Hostname: page-component-74656f9884-j7cmm Total loading time: 0 Render date: 2026-11-06T01:16:51.258Z Has data issue: false hasContentIssue false
  • English
  • Français

Expectations and the Limits of Legal Validity

Published online by Cambridge University Press:  06 March 2015

ANDREW T. FORCEHIMES
ANDREW T. FORCEHIMES*
Affiliation:
Vanderbilt Universityandrew.t.forcehimes@vanderbilt.edu
Rights & Permissions [Opens in a new window]

Abstract

Drawing on the work of Jeremy Bentham, we can forward a parity thesis concerning formal and substantive legal invalidity. Formal and substantive invalidity are, according to this thesis, traceable to the same source, namely, the sovereign's inability to adjust expectations to motivate obedience. The parity thesis, if defensible, has great appeal for positivists. Explaining why contradictory or contrary mandates yield invalidity is unproblematic. But providing an account of content-based invalidity invites the collapse of the separation between what the law is and what the law ought to be. Grounding formal and substantive invalidity in a unified source – the sovereign's inability to adjust expectations to motivate obedience – allows us to avoid bringing in any additional apparatus that might compromise this separation. This essay fleshes out and defends the parity thesis.

Information

Type
Research Article
Information
Utilitas , Volume 27 , Issue 3 , September 2015 , pp. 263 - 278
Check for updates
Copyright
Copyright © Cambridge University Press 2015 

INTRODUCTION

In Of Laws in General, Bentham develops ‘a hitherto undenominated branch of logic’,Footnote 1 the logic of imperation. This account, however, faces serious problems. Chief among these problems is Bentham's assumption that imperations are truth-apt. But the truth-aptness of imperations is far from obvious. We can address the truth-aptness problem by treating imperations as including a sanction to motivate obedience. Imperations are disjunctive descriptions concerning future states of affairs. One disjunct describes what the subject shall do. The other describes the undesirable state of affairs awaiting the disobedient.

This interpretation of Bentham's logic of imperation provides a compelling account of the formal limits of legal validity. Specifically, by issuing contradictory or contrary mandates, the sovereign lacks the ability to manipulate expectations to motivate obedience. Paralleling formal invalidity, mandates from the sovereign can, on Bentham's view, also be substantively invalid. A law is substantively invalid when it would disappoint expectations sufficiently to cause subjects to abandon the general habit of obedience. Accordingly, concerning such imperations, the sovereign lacks the ability to motivate obedience.

Call the claim that expectations limit the sovereign's ability to make valid law, both formally and substantively, the parity thesis. The goal of this essay is to flesh out and defend this thesis.

I proceed in four sections. In the first section, I provide an overview of Bentham's logic of imperation and the problems it faces. In the second section, by drawing on other aspects of Bentham's jurisprudence, I articulate a novel interpretation of imperations capable of resolving these problems. This interpretation gives us the resources to delineate the formal bounds of legal validity. In the third section, I turn to the parity thesis. Here I argue that mandates can be substantively invalid if their content would, by sufficiently disappointing subjects’ expectations, undermine the general habit of obedience to the sovereign. In so far as such mandates can be individuated from the general body of laws, they can, by failing this counterfactual test, be rendered invalid without undermining the sovereign's authority entirely. In the fourth section, I highlight an appealing implication of the parity thesis. Grounding formal and substantive invalidity in a unified source allows us to identify the problem with substantive invalidity without jeopardizing the separation between what the law is and what the law ought to be.Footnote 2

THE LOGIC OF IMPERATION

Bentham defines a valid law as follows:

[A]n assemblage of signs declarative of a volition conceived or adopted by the sovereign in a state, concerning the conduct to be observed in a certain case by a certain person or class of persons, who in the case in question are or are supposed to be subject to his power: such volition trusting for its accomplishment to the expectation of certain events which it is intended such declaration should upon occasion be a means of bringing to pass, and the prospect of which it is intended should act as a motive upon those whose conduct is in question.Footnote 3

We can break this definition into three parts. A valid law consists, first, in a statement asserting the volition of the sovereign that is to govern a person's conduct, where, second, the person is in a habit of obedience to the sovereign and, third, the volition is backed by certain expectations of pain intended to motivate obedience.Footnote 4

We can clarify the nature of a sovereign's mandates by turning to Bentham's account of the logic of imperation. This account demarcates four exhaustive relations of mandates. A command: one shall ϕ (e.g. every householder shall carry arms); a non-command: one need not ϕ (e.g. any householder may forbear to carry arms); a prohibition: one shall ~ϕ (e.g. no householder shall carry arms); and, a non-prohibition: one may ϕ (e.g. any householder may carry arms).Footnote 5

These four types of mandates exhaust the volitions of the sovereign. Consider any act. It will either be the wish of the sovereign to influence the subject's conduct concerning the act or it will not. If the sovereign wishes to influence the conduct of the subject concerning the act, it is considered decided; if not, then it is considered undecided.Footnote 6 Commands and prohibitions are decided; non-commands and non-prohibitions are undecided.

