of principles identified by Fuller and Bingham have intrinsic value, a value that dictatorial or arbitrary system of governance will obviously lack, but there are also those who will argue that the rule of law has instrumental value, either because it does (or is more likely to) lead to better kinds of outcome.

I can’t say exactly where I come out on all of this. I certainly think a “government of laws” is a better bet than “government of men” (going by my cursory sweep over the historical evidence). But I’m not sure why or what its precise constituents might be. Fortunately, I need neither of these things right now. For the purposes of this post, I will simply assume that the rule of law is morally desirable and that predictability, certainty, prospectivity and avoidance of discretion are part of the picture. Solum, I should add, seems to agree.


2. The Rule of Law Argument: What’s Originalism got to do with it?
With this theoretical background in place, the rule of law argument can be developed. The essence of the argument is twofold: (i) that an originalist method of interpretation can contribute to the rule of law; and (ii) that it does so better than other approaches to judicial decision-making. The first part of the argument is reasonable enough, but the second part is much trickier and also crucial to its overall success.

Before we get too bogged down in the details, let’s try to formalise the basic argument (numbering follows on from part one):

• (7) We ought to protect and sustain the rule of law. 

• (8) Originalist interpretation of the constitution helps to protect and sustain the rule of law. 

• (9) No other theory of judicial decision-making does a better job of protecting and sustaining the rule of law. 

• (10) Therefore, we ought to adopt an originalist approach to the interpretation of the law.

Now let’s get bogged down in the details. By which, of course, I mean the defence of each of the key premises. We’ve already looked at premise (7), so we’ll leave it to the side and focus on (8) and (9) instead (don’t worry we’ll be getting back to (7) when we look at criticisms of the argument as a whole).

Premise (8) is defended by the appeal to the rule of law virtues of predictability, certainty and non-arbitrariness. As stated above, what distinguishes the rule of law from, say, the rule of force is that everyone is subject to and accountable under the law and that no one individual has the power to change those laws at a whim. For this to work, we have to make sure that the laws are sufficiently stable and well-understood so that people can order their lives in relation to them, and that they are sufficiently constraining so that no one can alter or undermine them with ease. Solum, along with other originalists, argues that this is only possible when the laws have a stable core of fixed meaning. And, further, that this is only possible if we adopt an originalist approach to interpretation.

The argument being made here is an argument from constraint, which is probably the most popular and common way to defend originalism. The claim is that the originalist approach constrains judicial (and presumably other official) interpretations and applications of law within the — sometimes fuzzy — boundaries of original meaning. We state the argument thusly:

(11) If an approach to judicial decision-making provides predictability, certainty and non-arbitrariness, then it protects and sustains the rule of law. (12) Originalist interpretation of the constitution provides predictability, certainty and non-arbitrariness (by constraining judges and officials within the boundaries of a fixed original meaning). (8) Therefore, originalist interpretation of the constitution helps to protect and sustain the rule of law.

We shan’t say anything critical about this argument at this juncture. We’ll postpone that ’til the next section. We move on instead to the defence of premise (9). As I said, this is crucial to the overall argument: it’s not good enough to say that originalism protects the rule of law, it must be shown that it does a better job of this than alternative theories of judicial decision-making. Can an argument be made to that effect?

If we stick with the virtues of predictability and certainty, then it can. Consider the usual alternatives to originalism — living constitutionalism, judicial pragmatism, common law constitutionalism and so forth. Each of these theories distinguishes itself partly on the grounds that it provides greater flexibility to judges, that it frees them from the “dead hand of the past”. Indeed, as Solum notes, advocates of such theories are often keen to defend them precisely these grounds. But these grounds imply a lack of predictability and certainty in judicial decision-making. That would give us the following argument:

• (13) If an approach to judicial decision-making allows for some flexibility and discretion, it implies that such decisions will (potentially) lack predictability and certainty (which are needed in order to sustain the rule of law). 

• (14) Leading alternatives to originalism (living constitutionalism etc.) allow for greater flexibility and discretion than originalism. 

• (15) Therefore, leading alternatives to originalism will create a greater potential for unpredictability and uncertainty than originalism (which is damaging to the rule of law).

As can be seen, this argument isn’t yet a defence of premise (9). There are a couple of reasons for this. The most important one for now is that the argument says nothing about the arbitrariness of the decisions under the alternatives to originalism. As you recall, “arbitrariness” was also part of the rule of law virtues that originalism was supposed to protect. Why doesn’t the argument say anything about this? The answer lies in the nature of an “arbitrary” decision. To my mind, non-arbitrariness doesn’t simply equate with predictability. If it did there would be no reason to include it as a separate component in the rule of law. Instead, non-arbitrariness suggests a decision that is grounded in sound moral principle, that is not simply an exercise in vindictiveness, whimsy or caprice.

And this is where the defenders of the alternatives to originalism may gain some ground. Presumably such advocates don’t defend their theories because they are lovers of unpredictability and uncertainty; presumably, they do so because they think there is something morally superior to them. In particular, they might be inclined to argue that affording judges greater flexibility in their decision-making allows them to make decisions that avoid the moral mistakes of the past and are more attuned to the moral needs of the present. But in that case the decisions would be guided by moral principle and would thus be non-arbitrary.

Originalists will need some sort of response to this. Have they got one? Solum makes a few gestures in the right direction (pg. 41) by claiming that once we grant judges the power to overrule/amend the constitution they will exert that power in a political and legal vacuum. In other words, they will stand above the law - free to change it as they see fit - a situation we sought to avoid by promoting the rule of law. And while that won’t necessarily lead to problems, it increases the changes that problems will arise. We can well imagine what might happen: with the power vacuum opened up, politicians will seek to appoint ideologically driven judges and there’s no reason why these judges would have to adhere to a morally commendable ideological cause. That is to say, if we allow for discretion, we create the conditions in which arbitrary decision-making can flourish, even if we don’t necessitate that outcome.

• (16) If judges are given the freedom to amend or overrule the constitution, this will create the conditions in which the arbitrary exercise of power can flourish (which would be contrary to the rule of law).

• (17) The leading alternatives to originalism would allow judges the freedom to amend or overrule the constitution; originalism would not, or, at least, would do so to a lesser degree. 

• (18) Therefore, leading alternatives originalism would be more likely create the conditions in which the arbitrary exercise of power can flourish.

When you combine (18) and (15) you get a defence of premise (9). The rule of law argument is then complete.

This is a rather complex argument, so it might behoove us to consolidate the chain of reasoning into an argument map.

3. Problems with the Rule of Law Argument
I held off on criticism in previous section. Now it’s time to correct for that omission. As I see it, there are two types of criticism one can make of the rule of law argument. The first focuses initially on premise (7), but then has downstream effects on premise (9). The second follows a similar pattern only it starts off with premise (8). I’ll briefly outline both types of criticism here. I will not have the time to develop them in any detail, but I’ll lay the foundations at least.

Let’s kick things off with premise (7). As you recall, in its original form it stated:

• (7) We ought to protect and sustain the rule of law.

But this version of the premise lacks a crucial qualification. While it may be true, in general, that we ought to protect and sustain the rule of law, that isn’t necessarily true in all cases, particularly if the rule of law is taken to require fidelity to the original meaning of the text. Why not? Because the original meaning could require the perpetuation of great injustices. To take an extreme (and hypothetical) example, a constitution that said women and racial minorities were “inhuman” and could be used as “slaves” or “experimental subjects” “without their consent” would be unjust, and there would be no strong reasons to swear fealty to it. While this is an extreme example, there are some who would argue that the US constitution permitted injustices of this sort in its original form (e.g. by failing to outlaw slavery). So we need to modify premise (7*):

• (7*) We ought to protect and sustain the rule of law, provided that the law is not unjust.

Modifying the premise in this manner creates a number of problems. One of these is that it opens up a chink through which the non-originalist theories — banished by premise (9) — can creep back in. Proponents of those theories will argue that if a constitution does contain an original injustice, there is good reason to adopt an alternative to originalism, and to allow judges (and other officials) to correct for that injustice. Of course, this line of attack won’t always work — it all depends on whether the constitution contains an original injustice — but it does highlight the fact that fidelity to the law is only justified if certain preconditions are met.

How might an originalist respond? They will typically suggest that there is an alternative and better solution to the problem of original injustice: amend the constitution by plebiscite, not judicial opinion. Solum puts it this way: if you were to select, from behind a veil of ignorance, a procedure for amending the constitution, would that procedure grant final authority to a small number of judges or to the people en masse? The question has some rhetorical force, no doubt. I should also note that amendment by popular vote is quite common in many countries (certainly in my home country) and works quite successfully. Ironically, this does not seem to be true in the US. Mainly this is because amendment is more difficult there, but it could partly (as some have argued) be because judicial activism has drawn attention away from that possibility.

We turn now to the second type of criticism. This one focuses its ire on premise (8) which, as you’ll recall, stated that originalism did help to protect and sustain the rule of law. The criticism suggests that it does no such thing. To appreciate the criticism we need to refresh our memories about the broader defence of premise (8) which was outlined above. This defence specified that the protection of the rule of law was achieved by constraining decision-makers within the boundaries of original meaning and thereby helping to promote predictability, certainty and non-arbitrariness.

