Introduction
The expression ‘Federal-State relations’—along with parallels such as ‘federal-provincial relations’ and ‘Centre-State relations’—is ubiquitous in the discourse of federalism across jurisdictions. Indeed, it is often employed as synonymous with ‘federalism’. When compared to the more general term, ‘Federal-State relations’ has the advantage of emphasizing the bilateral character of federalism, i.e., of underlining the essential dichotomy that defines the federal phenomenon. Further, it also evokes the plurality of what it describes: to speak of ‘Federal-State relations’ is to imply that there may be many such relations, each one displaying this bilateral character. Finally, the expression ‘Federal-State relations’ seems more appropriate—and useful—if one wishes to analyse the legal structure of federal systems, which is organized around this essential dichotomy while also being its product.
By ‘the legal structure of federal systems’ or ‘federal structure’ I mean simply a political/legal system where a central/federal/national government coexists with state/regional/provincial governments, with both levels having jurisdiction over the same territory and the same population, and with a constitutional division of legislative powers. This structure in itself, whatever the political and social realities that surround it, systematically creates certain legal questions that politicians, lawyers, courts and scholars must regularly grapple with: most federations with a well-evolved constitutional jurisprudence present issues of concurrency, implied powers, preemption and supremacy, among others. These issues follow from the very logic of federalism, i.e., from the very logic of two governments plus one population plus division of powers. Since this division of powers is enshrined in a text, each federal constitutional court must eventually develop techniques of interpretation aimed at solving the problem of determining how constitutional provisions and the specific terms used therein—‘necessary and proper’, ‘commerce’, ‘peace, order and good government’, ‘incidental’, ‘repugnancy’—are to be read with respect to specific facts, i.e., specific State and Federal laws and executive acts. In other words, all federal systems face similar legal questions that flow from the federal structure. Whenever there is a Federal-State dispute regarding competence, the essential question to be answered is ‘Who does what?’ Which level of government has the legislative capacity to perform such-and-such legislative act, adopt such-and-such law? The problems of interpretation mentioned above all arise in the process of trying to answer this question.
Despite this commonality, there are few specifically legal, theoretical analyses of this logic in the existing literature on federalism. In this article I will propose a new approach that draws inspiration from W.N. Hohfeld’s theory of jural relations and builds upon it in the context of Federal-State relations. As the article aims to provide a universal, structural analysis, it will not be limited to any one jurisdiction but will draw on the positive law of several different federations.
The analysis proceeds in three Parts. Part I addresses Hohfeld’s theory and shows how it is relevant to the study of Federal-State relations, despite not having been referred to in this context in the existing literature. Part II explores the limits of this relevance, however, and lays the conceptual foundations for a new theory that can explain aspects of federalism falling beyond Hohfeld’s reach. Here one finds non-Hohfeldian tertiary jural relations. Part III elaborates this theory in detail and shows its actual usefulness in the study of the constitutional law of federal systems, concluding with a perspective on how the proposed concepts can help illuminate the complex dynamism of the federal legal order.
Part I: Hohfeld’s Theory and its Relevance to the Study of Federalism
1. Hohfeld’s fundamental legal conceptions
A brief reminder of Hohfeld’s work is not out of place. His theory, first published in two articles in the Yale Law Journal in 1913Footnote 1 and 1917Footnote 2 respectively, postulates that all legal problems can be expressed in terms of eight fundamental concepts, which may be arranged in the following table:
The jural opposition of two concepts means that the presence of one in a person’s legal situation necessarily implies the absence of the other: if a person has a certain right, then this person does not have a no-right with respect to the content of that right. The jural correlation of two concepts means that the presence of one for a person necessarily implies the presence of the other for another person: A’s right with respect to something correlates to B’s duty with respect to the content of that right.
‘Right’ is considered synonymous with ‘claim’; a person has a right or claim that another person do (or not do) something, e.g., pay a certain sum of money. This other person has a correlative duty to do or not do that something. A ‘privilege’ is similar to what we usually call a ‘liberty’, i.e., the freedom to perform (or not perform) a certain action, e.g., walk down the street. It corresponds to a ‘no-right’ for another person that we not perform that action. Thus, if A has the privilege with respect to B to walk down the street, B has a no-right that A not do it.Footnote 3 A generalized freedom or liberty—A’s freedom with respect to everyone to perform an action—can thus be analysed in terms of a (possibly infinite) number of bilateral relations.