Just as in categorical logic, among each of these four mandates ‘subsists such a relation that with respect to one another some of them are necessarily repugnant and exclusive; others as necessarily concomitant’.Footnote 7 This means each is subject to negation and translation. As Bentham explains, ‘A command . . . includes a permission: it excludes both a prohibition and a non-command. A prohibition includes a non-command and it excludes both a command and a permission.’Footnote 8 In addition, ‘a command may wear the form of a prohibition: and a prohibition that of a command. The law which prohibits the mother from starving her child commands her to take care that it be fed. The one may at pleasure be translated or converted into the other.’Footnote 9 Mandates can, accordingly, be diagrammed on the standard AEIO square of opposition.Footnote 10

The logic of imperation regulates the formal limits of legal validity. We can state the following of any self-consistent system of laws:

  1. 1. If the sovereign commands ϕ-ing,Footnote 11 then it also non-prohibits ϕ-ing; and cannot prohibit ϕ-ing, nor non-command ϕ-ing.

  2. 2. If the sovereign prohibits ϕ-ing, then it also non-commands ϕ-ing; and, cannot command ϕ-ing, nor non-prohibit ϕ-ing.

  3. 3. For any action, it is either non-commanded or non-prohibited, or both.

  4. 4. If the sovereign commands ϕ-ing, this is the same as prohibiting ~ϕ-ing (and vice versa).

  5. 5. If the sovereign non-prohibits ϕ-ing, this is the same as non-commanding ~ϕ-ing (and vice versa).

  6. 6. From 1–5 we can conclude that for any law concerning ϕ-ing it is always either:

    1. a. a command and non-prohibition (ϕ-ing is commanded);

    2. b. prohibition and non-command (ϕ-ing is prohibited); or

    3. c. non-prohibition and non-command (ϕ-ing is free or one is at liberty to ϕ).

  7. 7. From 1–6, faced with any action a person can always act within the law.Footnote 12

(7) is a virtue. In a properly functioning legal system, it should in principle be possible to avoid breaking the law. A legal system is seriously flawed when both the performance and non-performance of an act is illegal. Bentham's imperative theory eliminates such flaws. If a sovereign, for example, commands ϕ-ing and also prohibits ϕ-ing, then one of these laws is invalid. We can know this, even without knowing what ϕ-ing involves. There are, however, two problems that threaten Bentham's account.

The first, raised by Lyons,Footnote 13 is this. Non-prohibitions and non-commands, being undecided, are uncoercive. They fail to be backed by a motivation produced by certain expectations of pain or pleasure. Yet, Bentham still calls assertions of this sort laws. This appears to conflict with the third part of the definition of legal validity: a law is backed by the motivation produced through certain expectations of pain or pleasure. As Bentham writes, ‘A law by which nobody is bound, a law by which nobody is coerced, a law by which nobody's liberty is curtailed, all these phrases which come to the same thing would be so many contradictions in terms.’Footnote 14

We can address this worry. Laws such as ‘any householder may carry arms’ and ‘any householder may forbear to carry arms’ still require sanctions. Such a law ‘is addressed to subordinate power-holders in general restraining them from breaking in upon the liberty of the party whom the uncoercive mandate in question is meant to favour’.Footnote 15 Non-prohibitions and non-commands provide spheres of protection where a person, for instance, can participate in the act of carrying arms. If one tries to block another from participating in carrying arms when the sovereign has asserted that one may participate in this carrying, then the interferer will be met with a sanction.Footnote 16 Thus, although not directly, non-commands and non-prohibitions can still ‘expose [some person or persons] to suffer by it’.Footnote 17

Turning to the second worry, in classical logic relations such as opposition, concomitance, negation, etc., rely on the truth-values of the propositions. It is, however, unclear why a command is the negation of a non-command or a prohibition is the negation of a non-prohibition.Footnote 18 Imperations are not transparently truth-apt. Why can it not be true that a sovereign commands ϕ-ing and also non-commands or prohibits ϕ-ing? As Lyons critically notes, ‘[T]here is no formal contradiction between imperations such as “Jones is to [ϕ]” and “Jones may refrain from [ϕ-ing]”, or even between statements concerning imperatives, such as “Jones is required to [ϕ]” and “Jones is permitted to refrain from [ϕ-ing].”’Footnote 19 Call this the truth-aptness problem.

One attempt to address this problem, offered by Hart, is to take each mandate as a first-person descriptive declaration of the sovereign's will. On Hart's interpretation, by making each mandate a statement of fact that the sovereign has the relevant volition, we can treat imperatives as truth-apt.Footnote 20 Hart's interpretation has textual backing. First, Bentham claims that any utterance made by a speaker is merely the expression of the speaker's will or belief.Footnote 21 According to Bentham, ‘[N]o matter other than that which passes in a man's own mind can any immediate communication be made by language.’Footnote 22 This means when one asserts p, one is merely stating: ‘in my opinion, p’. Second, Bentham argues that mandates must have been taken up in the contemplation of the sovereign and positively chosen as expressions of its will.Footnote 23

From these points Hart maintains that if a sovereign commands ϕ-ing, this is an elliptical way of saying: it is my (the sovereign’s) wish that all ϕ. And, if the sovereign issues a non-command, this should be taken to mean: it is my wish that all may forbear from ϕ-ing.Footnote 24 Understood in this way, if it is true that the sovereign wishes that all ϕ, then it must be false that the sovereign wishes that all forbear from ϕ-ing.