But is that actually true? There are those who argue that the constraining powers of originalism are greatly exaggerated. That even if we do try to stick to the original meaning, there will be plenty of room for discretion and flexibility. The US Supreme Court decision in Heller is often thought to illustrate this: both sides focused on the historical meaning of the 2nd Amendment but both reached remarkably different conclusions.

There is much that could be said about this claim, and this example, but I want to sidestep all that and make a more general point. This point links directly into premise (9). I begin with this observation: even if originalism did contribute to predictability, certainty and non-arbitrariness, there doesn’t appear to be anything unique about originalism in terms of its power to do this. Judicial philosophies and ideologies will nearly always constrain judges in ways that make their rulings predictable and could thereby provide legal certainty. For example, Justice William O. Douglas was one of the more notorious “liberals” on the US Supreme Court in the mid-20th century. He famously purported to find rights hidden in the “emanations” and “penumbras” of constitutional clauses. But for all that, there was nothing very unpredictable about his decision-making: he followed a very clear and obvious ideology in his decision-making. One nearly always knew where he would come out on a controversial issue. This was true of other famous justices too, even when they didn’t follow an originalist approach.

The point is this: originalism doesn’t win the normative battle simply because it provides predictability and certainty. Most judicial theories and philosophies will do this. If originalism is to win, it will have to be on independent normative grounds. And as we saw the last day when investigating the argument from popular sovereignty, at least one of those independent grounds is not up to the job.


4. Conclusion
That brings us to the end of this series. As we saw in this post, one common defence of originalism holds that it both: (a) manages to protect and sustain the rule of law and (b) does so better than alternative theories of judicial decision-making. This is because it constrains judicial decision-making in such a way that promotes the rule of law virtues of predictability, certainty and non-arbitrariness. However, as we saw in the previous section, this defence of originalism is open to at least two criticisms. The first is that the rule of law ought only to be promoted if certain preconditions are met. The second is that the constraining powers of originalism — at least when contrasted with other approaches to judicial decision-making — are exaggerated.

Solum on the Normative Arguments for Originalism (Part One)

(Solum On Semantic Originalism)

In an earlier post (link is above) I looked at the theory of constitutional meaning known as semantic originalism. This theory maintains that the semantic content of a constitution (presumably any constitution although the focus is on the US constitution) is fixed by the public meaning of its clauses at the time of ratification. This theory has been developed to its most sophisticated form by Lawrence Solum and in the earlier post I addressed some of Solum’s arguments in its favour.

Solum argues that the semantic thesis is a strictly factual one. That is to say, whether or not the semantic content of the constitution is fixed by the original public meaning of its clauses is something that can be discovered as a matter of fact. We do not choose the semantic content on the basis of normative/ethical criteria. The semantic content is what it is, it is not something else. We can’t simply change it.

But, of course, the debate doesn’t end there. Even if it is true that the semantic content is fixed in this manner, it does not follow that we ought to respect it. So originalists need to go further and try to present good reasons for respecting the original public meaning of the constitution. Over the next two posts I want to take a look at some of those reasons.

Once again, I shall call upon Solum’s discussion of originalism in the debate book Constitutional Originalism. I have noted before that this book doesn’t contain the most elaborate defence of Solum’s theory. For that, you need to go to his article/monograph “Semantic Originalism”. Nevertheless, when it comes to the normative arguments for originalism, the debate book may have an advantage over the article. To me, at any rate, it lays out the arguments in a crisper and more emphatic manner than the article does. Still, Solum is somewhat half-hearted in his discussion of the normative arguments in the article, since his primary focus there is on the semantic thesis. Better arguments can be found elsewhere, but this is at least a start.

In the remainder of this series I’ll look at two of the normative arguments Solum uses to defend originalism. They are the popular sovereignty argument and the rule of law argument. I’ll roughly follow the presentation in the book (pgs. 36-54), but will also feel free to elaborate on the arguments presented therein and to offer some critical commentary as I go along. Bear in mind throughout this series that the goal of each argument is to defend something like the following proposition: “we ought to interpret the constitution in accordance with its semantic content (which is fixed by the original public meaning of its clauses)”. This proposition is assumes that Solum’s semantic thesis is correct and that is an assumption we shall share throughout.

This post will focus on the popular sovereignty argument. The rule of law argument will be discussed in part two.


1. The Popular Sovereignty Argument
That the popular will should be respected seems like too obvious to state. To the extent that we are committed to democratic governance — and I shall assume such a commitment here — it would seem like the law ought to both respect and implement the popular will. Solum sums this up by reference to what we might call “Lincoln’s Principle”:

Lincoln’s Principle: Government should be of the people, by the people, for the people.

Now, that’s probably not the most philosophically sophisticated way of expressing the core principle of popular sovereignty, but it’s enough to get the gist of the idea. The key question now is how this can be used as the basis of an argument in favour of originalism. The simplest argument, and the one I think is implicit in Solum’s comments (pg. 42) is the following:

• (1) We ought to respect the will of the people. 

• (2) The constitution, through its semantic content, expresses the will of the people. 

• (3) Therefore, we ought to respect the semantic content of the constitution (which happens to be fixed by the original public meaning).

As I said this is pretty simple. Premise (1) just states the core principle of popular sovereignty. Premise (2) is slightly trickier. I think it is fair to say that people express their wills by appealing to the semantic content of conventional signs and symbols, but it’s not right to say a priori that a constitution expresses the will of the people. After all, it’s possible for a constitution to be drafted and ratified by a dictator. So premise (2) can only be true of particular constitutions drafted in a particular manner. The particular constitution that Solum focuses on is, of course, the U.S. constitution. And he thinks there are good historical reasons for thinking it really did express the will of the people when they ratified it (we’ll get back to those historical reasons in a moment).

But this point — that the constitution was an expression of the will of the people at the time of ratification — presents a problem for the defender of originalism. The problem is this: premise (1) and (2) are ambiguous as to which people’s will needs to be respected. I presumes that premise (1) is referring to the people who are currently alive and kicking; but then premise (2) — if assumed to be a statement about the U.S. constitution — is surely referring to people who live over two hundred years ago. If the ambiguous temporal references are removed such that we have the following premises:

• (1*) We ought to respect the current (2012-ish) will of the people. 

• (2*) The U.S. constitution, through its semantic content, expresses the will of the people circa 1789.

But if we have these two premises, then the conclusion (3) no longer follows. It is certainly possible that the will of the people living in 1789 is the same as the people living in 2012, but this cannot be taken for granted. In other words, if we want to respect the current will of the people, we would have take positive steps to ensure that this was happening each generation (e.g. by going through a new constitutional convention). We couldn’t just assume that nothing had changed. This suggests that legislation, which is enacted to popularly elected representatives, might be a better legal mechanism for respecting the will of the people than a constitution.


2. Shoring Up the Popular Sovereignty Argument
Solum has a response to this objection. It is, however, a little…odd. I’ll do my best to do justice to his reasoning here. For those who want to check up on me, the relevant passage of Constitutional Originalism is on pgs. 42-43.

Solum’s argument is that there is an important sense in which the constitution of 1789 is more representative of the popular than contemporary legislation. As he sees it, the problem is that legislation is enacted by popularly elected officials and, to quote from the man himself:

“In a representative democracy such as ours [i.e. the US], political institutions are (at best) an imperfect reflection of the true preferences of the citizenry. The majority of citizens do not vote. Those who do vote are frequently unaware of the likely consequences of their ballots. Special interests dominate ordinary politics through lobbying, campaign donations, and the weak institutional structures of many legislative and executive institutions. The political branches are truly political, but that does not imply that they are deeply democratic. (pg. 43)”

I have no reason to doubt this claim. Certainly, the popular representations of the U.S. political system (in news media primarily) seem to bear this out. But just because the current situation is no good does not imply that the original enactment of the constitution was any better. Solum needs to provide additional evidence for this. As it happens, he does not do this. Instead, he simply asserts that “[b]y contrast, the Constitution was adopted through an intensely democratic process.” To be fair to him, providing an exhaustive history of the drafting and ratification process would be inapposite in a book like this, and there are some historical considerations (possibly well-known to the book’s target audience) that bear him out. Chief among them being the supermajoritarian process of ratification. But there are also some historical considerations that contradict that point, e.g. the absence of women, African-Americans, and men who did not meet the property qualification, from the process,. Solum acknowledges these flaws, but thinks they do not outweigh the positive aspects of the process. My feeling is that they cannot be so easily dismissed. However, I’m going to take up that line of criticism in a later post on a different article. Stay tuned.

Leaving that to the side for now, I’m going to suggest that there are still some troubling tensions in Solum’s line of reasoning. But first, I’m going to try to formalise his argument. I think the following does some justice to the position:

• (5) Current legislative mechanisms for enforcing the popular will are imperfect: many people do not vote, those who do are unaware of the consequences of doing so, special interests dominate the legislative process and so on. 