Hart's solution, however, is at best partial. Converting imperations to descriptive declarations can account for contradictories – commands and non-commands, prohibitions and non-prohibitions – but not contraries. Even after conversion, there is nothing in classical logic indicating that a command to ϕ and a prohibition to ϕ taken descriptively cannot both be true. The sovereign can truly assert both: ‘It is my wish every householder shall carry arms’ and ‘It is my wish no householder shall carry arms.’ Hart recognizes this problem. Failing to see a solution, he deems Bentham's account unworkable.Footnote 25

Hart's assessment is, however, too hasty. We can offer an interpretation of Bentham's account that captures contraries. Rather than translate imperatives into descriptive statements about the volition of the sovereign, we should translate them into disjunctive statements about future states of affairs.

THE MOTIVATIONAL DISJUNCT

Normally, a reduction of imperations to ‘do this . . . or else’ commands is implausible.Footnote 26 If I tell you ‘shut the door’, this is not (in every case) an elliptical way of asserting ‘shut the door or I will intend to harm you’.Footnote 27 But, importantly, I am not a sovereign. Bentham develops his account specifically to capture sovereign imperations.Footnote 28 Given that sovereign imperations always include a sanction to motivate obedience, a ‘do this . . . or else’ reduction is not implausible. We can, by attaching a motivational disjunct to the sovereign's mandates, solve the truth-aptness problem. Doing this correctly, however, requires turning to the other aspects of Bentham's definition of legal validity.

Start with the nature of sovereignty. A sovereign is a ‘person or assemblage of persons to whose will a whole political community are . . . supposed to be in a disposition to pay obedience: and that in preference to the will of any other person’.Footnote 29 Sovereignty, that is, consists in a community's habit of obedience exclusively to a certain person or group of persons. But the habit here needs to be reciprocated. As Bentham further explains, sovereignty is ultimately based on the ‘habit of commanding on [the] one side, accompanied by a habit of obeying on the other: and more immediately on the one part in a disposition on the one part to expect obedience, on the other part in the disposition to pay it’.Footnote 30

But why, if all are in a habit of obedience, must a law involve a coercive sanction? A sanction is required to ensure that the general habit of obedience holds for particular mandates. A general habit of obedience justifies the sovereign as a legitimate lawgiver, but the particular mandates, to dissuade acts of disobedience, must be accompanied by sanctions.Footnote 31 Citizens may recognize and be willing to obey an entity as their sovereign, generally, but be unmotivated to obey particular mandates. The sovereign, accordingly, motivates obedience for a particular mandate by tipping the tendency of pain in favour of compliance.Footnote 32 A law, recall, must produce some mischief. To make a law ‘is to do evil that good may come’.Footnote 33 Thus, to be a valid law, a sovereign's mandates must include a sanction that redirects the ‘flow . . . of natural motivation which is the fear of pain or suffering’Footnote 34 in favour of compliance.

The need to ensure obedience leads to a unique characteristic of sovereign imperations. A sovereign's assertions always include the intention to bring about some undesirable state of affairs for the disobedient.Footnote 35 Given this intention, the sovereign's assertion can be translated: one shall carry arms or the sovereign shall intend to bring about some undesirable state of affairs.Footnote 36 Put loosely, you will carry arms or I (the sovereign) will intend to bring about pain for you. One disjunct states what one shall do; the other tells of the undesirable alternative for failing to actualize the first disjunct. The ‘shall’ in both disjuncts can be taken descriptively. By attaching a motivational disjunct to a sovereign's imperations, truth-aptness ceases to be a problem. But the contraries problem remains.

On the straightforward disjunctive interpretation, a sovereign's command to ϕ should be converted into: one shall ϕ or the sovereign shall intend to bring about some undesirable state of affairs. A sovereign's prohibition to ϕ should be converted into: ‘one shall ~ϕ or the sovereign shall intend to bring about some undesirable state of affairs.’

This interpretation is a move in the right direction, but it still does not get us contraries. Consider, this time as conditionals, what is being asserted if ϕ-ing is commanded and prohibited:

  • • Command to ϕ: If one does not ϕ, the sovereign shall intend to bring about some undesirable state of affairs.

  • • Prohibition to ϕ: If one ϕs, the sovereign shall intend to bring about some undesirable state of affairs.

Unfortunately for subjects, these could both be true. But we can get contraries in a different way. We can work backward. We can ask: why can't contrary mandates be valid laws?

Return to Bentham's definition of a valid law. There are two plausible candidates for why a command and prohibition to ϕ, taken together, fail to be valid laws. First, they could fail to be ‘an assemblage of signs declarative of a volition conceived or adopted by the sovereign’.Footnote 37 Second, they could fail to bring about the ‘expectation of certain events which it is intended such declaration should upon occasion be a means of bringing to pass, and the prospect of which it is intended should act as a motive upon those whose conduct is in question’.Footnote 38

The first of these candidates is not promising. A sovereign might accurately capture its volition by declaring that, if a subject ϕs or not, an undesirable state of affairs is in the offing. Regardless of what a subject does, the sovereign may truly intend to bring about some undesirable state of affairs. We thus cannot formally rule out contrary mandates on the basis of a declaration concerning the sovereign's will. This is no surprise. Hart's interpretation floundered on precisely this point.

Turning to the motivational requirement, however, is more promising. A valid law requires not just a sanction, but a sanction intended to ‘influence the conduct of the party in question’.Footnote 39 If the same undesirable state of affairs were to be expected regardless of whether one ϕs or not, the motivational disjunct would not motivate. Disobedience, and the undesirable state of affairs that follows, is inevitable. With contrary laws, in other words, the sovereign intends to bring about an undesirable state of affairs regardless of how the subjects act. But if the sovereign both commands and prohibits ϕ-ing, the undesirable state of affairs floats free from ϕ-ing. Contrary mandates cannot motivate obedience. If, accordingly, the attached sanction is not motivationally efficacious, then one of the conditions requisite for validity is left unsatisfied.

In light of this point, we need to modify our interpretation. What needs to be captured is that the sovereign attaches an undesirable state of affairs when, say, commanding all to ϕ with the intention to motivate compliance with ϕ-ing over ~ϕ-ing. More precisely, the sovereign intends to artificially adjust expectations to secure obedience. So, when a sovereign commands one to ϕ, this is an elliptical way of stating: one shall ϕ or the sovereign shall intend to bring about a less desirable state of affairs than the state of affairs (with regard to the sovereign's actions) brought about by one's ~ϕ-ing.Footnote 40 The latter disjunct captures the sovereign's intention to tip the expectation of pain in favour of compliance.

This gives us what we need for contraries. Consider, again as conditionals, what is being asserted if ϕ-ing is commanded and prohibited. On this refined interpretation, we get:

  • • Command to ϕ: If one does not ϕ, the sovereign shall intend to bring about a less desirable state of affairs than the state of affairs (with regard to the sovereign's actions) brought about by ϕ-ing.

  • • Prohibition to ϕ: If one ϕs, the sovereign shall intend to bring about a less desirable state of affairs than the state of affairs (with regard to the sovereign's actions) brought about by ~ϕ-ing.

The motivationally efficacious disjunct attached to the command to ϕ is incompatible with the disjunct attached to the prohibition. The sovereign can adjust expectations to motivate obedience for ϕ-ing or ~ϕ-ing, but not both. Formal invalidity is thus a product of the sovereign's inability to adjust expectations to motivate obedience.

On this interpretation, the three parts of Bentham's definition of legal validity are interconnected. The issuing of a particular mandate must come from the sovereign and must be accompanied by the expectation of pain intended to motivate obedience. To repeat: formal invalidity is the product of the sovereign's inability to adjust expectations to motivate obedience. This same thought, or so I will argue in the next section, lies at the heart of substantive invalidity

SUBSTANTIVE LEGAL INVALIDITY

In this section, I defend the parity thesis: Expectations limit the sovereign's ability to make valid law both formally and substantively. We have already seen how the inability of the sovereign to adjust expectations can render certain mandates formally invalid. Here I argue that, like formal invalidity, certain mandates, because of their content, are such that the sovereign is unable to adjust expectations to motivate obedience. These laws are substantively invalid.

We can start with cases. Consider,

Covenant-Breaking Successor. A new sovereign mandates that no one (including the sovereign) shall keep past covenants made under its predecessor.

In this case, the original sovereign's covenants establish strong expectations. Such expectations

equal the expectation which is entertained of the prevalence of that disposition to obedience on the part of the people by which the sovereignty de facto is constituted: insomuch that the observance of the covenant on the one part shall be looked upon as a condition sine qua non to the obedience that is to be paid on the other.Footnote 41

The Covenant-Breaking Successor disappoints expectations sufficiently to destroy subjects’ general habit of obedience. So, if a sovereign came to the throne determined to make such a covenant-breaking mandate, ‘he would be told that he had adopted [the covenants of his predecessor] notwithstanding’.Footnote 42 The severity of the disappointed expectations put the covenant-breaking mandate outside the limits of the new sovereign's authority.

Consider a second case.

Anti-Kosher Mandate. A sovereign reigns over a (practicing) Jewish population. The sovereign issues a mandate that all persons shall eat pork twice weekly.

As in the previous case, the Jewish people's expectations might be so severely disappointed that the general habit of obedience is compromised. ‘The Jews’, as Bentham writes, ‘would have done anything else for Antiochus, but they would not eat his pork.’Footnote 43 Recall, as we saw above, that sovereignty is ultimately based on the reciprocal expectations: ‘on the one part in a disposition on the one part to expect obedience, on the other part in the disposition to pay it’.Footnote 44 In Anti-Kosher Mandate, disappointed expectations limit what the sovereign can take as ‘objects of his law’.Footnote 45

The rough picture we get from these cases is this. Expectations can be sufficiently disappointed to call into question the general habit of obedience, and this can limit the ability of the sovereign to make valid law. If we fill out the details of this picture, the parity thesis emerges.