• (6) The constitutional drafting and ratification process was, by contrast, intensely democratic and so the constitution is a better representation of the popular will: it was supermajoritarian, serious effort was made to allow people to participate in the process (despite notable absences) and so on. 

• (7) Therefore, the constitution is a better representation of the popular will than current legislation.

What to say about this argument? There are at least three significant flaws.

The first is that the argument doesn’t get the defender of originalism to where he/she needs to be. Remember, this counterargument is supposed to be helping them to defend the overall conclusion (3) that we ought to respect the original public meaning of the constitution. But this particular argument provides no support for that claim. The constitution may well be better at representing the popular will than legislation, but that does not imply that we ought to respect it. We could have good reasons for rejecting both legislation and the constitution. The least bad theory does not necessarily win out. For it to do that, we’d need to show that no other system for enacting the popular will is: (a) possible; and (b) capable of doing a better job.

The second flaw is that the argument does nothing to address the past vs. present problem highlighted earlier. All this argument really says response says is that the constitution did a better job of representing the popular will circa 1789 than legislation does circa 2012. But that doesn’t mean that the constitution does a better job of representing the popular will circa 2012. Again, additional argumentation would be needed to defend that conclusion.

The third flaw is that the argument seems to shoot itself in the foot. Solum is claiming that the current political system is imperfect in giving voice to the popular will. But surely the constitution itself must take a large portion of the blame for this? After all, one thing that the constitution does is to set up the political institutions that are currently failing us. (Speaking as a non-citizen and non-resident the use of the word “us” may be inappropriate but it just sounded nicer…) That’s not to say there aren’t other social and cultural pressures responsible for the current problems, but the fact that those pressures have the effect they do suggests, once more, that there might be a flaw in the original constitutional framework. Thus, to use the failures of the current legislative system to endorse the constitution seems self-defeating.


3. Conclusion: Looking Forward
That brings us to the end of part one. To sum up, in this part we have looked at one normative argument in favour of originalism: the popular sovereignty argument. That argument holds that because the semantic content of the constitution expresses the will of the people that semantic content ought to be respected. But this argument is problematic because the will of the people who drafted and ratified the constitutional framework is not necessarily the same as the will of the people who are currently subject to it. Thus, if popular will is the guiding consideration, current legislation would seem to have greater normative force than the constitution.

Solum tries to respond to this critique by offering some reasons for thinking that current legislation does a worse job of representing the popular will than the constitution did when it was drafted and ratified. But this argument has three significant flaws. It does nothing to support originalism since the least-bad system does not win by default. It does nothing to address the past vs. future problem raised by the critique. And it seems self-defeating since the constitution is partly to blame for the problems with the current system.

We’ll look at the rule of law argument the next day.

Addiction and Responsibility (Part Three)

(Series Index)

(Part one, Part Two)

This is the third and final part of my series on addiction and responsibility. The series is working off the article “Lowering the Bar for Addicts” by Gideon Yaffe. In this article, Yaffe defends the following two theses:

Unreasonable Compliance Thesis: An addict is not responsible for violating a norm if in order to comply with that norm they must cede control of their behaviour to something that is independent of their decision-making capacities.

Substitution Thesis: In cases where standard norm-compliance would be unreasonable for the addict, there is usually a lesser norm that the addict can be held responsible for violating.

Part two covered Yaffe’s general defence of the unreasonable compliance thesis, but did not explain how it applied to addicts. In this part, we shall consider Yaffe’s general defence of the substitution thesis and then, at last, consider how both theses might apply to the addict.


1. The Hopeless Romantic and the Substitution Thesis
The substitution thesis claims that even if compliance with a norm requires the unreasonable ceding of autonomous control to another, there is often a lesser norm that it is reasonable to comply with. But what exactly does this mean? How could there be a lesser norm? And what makes compliance with it reasonable?

Yaffe uses a thought experiment to explain the situation. Imagine a man — call him the “Hopeless Romantic” — who is constitutionally incapable of learning from his past romantic failures. No matter how many tragic, ill-conceived and acrimonious relationships he goes through, he continues to find himself drawn to unsuitable partners. This is because he has some sort of learning deficit. He learns more slowly and requires much more extreme signals in order to properly recognise the reasons for and against pursuing particular kinds of relationship.

What is this man to do if he is to avoid future disappointment? Well, one thing he could do is cede all relationship-related decision-making authority to a third party. For example, he could get one of his friends to choose his future romantic partners, or he could sign-up to a service that arranges marriages. But, as discussed in part two, it would be unreasonable to expect him to do so because this would require the renunciation of autonomous control (in this particular domain).

So what can we expect him to do? Here, Yaffe distinguishes between two norms that we could apply to our hopeless romantic:

Avoid zealous pursuit: This first norm counsels our hopeless romantic to avoid the zealous pursuit of particular romantic partners. 

Avoid weak pursuit: This second norm counsels our hopeless romantic to avoid all kinds of pursuit of particular romantic partner.

Obviously, the second of these norms is more stringent than the first. It demands that all forms of pursuit be avoided; the first simply counsels against the zealous forms of pursuit.
Yaffe’s argument is this: while it may be unreasonable to expect the Hopeless Romantic to comply with the norm of weak pursuit, it is not necessarily unreasonable to expect him to comply with the norm against zealous pursuit. This is because our romantic is not incapable of recognising reasons for and against pursuit, it’s just that he is less sensitive and takes longer to recognise them. Due to his past experience he knows that his romantic adventures tend to end in failure, and so he recognises reasons for being cautious and less “full-on” in his advances, but he can’t recognise reasons for avoiding pursuit altogether.

Much turns on the particulars of the individual case. Whether it is reasonable to expect compliance with the lesser norm will depend on the extent of the learning deficit and on what the individual has actually learned. If our hopeless romantic was completely insensitive to all reasons for and against pursuing particular romantic partners, then compliance with the lesser norm might be unreasonable too. Nevertheless, Yaffe thinks the lessons from this thought experiment are generalisable:

General Substitution Thesis: If a person suffers from a learning deficit which impedes or reduces their ability to recognise and respond to reasons for and against an act of type X, then it may be unreasonable to expect them to avoid performing all acts of type X, but it may be reasonable to expect them to avoid certain tokens of X.

Yaffe doesn’t formulate the thesis in this way, but I think this version, with its reference to the type/token distinction, captures what he is saying.


2. Addiction, Unreasonability and Substitution
So we come at last to the central topic of the paper: are addicts responsible for failing to comply with norms? The theoretical spadework has been done, and consequently this question becomes much easier to answer. What we need to know is: (i) whether addicts suffer from some learning deficit — similar to that of the hopeless romantic — which makes it unreasonable for us to expect them to comply with certain norms, and (ii) if this is true, whether there are nevertheless lesser norms which they can be held responsible for breaching.

With regard to the first of these matters, Yaffe is crystal clear: addicts do suffer from such a learning deficit. His argument can be set out in the following manner:

(1) If a person suffers from a learning deficit which impedes or reduces their ability to recognise and respond to reasons for and against some actions, then it is unreasonable to expect them to avoid such actions (because they would have to cede autonomy). (2) The dopamine signal is what allows people to attach values to outcomes and thereby recognise and respond to reasons for and against actions. (3) The dopamine signal is impaired in addicts in a such a way that their ability to recognise and respond to reasons for and against certain actions is impeded. (4) Therefore, it is unreasonable to expect addicts to avoid certain actions.

The argument for premise (1) has been laid out over the past two posts. I’ve expressed my doubts about it already but I will leave those to the side for now. The focus shall be on premises (2) and (3) respectively. One problem with these premises should be flagged at the outset. It is that Yaffe’s defence of both relies on the results of several neuroscientific studies and I have not evaluated those studies. What he says about the dopamine system sounds about right given my previous readings on it, but I’m afraid I’m his stuff on the effect of addiction on the dopamine system is unknown to me. I’ll simply assume he is fairly representing the evidence.

So let’s look at premise (2) more closely. As Yaffe notes, studies on the dopamine signal suggest that its function is to represent the value that a subject attaches to the difference between the value they associated with a particular outcome and the value they expected that outcome to have. Most such studies have been done on monkeys. If a monkey is shown a sign, and then a few moments later is given a reward, their dopamine signal goes up after receiving the reward. If this process is repeated, the monkey will learn to associate the sign with the reward. At this stage, the dopamine signal will goes up once the monkey perceives the sign, but will remain flat after the reward is received. What’s more, if the sign is perceived, but no reward is given, the dopamine signal will actually go down when the monkey realises no reward will be given. Thus, it does indeed appear to be true that the dopamine signal represents the difference between the value associated with an outcome and the expected value of that outcome.

The question is: why does such a mechanism exist in the brain? The answer, according to Yaffe, is that it exists to allow an organism to attach values to particular acts and outcomes in light of experience. In other words, the dopamine signal is what allows the organism to update the algorithm it uses to weigh the reasons for and against engaging in particular kinds of conduct. The key here is the notion of “updating”. Without that capacity for updating, organisms would be unable to recognise and respond to reasons for action. Thus, if the dopamine signal is indeed integral to this capacity, premise (2) would appear to be true.