We can begin by clarifying the general habit of obedience. Bentham contends that: ‘The obedience paid to any particular Law rests ultimately on a circumstance extrinsic to all Laws, a general habit of obedience. A habit of executing upon all occasions and in all points commands of some one person at least or body of persons in the state.’Footnote 46 ‘General’ here refers to the social convention of obedience for all (or most) members of society to all (or most) of some person's (or assemblage of persons’) mandates. The sovereign's ability to make valid law thus, in short, ‘necessarily depends on one's disposition to recognize it, which disposition depends on the expectation that most others in the community will or do also recognize it, where their recognition depends in part on the expectation that one will (or does) recognize it’.Footnote 47 Accordingly, a mandate that can call into question the general habit of obedience is a mandate that, if given, renders all (or most) members of society willing to disobey the sovereign's other mandates.

If the disposition to obedience ceases, then so too does the sovereign's ability to make valid law. And, sufficiently disappointed expectations can cause the general habit of obedience to cease.Footnote 48 For the effect of disappointing expectations, in Bentham's words, ‘is to weaken on the part of the people . . . that disposition to submission and obedience, by which the power of the sovereign, in point of fact, is constituted’.Footnote 49 Crucial to maintaining the general habit of obedience is avoiding a mismatch between existing and new expectations.Footnote 50 As Bentham writes:

If we could suppose a new people, a generation of children: the legislator, finding no expectations formed which could oppose his views, might fashion them at his pleasure, as the sculptor fashions a block of marble. But as there already exists among all people a multitude of expectations, founded upon ancient laws or ancient usages, the legislator is obliged to employ a system of conciliations and concessions, which constantly restrain him.Footnote 51

Return to Anti-Kosher Mandate. The Jewish people have natural expectations, founded on religious customs, not to eat pork. If the sovereign commands that all must eat pork, these expectations will be disappointed. And if this disappointment is sufficiently severe the very general habit of obedience is in jeopardy.Footnote 52

Though disappointing existing expectations may threaten the general habit of obedience, importantly, this does not challenge sovereignty wholesale. In the Fragment on Government, Bentham writes: ‘For a body then, which is in other respects supreme, to be conceived as being with respect to a certain sort of acts, limited, all that is necessary is, that this sort of acts be in its description distinguishable from every other.’Footnote 53 In other words, if a mandate is severable from the general body of laws, then a limitation on the sovereign's ability to mandate a certain act does not undermine the sovereign's ability to make other valid laws. For example, in so far as the mandate to eat pork is detachable from the general body of laws, one may reject the sovereign's mandate concerning eating pork without undermining sovereignty entirely. As Bentham clarifies,

A certain act is in the instrument of convention specified, with respect to which the government is therein precluded from issuing a law to a certain effect: whether to the effect of commanding the act, of permitting it, or of forbidding it. . . . These bounds the supreme body in question has marked out to its authority: of such a demarcation, then, what is the effect? . . . that the disposition to obedience confines itself within these bounds. Beyond them the disposition is stopped from extending: beyond them the subject is no more prepared to obey the governing body of his own state, than that of any other.Footnote 54

Individuatable mandates can fall outside the jurisdiction of the sovereign's authority to make valid law; specifically, those mandates whose content is such that they would undermine the general habit of obedience.

This gives us a counterfactual test for determining the limits of legal validity. If mandating an act would call into question the general habit of obedience,Footnote 55 then in so far as this mandate is distinguishable from the general body of laws,Footnote 56 the mandate fails to be a valid law.Footnote 57 For example, Anti-Kosher Mandate would fail to be a valid law if, were the sovereign to command it, the Jewish population's expectations would be disappointed to the point where they would be willing to disobey the sovereign's other commands. But what stands behind this test?

Enter the parity thesis. As we saw in the previous section, formal invalidity is a product of the sovereign's inability to adjust expectations to motivate obedience. The same holds for substantive invalidity. Take the command that all persons shall eat pork twice weekly. This should be translated to: one shall eat pork twice weekly or the sovereign shall intend to bring about a state of affairs less desirable than the state of affairs (with regard to the sovereign's actions) brought about by eating pork twice weekly. There is nothing formally problematic with this mandate. If, for example, the sovereign's subjects were happy to engage in pork eating, then the law would be valid. If, however, the sovereign ruled over a devout Jewish population, willing en masse to discontinue allegiance rather than eat pork, the sovereign cannot adjust expectations to motivate obedience. The issuing of the mandate would undermine the very habit of obedience upon which the sovereign's authority to make valid law rests.Footnote 58

By focusing on the perspective of the subjects, we can see the parity thesis in a different, better way. Formal validity, as when the sovereign issues contrary mandates concerning ϕ-ing, ensures an incompatible set of artificial expectations. Issuing contrary or contradictory mandates necessarily disappoints expectations such that the sovereign cannot motivate obedience. Substantive invalidity, by contrast, does not ensure an incompatible set of expectations. Rather, established expectations are such that the issuing of certain mandates introduces a conflict sufficient to undermine the general habit of obedience. The strength of existing expectations is such that, like formal invalidity, the sovereign (qua sovereign) cannot motivate compliance.

The parity thesis holds. The inability of the sovereign to adjust expectations to motivate obedience grounds both formal and substantive invalidity.

THE APPEAL OF THE PARITY THESIS

The above analysis, if correct, clarifies Bentham's jurisprudence. Given the place Bentham holds in the history of legal philosophy, this, alone, is significant. But the parity thesis is, I believe, attractive in its own right.

Many people reject Bentham's account.Footnote 59 The most common charge, broadly put, is that the account leaves out certain aspects that the concept of law intuitively has.Footnote 60 It is fortunate, then, that we have focused on invalidity. So long as Bentham's account accurately captures the minimal conditions of legality, these objections do not impugn the parity thesis. The parity thesis is detachable.

That there is a problem when a sovereign asserts contradictory or contrary mandates is manifest. This, however, is not the case for substantive invalidity, especially for positivists. If we want to maintain that what the law is and what the law ought to be are separate questions,Footnote 61 grounding formal and substantive invalidity in a unified source allows us to identify the substantive problems without jeopardizing this separation.

To see this, return to the Hart–Fuller debate. In reading this debate, most are struck by a set of conflicting intuitions. On the one hand, it seems that legal validity is not held hostage to morality; a mandate's morality and legality are distinct questions. On the other hand, it seems the immorality of certain mandates ensures their invalidity; valid laws cannot be morally unacceptable. The parity thesis – the inability of the sovereign to adjust expectations to motivate obedience – provides the resources to capture both of these intuitions.

Imagine that the current British government tried to enact the same set of laws as were in place in 1941 Germany. Given the existing expectations of the British people, such laws, it seems safe to assume, would undermine the general habit of obedience. That is, mandates to carry out the mass extermination of a group of people would cause the population to end their allegiance to the Crown. We can, given the above analysis, accordingly, rule the 1941 Germany laws substantively invalid for the current British population. Their immorality would sufficiently disappoint expectations such that the sovereign would be unable to motivate compliance.

This allows us to explain our intuition that these laws are invalid, while retaining the separation between laws and morals.Footnote 62 This is the parity thesis's great appeal. By relying on the same explanation as formal invalidity, we can deem laws substantively invalid without bringing in any additional apparatus that might compromise the separation thesis.Footnote 63

References

1 Bentham, J., Of Laws in General, ed. Hart, H. L. A. (London, 1970), p. 15 n. hGoogle Scholar.

2 Or, in Benthamite argot, between ‘expository’ and ‘censorial’ jurisprudence. See Bentham, J., The Works of Jeremy Bentham, 11 vols., ed. Bowring, J. (New York, 1962), vol. 1, p. 238Google Scholar.

3 Bentham, Of Laws, p. 1.

4 There is, to be precise, more than one way a party can be affected by a law: ‘by being exposed at least to suffer by it, [and] by being favoured or intended to be favoured by it’ (Bentham, Of Laws, p. 53). To save words, however, I will focus exclusively on the former. But nothing I say in what follows is incompatible with motivating through some favourable enticement.

5 Bentham, Of Laws, pp. 95–7.

6 Bentham, Of Laws, pp. 95.

7 Bentham, Of Laws, pp. 97.

8 Bentham, Of Laws, pp. 97.

9 Bentham, Of Laws, pp. 96.

10 For illustration, see Mullock, P., ‘Logic and Liberty’, Philosophical Studies, 35 (1979), pp. 217–38, at 224CrossRefGoogle Scholar.

11 We can avoid cumbersome locutions by letting ϕ-ing stand for an action applicable to all subjects. A command to ϕ, for example, should be taken as: ‘All subjects shall ϕ’. Were we to limit, as most laws are limited, the law to a class of persons, we would need to add: ‘for all x, if x is of such-and-such a description x shall ϕ’.

12 The following draws on L. J. Lysaght, ‘Bentham on the Aspect of a Law’, Northern Ireland Legal Quarterly, 24 (1973), pp. 383–98, at 390–6; Lyons, D., In the Interest of the Governed: A Study in Bentham's Philosophy of Utility and Law (Oxford, 1973), pp. 117–20Google Scholar; Mullock, ‘Logic and Liberty’, p. 224; Raz, J., The Concept of a Legal System: An Introduction to the Theory of Legal System, 2nd edn. (Oxford, 1980), pp. 55–9CrossRefGoogle Scholar; and Hart, H. L. A, Essays on Bentham: Jurisprudence and Political Theory (Oxford, 1982), pp. 114–16CrossRefGoogle Scholar.

13 Lyons, In the Interest, p. 112.

14 Bentham, Of Laws, p. 54.

15 Bentham, Of Laws, p. 99.

16 The implication here is that all undecided mandates are accompanied by decided ones, as suggested by Lysaght, ‘Aspect of a Law’, p. 398.