This brings us nicely on to premise (3). Is it true that the dopamine signal is impaired in addicts? The answer appears to be “yes”. This is chiefly because many drugs work to artificially increase the dopamine signal, independently of the experienced and expected value of using those drugs. Cocaine is the classic example and the case of the cocaine user can be contrasted directly with that of the monkeys used in the studies on the dopamine system.

As I just mentioned, the monkey, when repeatedly presented with a sign followed by a reward, learns to associate that sign with the value of the reward. This results in their dopamine signal altering from one that it is initially raised by the receipt of the reward to one that remains flat after the receipt of the reward. The flattening of the dopamine signal being indicative of the fact that the monkey has learned what value to attach to that outcome, as well as the expected value of the events that precede that outcome. Contrast this with the experience of the cocaine user. Cocaine works by giving the brain an artificial dopamine boost. If a person repeatedly uses cocaine, they should learn to associate certain kinds of behaviour (drug-use related behaviour) with a pleasant outcome. And this should lead, over time, to the flattening of the dopamine signal arising after that outcome. This would indicate that normal evaluative-learning has taken place. But the effects of the drug prevent this from happening. Instead, when the user takes cocaine they get a higher dopamine signal than the one they expected. Their ability to recognise and respond to the reasons for and against drug use behaviour is impaired as a result.

Yaffe is keen to point out that this doesn’t just apply to cocaine users nor just to drug use behaviours. Studies have found that the impairment in the dopamine signal among addicts applies to thier non-drug-using behaviours too. Thus, it appears that premise (3) is true. And if this is so the argument outlined above would appear to be correct.

But that’s only the first piece of the puzzle — the unreasonable compliance thesis — what about the second one. Is it true that there is usually a lesser norm that the addict could be expected to follow? On this matter, Yaffe is rather more circumspect. He mentions the possibility of zealous/weak dichotomy in drug pursuing behaviours, which would make the case of the drug user and the hopeless romantic analogous. For example, a drug user could pursue drugs zealously, by killing and maiming in order to secure the drugs, or more weakly, by stealing money from friends and family. We could expect him to avoid the former but not necessarily the latter.

But this is just a possibility and the link between those kinds of cases, and the kinds of cases that might arise in the criminal law are sketchy. Yaffe acknowledges this deficiency. His general position is that addicts can be held responsible for whatever it is they can do without being forced to give up autonomy. But it is unclear what that would be in particular cases. More research needs to be done on the actual effects of drug use on specific kinds of learning.

Okay, we shall leave it there for now. I would recommend checking out some of Yaffe’s other work on addiction and responsibility. They can be found on his homepage.

Addiction and Responsibility (Part Two)

(Series Index)

(Part one)

This is the second part in a brief series on addiction and responsibility. The series is working off the article “Lowering the Bar for Addicts” by Gideon Yaffe. Yaffe defends two central theses in his article. They are:

Unreasonable Compliance Thesis: An addict is not responsible for violating a norm if in order to comply with that norm they must cede control of their behaviour to something that is independent of their decision-making capacities.

Substitution Thesis: In cases where standard norm-compliance would be unreasonable for the addict, there is usually a lesser norm that the addict can be held responsible for violating.

As part of his defence of these theses, Yaffe discusses alternative views on the relationship between addiction and responsibility. One of those views claims that the situation faced by the addict is akin to that faced by the person who is under duress. Both face an unreasonable burden if they comply with a norm and so both should be excused if they fail to comply. In part one, we identified some flaws in this analogy. In brief, there appear to be a number of significant structural differences between the situation faced by the addict and that faced by the person under duress.

However, Yaffe is keen not to throw the baby out with the bathwater. There is something about addiction that involves unreasonable burdens, and those burdens are relevant to responsibility. It’s just that the duress-analogy fails to specify exactly how this works. Yaffe has an alternative theory that does. In the remainder of this post, I’ll try to explain what it is.

The post is divided into three sections. The first introduces Yaffe’s concept of burdens of compensation and contrasts them with burdens of duress. The second part introduces some thought experiments that suggest that certain burdens of compensation are unreasonable and tries to specify a condition which, if met by such burdens, is sufficient to undermine responsibility. And the third part outlines Yaffe’s argument for this sufficient condition. This constitutes the theoretical core of Yaffe’s paper, all that remains to be done in the final part to this series is to explain how the sufficient condition might apply to addicts.


1. Burdens of Compensation
We all have certain innate foibles and defects that make it difficult for us to do certain things. Some of us have more of these defects than others. Usually, we accept that we have to undertake a burden to compensate or correct for these defects. Consider, for example, the man who finds it difficult to wake up and get out of bed in the morning. He has an important job and he has to make sure he is there on time. So, he finds it necessary to take steps to correct for his problem. This could consist of buying an alarm clock, getting his wife to pour cold water on his face, or something along these lines.
The man in this example faces what Yaffe calls a “burden of compensation”, which can be defined as follows:

Burden of Compensation (BoC): Is any step that an agent has to take in order to compensate for innate/intrinsic obstacles to norm-compliance.

A burden of compensation is importantly different from the kinds of burdens that arise in duress. Take the example once more of the person who is threatened with death or harm unless they agree to perform a burglary. The threat presents a burden that we would not expect the person to bear. But the burden in question would arise after the person complied with the norm, not before. This is distinct from a BoC, which arises before norm-compliance, not after. That is to say, in the case of a BoC, the agent must take on the burden in order to comply with the norm; whereas in the case of duress, the agent must breach the norm in order to avoid the burden. Whether this distinction is morally significant is questionable, but it is, at the very least, structurally different.

In addition to this, Yaffe argues that close analysis of BoCs reveal that there is sometimes a fuzzy boundary between saying that someone “Cannot do X” and saying that someone “Cannot be expected to do X”. The person who faced a BoC but did not undertake it could not have complied with the relevant norm because they then lack a certain capacity or ability. Thus, we might be tempted to say that they could not comply and so should be excused. But this is not quite right since they still had the opportunity to comply: if they had undertaken the BoC they could have complied. So what really seems to matter here is whether the burden in question was one they could reasonably be expected to bear.


2. What kinds of Burdens are Unreasonable?
Consequently, we need to figure out what kinds of burden are unreasonable. In the duress case, this might be straightforward enough: a burden is unreasonably high when the moral wrong (or harm) that would arise if one complied with the norm is greater than the moral wrong that arises if the norm is breached. So, for example, if someone is forced to commit a burglary in order to avoid their child being harmed or severely injured, we excuse them because having a child killed or injured is morally worse than allowing a burglary to be committed. There is a straightforward moral cost/benefit analysis taking place here.

A similar calculus might work in the case of BoCs, but Yaffe isn’t particularly optimistic about developing a general theory of unreasonable burdens. Instead, he is content to identify one condition which if satisfied is sufficient to undermine responsibility, without ruling out other possible conditions. The condition in question is a complex one, and so we need to take care in how we approach it.

Let's start with some intuition-pumping. Imagine there is a person with a severe mental impairment. This impairment reduces their ability to recognise and respond to reasons for complying with certain social, legal and moral norms. However, there is a way for them to overcome this obstacle to norm-compliance: they can cede all decision-making authority and control to another agent (e.g. by granting power of attorney, or something akin to this). But, as it happens, this person fails to cede decision-making authority and control to another person and thus ends up breaking some norm. Now imagine a second person, with normal mental faculties, who also breaches that norm. Which of the two deserves blame for failing to be norm-compliant?

The intuition is that the person with the severe mental impairment, even if they do not entirely escape blame, is much less blameworthy than the normal person. Why so? Well, the normal person presumably recognised the reasons for being norm-compliant but decided to ignore them, whereas the impaired person could not even recognise those reasons in the first place. There’s something intuitively worse about the former, which suggests there is something about the predicament faced by the latter that lessens the degree to which they are responsible.

But what is this? The answer, according to Yaffe, is that the impaired person is under an unreasonable BoC. They can only comply with the norm by ceding their autonomy to another agent or entity, and it is unreasonable, ceteris paribus to expect someone to renounce their autonomy in this manner. The ceteris paribus clause is significant here. On some occasions, ceding autonomy can itself be an exercise of autonomy. One thinks of Ulysses binding himself to the mast in order to avoid being lured away by the Sirens. In doing so, he compromised the autonomy of his future self, but he was still exercising autonomous control.

What’s more, it is sometimes reasonable to expect people to cede autonomy in this manner. Yaffe uses the example of the panicky soldier to make this point:

Panicky Soldier: D joined the army voluntarily and upon joining committed himself to following any orders given by his commanding officer. The problem is that D suffers from severe panic attacks. In the midst of a heated battle, he is apt to panic and flee (a breach of the military code). The only way D can overcome this is to focus on his commanding sergeant and copy whatever he is doing.

In this example, the soldier can only comply with the “Do not flee the battlefield”-norm by ceding his decision-making power to his commanding officer. However, since he voluntarily agreed to do this when he joined the army, it seems reasonable to expect him to do so via the specific mechanism (copying movements) identified here. In other words, the soldier seems to be under an independent normative obligation to cede his autonomy to his commanding officer — which was undertaken upon joining the army — and this independent normative obligation seems to make it reasonable to expect him to take steps to do just that when he’s on the battlefield.