17 Bentham, Of Laws, p. 53.

18 See Lyons, In the Interest, p. 120; Mullock, ‘ Logic and Liberty’, p. 222; Hart, Essays on Bentham, p. 114.

19 Lyons, In the Interest, pp. 122–3. Lyons continues: ‘[I]mperational contradiction requires that a restrictive law exists if and only if the corresponding permissive law does not exist’ (Lyons, In the Interest, pp. 122–3). Bentham's logic, hence, denies the very possibility of conflict within a legal system. But, since such conflicts do obtain, ‘Bentham's logic is absurd’ (Lyons, In the Interest, p. 123). This objection is not compelling. Conflicts may obtain within legal systems but these are merely dysfunctional systems. If a person denies the antecedent it does not mean that there is something wrong with logic; there is something wrong with this person's reasoning.

20 Hart, Essays on Bentham, p. 248. For an earlier attempt to address this problem by using descriptions of sovereigns’ attitudes, see Lyons, D., ‘Logic and Coercion in Bentham's Theory of Law’, Cornell Law Review 57 (1972), pp. 335–62Google Scholar, at 341–2.

21 Hart, Essays on Bentham, p. 248.

22 Bentham, Works, vol. 8, p. 321.

23 Bentham, Of Laws, pp. 97–8.

24 Hart, Essays on Bentham, p. 249.

25 Hart, Essays on Bentham, p. 115.

26 For a contemporary overview of the meaning of imperatives, see Charlow, N., ‘The Meaning of Imperatives’, Philosophy Compass 9 (2014), pp. 540–55CrossRefGoogle Scholar.

27 For a proposal to translate all imperations in this disjunctive fashion, see Bohnert, H. G., ‘The Semiotic Status of Commands’, Philosophy of Science 12 (1945), pp. 302–15Google Scholar.

28 When selecting a common appellation for what to call an imperatival utterance, for example, Bentham rejects a number of candidates on the ground that they give ‘no intimation that the will in question is in any manner referable to the sovereign’ (Bentham, Of Laws, p. 10). And nowhere, to my knowledge, does Bentham indicate that the logic of imperation he develops extends to everyday imperations.

29 Bentham, Of Laws, p. 18.

30 Bentham, Of Laws, p. 109, 139.

31 Postema, G. J., Bentham and the Common Law Tradition (Oxford, 1986), p. 245Google Scholar.

32 Bentham, Of Laws, p. 31, 133.

33 Bentham, Of Laws, p. 54. By mischief, Bentham simply means that one is forced to suffer by the coercive sanction attached to the law, through either ‘punishment or vindictive compensation’ (Bentham, Of Laws, p. 55).

34 Hart, Essays on Bentham, p. 222.

35 Bentham, Of Laws, p. 56; see also Lyons, In the Interest, p. 116. Bentham, in The Principles of Morals and Legislation, writes that if the sovereign says: ‘Cause to be hanged whoever in due form of law is convicted of stealing, is, though not a direct, yet as intelligible a way of intimating to men in general that they must not steal, as to say to them directly, Do not steal: and one sees, how much more likely to be efficacious’ (Bentham, Works, vol. 1, p. 151 n. 8).

36 Put differently: the first disjunction is directive; the second is comminative. Laws, Bentham argues, always have this two-part structure; see Bentham, Of Laws, p. 167.

37 Bentham, Of Laws, p. 1.

38 Bentham, Of Laws, p. 1.

39 Bentham, Of Laws, p. 95.

40 Understanding Bentham's account in this way has other advantages. For example, Hart charged Bentham's jurisprudence with extensional inadequacy, see Hart, H. L. A., The Concept of Law, 2nd edn. (Oxford, 1994)Google Scholar, ch. 3. Bentham's account cannot, Hart argued, capture all the different varieties of laws, such as the rules for making contracts. Treating mandates as disjunctive, however, provides a first step in meeting Hart's challenge. We can translate such rules into the Benthamite account as follows: for any x, if x wishes to make a contract, x shall follow rule (r) or the sovereign shall intend to bring about a less desirable state of affairs than the state of affairs (with regard to the sovereign's actions) brought about by following r; namely, the sovereign will not treat the contract x desired to make legally binding as legally binding. We can offer similar translations for the conferral of legal powers. Perhaps these translations distort ‘the different social functions which different types of legal rules perform’ (Hart, Concept of Law, p. 38). But that is a different, less serious charge.

41 Bentham, Of Laws, p. 66.

42 Bentham, Of Laws, p. 66.

43 Bentham, Of Laws, p. 19 n.b.

44 Bentham, Of Laws, p. 109.

45 Bentham, Of Laws, p. 19n.b.

46 Bentham MSS in the University College, London Library: UC lxix. 204.

47 Postema, Bentham and the Common Law, p. 249. Determining the general habit of obedience, accordingly, is a matter of ‘looking to the general dispositions of the people in the community’ (Postema, Bentham and the Common Law, pp. 242, 237). It is only from this general disposition to obedience that ‘government, and what is called political society, and the only real laws, result’ (Bentham, Works, vol. 3, p. 219).