Our feelings would probably be different if the soldier had been conscripted into the army, or had otherwise been forced to participate in the battle. For if this happened, there would be no independent normative obligation to cede autonomy to another, and hence nothing to justify the reasonable expectation.


3. Ceding Autonomy and the Argument from Blame
With the lessons from the preceding thought experiments under our belts, we can now proceed to outline the condition which, if met, is sufficient to undermine responsibility:

Ceding Autonomy Principle: If S, in order to do A in circumstances C, would have to give up his power to decide whether or not to do A in C on the basis of appreciation of the reasons for and against A-ing in C, and if S is not under independent normative pressure to give up his autonomy in this way, then S is not responsible for failing to A in violation of a norm requiring agents to A in C.

That’s quite a mouthful, but you can blame Yaffe for this since it’s his wording (the name is mine). I can live with the wording of it though, what I’m more interested in is the argument for this principle. Why is it that ceding autonomy is an unreasonable burden to bear?

Yaffe’s argument is based on the nature of blame and punishment. He argues that in blaming and punishing an agent for failing to comply with a norm we are, in effect, trying to give them an additional reason for not having acted as they did. (I think this is mildly controversial, but let’s run with it for now.) Next, he argues that in order for blame and punishment to make sense (or, maybe “to be justified”) the kind of reason-for-action that they supply must be one that would have worked at the time of the original norm-violation. But, of course, if someone has to cede their autonomy to another in order to comply with a norm, they would not be receptive or responsive to reasons-for-action at the time in question. After all, they have some deficit that prevents them from being receptive and responsive. Thus, blaming and punishing them for failing to comply with the norm is nonsensical/unjustifiable.

I have some doubts about this argument, but I’m going to try to formalise it as best I can before I say anymore:

• (1) It is only justifiable/sensible to blame and punish X for failing to comply with a norm if, at the time at which they failed to comply with the norm, they could have been receptive to the reason for compliance that is supplied by the blame and punishment. 

• (2) If X had to cede autonomy to another in order to comply with a norm, then they could not have been receptive to the reason for compliance that is supplied by the blame and punishment.

• (3) Therefore, if X had to cede autonomy to another in order to comply with a norm, it is not justifiable/sensible to blame and punish X them for failing to comply.

As you can see, I’ve equivocated here between whether the blame and punishment is “sensible” or whether it is “justifiable”. I do this because Yaffe uses the phrase “makes sense” in his discussion, which suggests to me that he is making a conceptual argument, but I think it would be better to construe the argument as a normative one, so I've substituted in the term “justifiable”. That said, there may not be a huge difference between conceptual and normative arguments in this instance. If the upshot of the conceptual argument is that it is inappropriate to apply certain terms or concepts in the specified cases, and if those concepts describe inherently normative practices, then it may just amount to saying that it is unjustifiable to carry out those practices in the specified cases.

That’s a somewhat technical point. A more interesting, and ultimately more significant, point is this: there’s something suspicious about the account of blame and punishment that Yaffe is adopting here (premise (1)). In particular, it seems that in saying that blame and punishment must provide reasons for norm-compliance that would have been operative at the time of norm-violation, Yaffe is relying implicitly on a retributive conception of blame and punishment. This is controversial.

According to retributive theories, it is right and proper to blame/punish an individual if and only if they have engaged in culpable wrongdoing. Which is to say, what matters from the perspective of retributive punishment is how an agent actually violated a norm. Thus, this is a necessarily historically-oriented justification for punishment. This fits well with the account of blame and punishment offered by Yaffe. If blame and punishment are supposed to provide reasons for norm-compliance, and if punishment is justifiable only by reference to what happened in the past, then it makes sense to demand that the reason for compliance would have been effective in the past.

But a historical theory of blame and punishment like retributivism is not the only game in town. If those practices are justified on consequentialist or utilitarian grounds, then they become future-oriented. And if they become future-oriented, one wonders whether the demand that the reasons for action that they supply be operative in the past is needed.

To be clear, I’m not entirely sure about this line of criticism. It’s possible that Yaffe’s argument could be repaired or refined in some way that allows it to encompass the competing theories of blame and punishment. It’s also possible that the ceding autonomy principle could be defended in other ways. Who knows? From now until the end of this series we’ll assume it’s correct and turn our attention instead to whether it applies to drug addicts. That’s a topic we’ll take up in the next post.

Addiction and Responsibility (Part One)

When it comes to assessing the impact of addiction on responsibility, there’s a longstanding debate between two schools of thought. The first, which we might call the “Moral Failing” school, holds that addiction does nothing to lessen responsibility: an addict is someone who exhibits a lack of moral virtue, and so they should not be excused from responsibility. The second, which we might call the “Disease” school, holds that addiction is an action-related disease and, as a consequence, we cannot hold addicts responsible for what they do.

In a recent article entitled “Lowering the Bar for Addicts”, Gideon Yaffe argues that both schools of thought are wrong. Addicts are neither completely responsible, nor completely absolved. Instead, he argues that there is a middle ground, which holds that an addict “should be given some kind of break but not excused entirely.” In essence, this middle ground consists of two theses (the names are mine, not Yaffe’s):

Unreasonable Compliance Thesis: An addict is not responsible for violating a norm if in order to comply with that norm they must cede control of their behaviour to something that is independent of their decision-making capacities.

Substitution Thesis: In cases where standard norm-compliance would be unreasonable for the addict, there is usually a lesser norm that the addict can be held responsible for violating.

The first thesis is so named because Yaffe’s argues that the reasonability of the burdens associated with norm compliance plays a key role in our assessments of responsibility. The naming of the second thesis should apparent.

Over the next few blog posts, I want to outline Yaffe’s defence of these two theses. That defence comes in four parts. The first part highlights the importance of reasonable expectations and burden-sharing in assessments of norm-compliance and responsibility. The second part notes how people are sometimes expected to take on the burden of correcting for their defects, and how a failure to meet that burden can lead to the ascription of responsibility. The third part specifies a key condition that makes it unreasonable for an agent to take on the burden of correcting for their defects. And, finally, the fourth part argues that addicts meet that condition. This then leads to the conclusion that the two theses mentioned above are correct.

In the remainder of this post, I will focus on the first part of Yaffe’s defence.


1. Duress and Unreasonable Burdens
An argument that is sometimes mooted is that an addict’s responsibility is undermined because they are under duress. This raises the obvious question: what is duress? Put simply, duress is a defence to a criminal charge. Duress arises when a person is put in a situation in which they are forced to choose between violating a norm (i.e. commit a crime) and doing something that seems to place an unreasonable burden on them. In practice, it is quite difficult to prove, but it is still generally recognised as having the power to excuse.

An example will be helpful at this point. In the English case of R v. Hasan (2005) the defendant pleaded duress to a charge of burglary. He alleged that a violent drug dealer had threatened to harm both himself and his family if he did not carry out this burglary. This is a classic example of duress: D, or someone close to D, is threatened with being severely harmed unless D commits a crime. As it happened, the defence was not successful in this case. The court found that D had voluntarily associated with the drug dealer and would reasonably have expected to be threatened or compelled by them at some point. In other words, the court found that if you voluntarily put yourself in a situation in which you are likely to be placed under duress, you cannot avail of the defence.

The court’s conclusion might provide an interesting lesson for those who argue that addicts are under duress, but we shall not dwell on that here. There are other, deeper, problems with that view. To see what these are we need to map out the basic structural features of duress. Basically, duress arises when the following conditions are met: (i) a person is (involuntarily?) confronted with a decision problem; (ii) in that decision problem they have a choice between complying with a norm, and not complying with a norm; and (iii) there is a heavy, arguably unreasonable, burden attached to compliance with the norm. So even though the person had a choice of whether or not to comply, we think they are excused for their non-compliance because of the unreasonable burden. This gives us the following principle:

Duress Principle: If D is faced with a choice between norm-compliance and non-compliance, and if there is a heavy, unreasonable burden associated with compliance, they can be excused for failing to comply with the norm.

This principle isn’t perfect, by any means. I think it would need to be qualified and refined in a number of ways before it became satisfactory. However, for the time being, let’s assume its accurate and see what follows. For ease of analysis, the structural elements of duress are represented in the diagram below. The diagram uses the case of R v. Hasan for illustrative purposes, but ignores the issue of voluntary association that arose therein.

2. Addiction and Duress
From the preceding analysis, it would seem that in order for the claim that addicts are under duress to be borne out, we would need to locate some heavy burden that makes norm-compliance unreasonable for them. What might this be? Yaffe considers the suggestion that the burden in question is that of withdrawal. In this context, “withdrawal” is a general term given to a complex of symptoms experienced by addicts once they withdraw from consumption of their preferred substance. If we compare non-addicts to addicts, we find that the latter are (sometimes) faced with a choice between norm-compliance plus withdrawal, and non-compliance, whereas the former are not. For instance, a heroin addict might be forced to steal in order to feed their addiction; they have the option of not-stealing, but if they chose it they would be forced to bear the burden of unpleasant withdrawal symptoms. Non-addicts have no similar burden attaching to compliance.

This leads to the following argument:

• (1) If D is faced with a choice between norm-compliance and non-compliance, and if there is a heavy, unreasonable burden associated with compliance, they can be excused for failing to comply. 

• (2) In some situations, addicts are faced with a choice between: (a) complying with a norm and undergoing withdrawal symptoms; and (b) not complying with a norm and avoiding withdrawal symptoms. 

• (3) Withdrawal symptoms are a heavy, unreasonable burden to bear. 

• (4) Therefore, in some situations, addicts can be excused for failing to comply with norms.

Is this a plausible argument? Is an addict really under duress because of the burden of withdrawal? Yaffe detects three flaws in this line of reasoning.

The first has to do with the nature of withdrawal symptoms. As Yaffe notes, these vary greatly depending on the substance to which a person is addicted. For instance, a common methamphetamine withdrawal symptom is sleeplessness, while a common cocaine withdrawal symptom is craving. No doubt both are unpleasant, but are they really unreasonable burdens? Would we not expect people to comply with norms even if such unpleasantries were attached to compliance? Furthermore, and this is the angle of attack Yaffe prefers, the variation in symptoms is greater than the variation in beliefs about the responsibility-undermining potential of addiction. In other words, people seem to think that addiction per se undermines responsibility, and they do not seem to pay attention to the specific nature of the withdrawal symptoms associated with different kinds of addiction. Thus, to claim that withdrawal symptoms explain or justify why addiction might undermine responsibility seems wrong.

The second problem is that addiction seems to affect responsibility even when an addict violates a norm that has nothing to do with drug use. He gives the example of a heroin addict who has a $30 a day habit, and who calls on family favours to secure himself a job that pays $130 a day (not too shabby for a heroin addict). However, he cannot perform his duties at work competently or efficiently, and he is fired. He has violated a norm (perhaps a social one, but also a legal/contractual one). Yaffe suggests that his addiction surely has some role to play in this, and although we may think he still deserves to be fired, it nevertheless seems to affect the degree to which we hold him responsible for being fired. In this example, addiction plays a part in our responsibility assessments, but that part has nothing to do with the fact that the addict faced a stark choice between non-compliance and withdrawal. Indeed, quite the opposite since the job would have helped him feed his habit.

The third problem is that to say that an addict is under duress is a mischaracterisation of how they make decisions. If the addict chooses to violate a norm in order to (say) steal money to feed their addiction, they are not acting so as to avoid withdrawal symptoms. Instead, they are acting because they enjoy, or derive pleasure (or simply need) the drugs, “avoiding withdrawal symptoms” does not really figure among their motivations. This stands in stark contrast to the man who commits a burglary in order to avoid harm to himself or his family. There, the primary motivation for non-compliance comes from the heavy burden of compliance; in the case of the addict, the primary motivation comes the attractive features of non-compliance. There is something Frankfurter-esque in this line of reasoning, and it chimes with certain intuitions and theories we have about what makes someone responsible. Roughly, how they perform an act seems to count for more than what might have happened if they failed to perform the act.

If this right, then this duress-based analysis of addiction and responsibility fails. Reasonable expectations and burden-sharing play important roles in how we understand responsibility, but addicts cannot be excused simply because they are under the burden of withdrawal symptoms. If they are to be excused, some other rationale must be sought. We’ll find out more of Yaffe’s thoughts on this rationale in the next post.

Solum on Semantic Originalism

This post is slightly unusual, covering as it does a slightly recondite issue in legal philosophy rather than something in ethics/religion which is the norm for this blog. Nevertheless, it has great personal significance for me since I teach legal philosophy and work in a law school. Also, I think it is an interesting topic irrespective of its recondite nature.

So what is this topic that I speak of? Many living in the US will be aware of the ongoing, often-heated, debate over the correct interpretation of the constitution. Since roughly the late 1970s (although one can find precursors before that) there has been a movement calling for an originalist interpretation of the constitution. This movement is typically associated with the conservative side of the political debate, although more recently some liberals have endorsed a form of originalism (Jack Balkin being the exemplar of this).

The doctrine of originalism has gone through a number of transformations over the years. It started off being about interpreting the constitution in light of the intentions of the framers, before eventually settling on its most stable and popular form which calls for interpreting the constitution in light of its original public meaning. By far the most sophisticated and compelling defence of this brand of originalism can be found in the work of Lawrence Solum.

Solum defends something he calls “semantic originalism” which is a partly normative, partly semantic interpretive theory. In this post, I want to outline some of his basic arguments for this position. I must confess, however, that I will be relying on a slightly dumbed down version of those arguments, which can be found in Solum’s debate-book with Robert Bennett. The full version of Solum’s argument can be found in his paper/monograph “Semantic Originalism”. I doubt I’ll cover that on the blog though, since it is quite a dense piece of work (in the best possible sense of the word “dense”).

The post will be divided into three parts. The first will outline the four central theses of Solum’s semantic originalism and will present the basic argument for that position. As will become clear, the theory has a normative and a semantic component to it. The remainder of the post will focus on the semantic component. To this end, the second part will look at one of Solum’s arguments for this component, something I dub the constitutional communication-argument. While the third part will look at another argument, which I dub the absurdity argument.


1. Semantic Originalism
Solum’s unique brand of originalism centres on four interlocking, though distinct theses. They are:

The Fixity Thesis: The semantic content of the constitution is fixed at the time of ratification. 

The Clause-Meaning Thesis: The semantic content of the constitution at the time of ratification was the original public meaning or original conventional meaning of the clauses contained within. (There are four qualifications of this thesis which I shall not discuss at length here, they cover things like the possibility that new terms were created by stipulation within the constitution, and that legal terms of art were adopted by the constitution.) 

The Contribution Thesis: The semantic content of the constitution contributes to, but does not fully determine, the application of the law. 

The Fidelity Thesis: We have good reasons to affirm, or be loyal to, the semantic content of the constitution.

Solum qualifies and develops each of these theses at considerable length. Of necessity, I will be much briefer. Broadly speaking, the first two theses constitute the “semantic” component of Solum’s theory because they both make claims about the meaning of the clauses contained within the constitution. By contrast, the latter two theses would appear to constitute the “normative” component of Solum’s theory because they make claims about the kind of legal order we ought to be committed to.

Admittedly, the contribution thesis is not obviously normative in nature. It seems more to be making a point about how the law works: legal rules have a semantic content which is important, but which does not fully determine the kinds of institutions or legal structures we should construct out of those rules. This is because meaning is ambiguous, open to multiple applications, depending on the circumstances. Despite this, I take it that the contribution thesis is still part of the normative case for originalism because it tries to reassure certain critics of that position. Critics will argue that originalism, if taken seriously, will turn the constitution into a straitjacket (or a “suicide pact”), something which cannot be sustained in the face of a ever-changing and evolving society. But the contribution thesis suggests that this need not be the case. Because there is a gap between the interpretation of the constitution (which depends on semantic content) and the application of the constitution (which depends on interpretation and more), there is some room for flexibility and responsiveness. This gap is something that Jack Balkin has run with in his book Living Originalism, which tries to square the circle between originalism and living constitutionalism.

Combining the semantic and normative components, we get a very simple argument in favour of semantic originalism:

• (1) We ought to interpret the constitution according to its semantic content. 

• (2) The semantic content of the constitution is fixed by the original public meaning of its clauses. 

• (3) Therefore, we ought to interpret the constitution according to the original public meaning of its clauses.

I shan’t discuss the defence of premise (1) in this post. The reader is free to speculate and comment as they see fit. I shall, however, discuss two of Solum’s arguments in favour of the second premise. The first of those arguments comes from considering the nature of constitutional communication. The second is effectively a reductio ad absurdum of the leading alternative theory, living constitutionalism. Both are taken from pgs. 13-17 of Constitutional Originalism.


2. The Constitutional Communication Argument
Laws can be viewed as speech acts, as attempts to do things with words. After all, a law is a written or otherwise symbolically-encoded rule that sets out demands, imperatives, commitments and so forth. Speech acts have conditions of success associated with them. Those conditions of success relate to both the semantic content of the act, as well as its purpose (or, to use a fancier term, its illocutionary point).

Consider an ordinary conversational communication between two people. The first, the Speaker, is trying to perform a speech act. In this instance, the speech act is a request. The Speaker asks the Listener to “Pass the chalk”. In order for the request to be successful, the Listener must understand both the point of the speech act and as well as its semantic content. How is this possible? The standard Gricean analysis holds that this is possible when the listener understands the speaker’s intentions. In other words, when the speaker refers to “chalk”, the listener knows which object in the world the speaker is referring to; and when the speaker says “pass me”, the listener knows which actions he or she is expected to perform. Furthermore, in order for the request to be fully successful, there must actually be chalk in the room that the listener can pass to the speaker. So the Gricean analysis of meaning effectively reduces meaning to knowledge of a speaker’s intentions.