48 Bentham rightly affords disappointed expectations great disutility. For an overview, see Postema, Bentham and the Common Law, pp. 161–2.

49 Bentham, Of Laws, p. 16. As Bentham emphasizes, ‘the efficacy of the law . . . depend[s] in a considerable degree on the existence, real or supposed, of some customs to which it is or pretends to be conformable’ (Bentham, Of Laws, p. 109)

50 Expectations can be natural (such as expectations that arise spontaneously through habits, customs or convention) or artificial (such as when the sovereign announces that an undesirable state of affairs shall accompany non-compliance); see Bentham, Works, vol. 1, p. 323.

51 Bentham, Works, vol. 1, p. 323.

52 We can, perhaps, better understand Bentham's point by drawing a parallel with Hart's discussion of the internal and external perspective one can take on a legal system. According to Hart, we can look on a legal system, externally, ‘as an observer who does not himself accept [the laws], or [internally] as a member of the group which accepts and uses them as guides to conduct’ (Hart, Concept of Law, p. 89). To create obligations, the subjects must see the sovereign's mandates from the internal point of view. However, the content of certain mandates, as in Covenant-Breaking Successor and Anti-Kosher Mandate, shift subjects’ perspectives from the internal to external point of view. The content of such mandates falls outside the sovereign's authority, and hence cannot qualify as valid laws.

53 Bentham, Works, vol. 1, p. 290.

54 Bentham, Works, vol. 1, p. 290.

55 Some laws might disappoint expectations, but not sufficiently to make all (or most) subjects willing to disobey the sovereign's other mandates. These laws would not be rendered invalid by this counterfactual test. I thank an anonymous referee for prompting me to clarify this point.

56 If not, then sovereignty itself is in jeopardy. As Bentham writes: ‘The true rampart, the only rampart, against a tyrannical government has always been, and still is, the faculty of allowing this disposition to obedience – without which there is no government – either to subsist or to cease . . . Shall this habit of obedience be continued unbroken, or shall it be discontinued upon a certain occasion? Is there more to be gained than to be lost in point of happiness, by its discontinuance? Of the two masses of evil, – intensity, duration, certainty, all included – which appears to be the greatest, that to which one believes one's self exposed from continued obedience, or that to which one believes one's self exposed by its discontinuance?’ (Bentham, Works, vol. 3, p. 219).

57 This softens Bentham's positivism significantly. It is possible, for example, that a law allowing one to profit from wrongdoing would fail this counterfactual test.

58 A brick cannot be added to an edifice, if adding it would bring the whole structure down.

59 This is, perhaps, an understatement. As Green writes, ‘Imperatival theories are now without influence in legal philosophy’ (L. Green, ‘Legal Positivism’, Stanford Encyclopedia of Philosophy, <http://www.plato.stanford.edu/archives/fall2009/entries/legal-positivism/> (2009)).

60 Many believe the following objection to Bentham's account is damning: ‘[The] analysis fails entirely to account for, even to recognize, certain striking facts about the attitudes we take toward “the law.” We make an important distinction between law and even the general orders of a gangster. We feel that the law's strictures – and its sanctions – are different in that they are obligatory in a way that the outlaw's commands are not’ (Dworkin, R., ‘The Model of Rules’, The University of Chicago Law Review 35 (1967), pp. 14–46Google Scholar, at 19; cf. Hart, Essays on Bentham, pp. 253–5; Hart, Concept of Law, pp. 50–61). According to this objection, pace Bentham, sovereign authority is more than just adding reasons in favour of obedience; possessing authority consist in offering pre-empting reasons. See, for example, Leiter, B., ‘Legal Realism, Hard Positivism, and the Limits of Conceptual Analysis’, Hart's Postscript, ed. Coleman, J. (Oxford, 2001), pp. 355–70Google Scholar, at 363–4; Raz, J., The Morality of Freedom (Oxford, 1986), pp. 57–61Google Scholar. However, for reasons that go beyond the scope of this article, I do not regard this objection as decisive. In any case, even if this objection shows Bentham's account is, in certain respects, inadequate, the significance of the parity thesis remains.

61 For the seminal argument in favour of this separation, see Hart, ‘Positivism and the Separation of Law and Morals’, Harvard Law Review 71 (1958), pp. 593–629.

62 As should be clear, the connection between legal validity and morality is not conceptual. It is contingent. The expectations (that happen to be partially informed by a society's moral views) indirectly invalidate the law. This indirectness is a virtue. If we can invalidate a law in this indirect way, we avoid violating the separation thesis. We can thus explain our intuition that the immorality of certain mandates ensures their invalidity. This intuition is tracking the thought that for us, given our expectations, that kind of immoral mandate couldn't be a valid law. If mandated, we would discontinue our general habit of obedience.

63 Many thanks to Luke Semrau, Shannon Fyfe, Larry May and an anonymous referee for helpful comments on earlier drafts.