This has an interesting consequence when applied to the ordinary conversational context. Obviously, words like “chalk” and “pass” have conventional meanings — meanings that are understood and recognised by most people in society. The speaker in our hypothetical conversation relies upon those conventions in making his/her request. However, in that setting, the conventional meaning is not necessary for successful communication (though it may often be sufficient). Imagine, for instance, that the request for chalk took place in a room with a whiteboard and a marker, but no chalk. In that case, the listener would probably be able to guess that the speaker really intended for them to pass the marker (they could ask a clarifying question if needs be). So the mere fact that the conventional meaning of “chalk” didn’t apply to the actual situation in the room, would not imply that a successful communication could not take place. All that matters is knowledge of speaker’s intentions.

So much for the ordinary conversational context. What about the context in which legal speech acts are made? Solum argues that this context is so radically different from that of everyday conversation that the Gricean analysis cannot apply. Legal texts, such as constitutions, have multiple authors (multiple speakers) all of whom may have had different, perhaps inconsistent, intentions in mind when they performed the speech act (i.e. drafted and ratified the constitution). Furthermore, unless they make some public record of their intentions, speculation and inference will be the only way to figure out what they may have intended. Both of these features makes it very difficult for listeners to know what the speaker’s intention was in the case of legal texts. What’s more, the authors would have known that this was the case, so they couldn’t have honestly expected people to interpret those texts in light of their intentions. They would have expected them to focus on original public meaning.

Another analogy, proposed by Solum, can be used to underscore this point and to endorse the originalist approach to interpretation:

Suppose you wanted to send a message in a bottle and thereby successfully communicate with an unknown reader, perhaps in a distant land generations from now. You couldn’t rely on the reader’s knowing anything about you, your intentions, or the context in which you wrote the message. You would have to rely on the plain [i.e. public/conventional] meaning of the words you used and the rules of English syntax and grammar. Of course, those meanings and rules might change over time, so it would be a good idea for you to date your message: if the reader were interested enough, he or she could check his or her assumptions about the plain meaning of your text against historical evidence of linguistic practices (pg. 14-15).

The point is that legal texts, like constitutions, are akin to messages in a bottle. They are speech acts, but when they are made, their speakers cannot rely on knowledge of speaker’s intention when trying to make a successful communication. Instead, they must rely entirely on the conventional public meaning of the language used, at the time of the speech act. All of which suggests the following argument in favour of premise (1):

• (4) In order for a speech act to be successful (i.e. in order for its listeners to understand its content and its point) it must either: (a) rely on the conventional meaning of its contents at the time it was spoken; or (b) rely on the listener having epistemic access to the speaker’s intentions. 

• (5) Legal speech acts (such as constitutions) have multiple authors (speakers) and an unknown and oftentimes temporally distant set of interpreters (listeners). 

• (6) If a speech act has multiple speakers and an unknown and temporally distant set of listeners, it cannot depend on the listeners having epistemic access to the speakers’ intentions.

• (7) Therefore, in order for a legal speech act to be successful, it must rely on the conventional public meaning of its contents at the time it was spoken.

This gets us to premise (2), with some relatively minor, and easily reparable, slippage between (7) and (2). As you can see, the argument is normative in the sense that it assumes we want to have successful speech acts, but it is still primarily about what determines the semantic content of different kinds of speech act (in any event, the line between semanticity and normativity is a thin one and, somewhat ironically, how thin it is depends on what one understands by the term “normative”).


3. The Absurdity Argument
The preceding argument offers some reasonable grounds for thinking that the semantic content of a legal text depends on conventional meaning. This provides support for Solum’s clause-meaning thesis. But snuck into the argument at the last moment was support for fixity thesis too. While that support may be sufficient, additional support can be derived from a reductio ad absurdum of the opposing view, namely: that the meaning of a term in a text changes along with any changes in the conventional meaning of that term. Call this the “living meaning”-thesis.

Solum has a nice example he uses to motivate this reductio. It concerns the use of the word “deer”. Apparently, in the 12th C, the word “deer” was a general term used to refer to any four-legged mammal, not just the mammals that we would nowadays call “deers”. That is to say, the conventional meaning of the word “deer” has evolved over time. Now imagine that we have a text dating from the 12th C that uses the word “deer”. How are we, in the 21st C to interpret that text? If we adopted the living meaning-thesis, we would say that the text was referring to the kinds of four-legged mammal beloved by Santa Claus. But this would surely be absurd, or so the argument goes.

I should say: I have no idea whether this example correctly represents the linguistic conventions of the 12th C. I assume it does, but in some ways it doesn’t matter. Solum has another, purely hypothetical example, which also suggests that living meaning, at least when it comes to historical texts, is an absurdity. Consider the term “freedom of speech”. This has a broadly accepted connotation in politics and law right now. But suppose, in the future, linguistic practice shifts, and the term “freedom of speech” becomes street slang for a forced confession. One would hear criminals conversing with one another talking about how a police officer “freedom of speeched” them, and so forth. If we (in the future) have legal texts referring to “freedom of speech” that predate this shift in linguistic practice, should we interpret them to be referring to “forced confession”? Again, to suggest that we should seems absurd.

We can summarise these thoughts as a simple argument. The argument mentions only the deer-example but could easily be reformulated to include the freedom of speech one. It runs as follows:

• (8) In the 12th C the word “deer” was conventionally understood to refer to all four-legged mammals; over time the conventional meaning has shifted to a particular subset of four-legged mammals. 

• (9) Suppose the living meaning thesis is correct. 

• (10) Then we would have to assume that a 12th C text using the word “deer” was not referring to all four-legged mammals but only the particular subset that we nowadays apply that label to. 

• (11) It is absurd to interpret a 12th C text in this manner. 

• (12) Therefore, the living meaning thesis is false. 

• (13) If the living meaning thesis is false, the fixity thesis must be true. 

• (14) Therefore, the fixity thesis is true.

Although premise (13) looks slightly dodgy here — in that it assumes there are only two possible theories — I still find this argument mildly compelling. However, this may be due to the simplicity of the particular example. I doubt there is much reason to dispute this kind of interpretation of the word “deer” from a 12th C text, just as there isn’t much reason to dispute the interpretation of the phrase “domestic violence” in the US Constitution (hint: it doesn’t mean what you might think it means). The real controversy arises with the more general, evaluative and ambiguous phrases like “cruel and unusual punishment”. No doubt there were a limited number of things that the public to whom the Constitution was originally addressed deemed “cruel and unusual”, but how they might have applied that phrase isn’t quite the same thing as the meaning they attached to it. The meaning is quite abstract and general; the application might be narrow and context-specific.

In fairness, Solum is admirably aware of this point. Indeed, this is exactly what his contribution thesis is getting at: to say that the semantic content of the constitution was fixed at the time of its ratification, is not to say that the expected legal applications of that content, at the time of ratification, should fully determine the law. We can apply terms like “unreasonable search and seizure” to new technologies and circumstances, ones maybe not envisaged by the original public, because the semantic content of those terms is broader than the original expected application. That must be borne in mind at all times when evaluating the semantic originalism defended by the likes of Solum.

Okay, so that’s it for this post. Hopefully, this has been a decent enough overview of semantic originalism and the basic arguments in its favour. If you want to learn more, I would recommend the book Constitutional Originalism or Solum’s article “Semantic Originalism”.

The Epistemic Objection to Torture (Part Two)

(Part one)

This is the second part in my series on the epistemic objection to the use of torture. The objection is derived from the work of Roger Koppl on epistemic systems. In part one, I formulated the argument and laid out Koppl’s defence of its first premise (note: Koppl uses mathematical modelling to make his case, I tried to simplify things with some elementary logic and decision theory). In this part, I will discuss the second premise of the argument.

For ease of reference, I will restate the epistemic argument here:

• (1) Torture can only be an effective and reliable means of obtaining information if: (a) torturers can recognise the truth once it is spoken; and (b) suspects undergoing torture believe that the torturer will stop once they speak the truth. 

• (2) Conditions (a) and (b) are unlikely to be met in empirically plausible cases. 

• (3) Therefore, torture is unlikely to be an effective and reliable means of obtaining information.

• (4) Torture is only permissible (in ticking bomb scenarios) if it is an effective and reliable means of obtaining information. (5) Therefore, torture is impermissible (in ticking bomb scenarios).

Now, we see more clearly the focus of this post. We must ask: why is that torturers are unlikely to be able to recognise the truth once given and why are suspects undergoing torture unlikely to believe that they can?


1. A Simple Defence of Premise 2
Interestingly, Koppl never provides an explicit defence of premise (2). Instead, he defends it from a series of counterexamples. I presume this is because he either thinks the defence of premise (2) is obvious, or he thinks that if it can fend off counterattacks it will be deemed plausible. He may be right about this, but I still think it’s worth pausing briefly to consider how one might provide a positive defence of premise (2).

The defence would, I believe, run something like this. If a torturer could reliably distinguish true answers from false ones, they would probably not need to threaten the use of torture. Why not? Because then they would either know the truth, or have narrowed down possible truths to such an extent that they could check themselves (without getting the information from the suspect) or could use other less severe threats to extract the truth. It is only because they don’t know what the correct answer is, and have no other means of finding out, that torture becomes an option.

What’s more, the suspect would presumably know that the torturer was in this epistemic predicament. They would reason to themselves that the torturer would not be threatening to torture them if they could reliably tell the truth from a lie. Hence, they would be encouraged to follow the decision-theoretical reasoning I laid out in part one.

In other words, the following two propositions seem likely to be true and conjointly they give us reason to endorse premise (2) (numbering follows from part one):

• (12) If the torture knew or could recognise the truth, they would not probably not need to torture. 

• (13) If a suspect is threatened with torture, then they would know (or have strong reason to believe) that the torturer could not recognise the true answer once given.

Thus, it seems that in practice Koppl is right: the two conditions needed for epistemic success would probably not be met. One can understand the argument here as presenting something akin to a common knowledge problem. If the suspect can be expected to know that we have no way of telling the truth from a lie, then we should know that he will know this (and vice versa ad infinitum). Hence, we would know that we could gain no epistemic advantage by using torture.

There are, however, two potential objections, both discussed by Koppl. The first, and most interesting, objection concerns the scenario in which there are two (or more) suspects with access to the relevant information, both of whom can be threatened with torture. The second, and less interesting objection, concerns the possible use of feedback in the torture process. Let’s consider both in turn.


2. Truth Recognition in the Two-Person Scenario
There is an old logic puzzle involving two guards standing in front of two doors, each leading to a different location. One of the guards always tells the truth; the other always tells a lie. You do not know which guard is which and you need to find out the location behind each door. Is there anyway that you can do this? Yes there is. You can simply go up to one of the guards and ask him: what will the other guard say if I ask him what’s behind his door? I won’t explain exactly why this works now, but you can read about it here, or view a video about it here. What’s interesting about this puzzle is that it points to a scenario in which the truth is seemingly concealed from us, but in which the truth can nevertheless be revealed by carefully analysing the logical structure of the scenario.

Could something similar be true in the case of torture? In other words, could it be that, despite the epistemic impediments outlined above (and in part one), there is some way in which we can bring the truth to the surface? Some people argue that there is, provided we have two or more suspects available to us, both of whom have access to the truth. To understand how this might work, we need to introduce a new concept called a message set:

Message Set: This is the set of credible messages (i.e. the set of believable lies and truths) that the torture victim could “send” to the torturer.

Now, imagine that we have two suspects in custody, both of whom have access to the truth we wish to know, and whose message sets only overlap on this truth. In other words, the union of the message sets has only one member: the truth (although, admittedly the truth may consist of more than one message).

If we were presented with two such suspects, then we could indeed use torture as means of eliciting the truth. How so? Well, we could simply continue to torture them until they both gave us the same answer. We would then know that this had to be the truth. If you bear with me for a moment and assume that it is likely for us to run into two (or more) such suspects, we can make the following argument against premise (2):

• (14) If a torturer had two or more suspects in custody, both of whom had access to the truth, but whose message sets only overlapped on the truth, then he could torture them both until they gave the same response and he would know that this response was the truth (i.e. the torturer would be able to independently verify the truth). 

• (15) If the suspects could be made aware that this was the nature of the scenario, they would have reason to believe that the torturer could reliably distinguish the truth from a lie. 

• (16) The antecedents of the conditionals expressed in (14) and (15) are likely to be true in empirically plausible circumstances. 

• (17) Therefore, (2) is false.

3. Objections to the Two-Person Argument
Thank you for bearing with me as I wrote that out. No doubt, a whole raft of objections occurred to you as you read through it. Don’t worry, I spotted (at least some of) them too. Let’s run down the list of objections now.

The first worry is in my characterisation of premise (14). It is not just that we need to have two suspects with messages sets that only overlap on the truth, we also have to know that we have two such suspects. That is to say, it’s not enough for this to be the case, we also have to know that it is the case. This is similar to the logic puzzle involving the two guards. In order for the solution to work, we have to know that one guard always tells the truth and the other always tells a lie.

This knowledge-requirement has a direct knock-on effect on the plausibility of premise (16). After all, no matter how unlikely it is that we have in our custody two suspects, from the same terrorist cell, who have interacted in such a manner that their messages sets only overlap on the truth, it is surely even more unlikely that we would actually know that we had two such suspects in custody.

There are other objections to (16) as well. Koppl mentions several. One obvious one is that terrorist suspects are more likely to have pre-coordinated on a common lie before being taken into custody. This is true even if they operate within a cell-structure which seals them off from certain information. Why so? Because even within that system those who are higher-up in the structure could filter some agreed common response down to the underlings within the cells. Indeed, they are more likely to do this if they are aware of the possibility of the double-torture scenario. So even if the two-person situation arose once and was successful, it would be unlikely to work again in the future as the terrorist cell adapts to the problem.

Finally, even supposing that the terrorists could not pre-coordinate on a common response, it is possible that they could give a common lie in response to being tortured. This could happen if there is some common non-truthful message in the message sets that forms a “Schelling point”. A Schelling point is effectively an attractor point that two independent, non-colluding agents are likely to coordinate on. It was introduced by the game theorist and strategist Thomas Schelling in the mid-20th Century. Schelling proposed that two travellers who had agreed to meet each other in New York on a particular day, but who had failed to agree a time and place for this meeting, might independently arrive in Grand Central Station at 12 noon. This is because that place and time forms a common attractor point for travellers to New York. Is it possible that there is something akin to “12 noon, Grand Central Station” in the message set available to potential terrorist suspects? Arguably it is, and arguably this would be more likely than there being two terrorists with message sets that only overlap on the truth.

Combined, these objections seem to throw the two-person defence of the epistemic efficiency of torture into some doubt.


4. The Feedback Objection
The two-person objection to premise (2) was a little intricate; the feedback objection is much more straightforward. It argues that torturers, in empirically plausible scenarios, do have ways in which to independently verify whether the suspect is giving them truthful information. Quite simply: once an answer has been given to them, they can stop torturing the suspect, and go and check whether the answer is correct (remember: we are talking about ticking bomb scenarios where the relevant information is going to be a bomb-location or something along those lines). If it turns out to be correct, the torture will cease; if it turns out to be false, they can recommence torturing the suspect.

Under those conditions, the torturer should be able to (a) recognise the truth and (b) convince the suspect that they will stop torturing them once the correct answer has been given. And assuming independent verification of this sort would be quite normal in empirically plausible circumstances, we get the following argument:

• (18) If a torturer can independently verify the answers given to them by the suspect (e.g. by checking locations for ticking bombs), then they could distinguish the truth from a lie and, what’s more, they could convince the suspect that they had this ability. 

• (19) Independent verification of answers is possible in most empirically plausible scenarios.

• (20) Therefore, premise (2) is false.

Two responses to this argument can be mentioned here. The first is that the time pressures are such in ticking bomb scenarios that checking multiple locations is not practical. In other words, it may be that you only really get “one shot” at correctly identifying the location. If the suspect was aware that the time pressures were of this sort, then the argument would not go through (i.e. premise (19) would be rebutted).

In a similar vein, Koppl argues that many terrorist organisations will function in a fluidic and responsive manner. Once they know that a member of theirs has been arrested and is likely to be tortured for relevant information, they could quickly change the proposed location of their attack. Thus, whatever information was available to the suspect would quickly become obsolete (this argument would also work against the two-person scenario). I’m not quite sure what to say about this. Many of the most famous terrorist attacks (e.g. 9/11 and 7/7) did involve some partial last minute changes of plan, but this was usually due to unforeseen obstacles or accidents. Still, this would seem to be enough to suggest that terrorist groups can operate in the fluidic and responsive manner envisaged by Koppl. The key question then is how likely it is that members of the terrorist group would know that one of their own had been arrested. After all, if they don’t know that their plan might be foiled by someone divulging relevant information they can’t be responsive to this possibility. Whether they have such knowledge or not seems like something that would vary greatly from case-to-case.


5. Conclusion
To sum up, the epistemic objection to torture undercuts one of the key assumptions of those who would defend its permissibility, namely: that torture is an effective an reliable means of obtaining information. In order to work, the proponent of the objection needs to show that, in empirically plausible cases, (a) it is unlikely that a torturer would be able to recognise when a suspect is giving them a truthful answer and (b) a suspect would be unlikely to believe that the torturer had such an ability.

In this post, we considered a prima facie defence of this claim, as well as two responses to it. Both of these responses were found to be lacking, though some version of the feedback argument might be plausible, depending on the case.

There is one final point I want to address here. This relates back to premise (4) of the epistemic argument against torture. As you recall, this premise stated:

• (4) Torture is only permissible (in ticking bomb scenarios) if it is an effective and reliable means of obtaining information.

I think there might be some reason to doubt this. In the kinds of “high stakes” scenarios envisaged in the torture debate — i.e. scenarios in which some disastrous consequence is going to arise unless the information is obtained — it’s at least arguable that torture need not be effective or reliable in order to be permissible. It could be argued that as long as there is some small chance that torture would elicit the relevant information, it is worth giving it a shot, given that otherwise a disastrous consequence is going to occur. In other words, it could be that it is the proponent of the epistemic objection, not the proponent of the permissibility of the torture, whose argument rests on a faulty assumption.