Hohfeld’s concept of ‘power’ is a person’s capacity to modify existing legal relations, and a ‘liability’ is the correlative susceptibility that another person has to having their legal situation modified. For instance, A’s power to arrest B implies the capacity to modify B’s rights and privileges. This subjection is not necessarily undesirable; for instance, A’s power to sell B a piece of land corresponds to B’s ‘liability’ to acquiring legal rights and privileges with respect to that land. Finally, a Hohfeldian ‘immunity’ is the freedom from subjection to a power, and correlates to a ‘disability’. B’s immunity from arrest corresponds to A’s disability to modify his (B’s) rights and privileges.
The ‘power—liability’ and ‘immunity—disability’ relations may be considered ‘secondary’ in that they have other relations as their content: the power to grant a right is correlative to the liability to have a duty imposed.Footnote 4 Right, duty, privilege and no-right are thus ‘primary’ concepts dealing directly with human actions, such as the act of paying a debt. This distinction will become particularly useful further on in this analysis.
2. Hohfeld’s concepts and public law
While he does not make any express statements on the subject, it appears that Hohfeld only seriously considered the application of his schema to private law legal relations. However, he does mention issues involving public law on a number of occasions, creating the implication that his theory is indeed relevant for them. Intuitively, this seems quite natural; the idea that public law actors have rights and powers, duties and immunities that may be analysed in Hohfeldian terms does not seem to be a stretch of reasoning. Indeed, many authors have referred to Hohfeld’s theory in public law contexts, especially with respect to fundamental rights.Footnote 5
Thinking of legal relations between governments within the same nation is nothing but a further extension. Since Federal and State governments are legal entities capable of entering into legal relations—between themselves and with others—there is no conceptual difficulty in applying Hohfeld’s ideas to them.Footnote 6 If a government can have powers, impose duties etc. with respect to its subjects, there is no reason it cannot do so with respect to other governments. As mentioned earlier, the expression ‘Federal-State relations’ is common; constitutions distinguish between the powers of the two levels of government; constitutional litigation involves them opposing each other in the courts. These legal situations can clearly be thought of in terms of bilateral jural relations between the Federation and the States.
It is important to note one major difference between Federal-State jural relations and jural relations between individuals. Strictly speaking, there are as many Federal-State relations as there are individual States; thinking of these relations in general as bilateral may occasionally lead to some confusion. However, it is a very useful conceptual shorthand, and is appropriate because jural relations between the Federation and one State (as clarified, say, in a judicial decision), are relevant for defining jural relations between the Federation and all the States taken together. A U.S. Supreme Court decision in a particular dispute between the U.S. Federal government and one State determining the scope of federal competence under the Commerce Clause is relevant for understanding the legal relations between the Federal government and all States. This is especially true when legislative competence is at issue. For present purposes, then, one may consider all the States or provinces in a particular federation together under the rubric ‘States’.Footnote 7
3. Applying Hohfeld’s concepts to Federal-State relations
Federal and State governments often find themselves in legal situations where their respective rights and obligations resemble those arising between individuals. Modern federalism is a complex phenomenon with various overlapping spheres of government activity and regulation where, even if the distinction between public law and private law is maintained, the fundamental nature of legal relationships in the two contexts is not always distinct. Federal and State governments have a host of administrative and financial relations and, further, each regulates activities in which the other might participate; these situations often mirror those that occur between individuals or between individuals and governments. The two levels of government impose taxes and other obligations on each other, enter into relations resembling private law contracts, enforce their laws against each other’s agents, and so on. We will see that Hohfeld’s theory helps break these situations down into their fundamentals.
Before analysing this in more detail, it is worth looking at a few extracts from judicial opinions in order to show that judges do indeed think using concepts similar to Hohfeld’s jural relations while deciding on federalism issues. This exercise is inspired by Hohfeld’s own emphasis on identifying his concepts in the processes of judicial reasoning.
One may proceed by looking at extracts from two famous decisions that need no introduction to students of Australian or American constitutional law: the Engineers decision of the Australian High Court and the McCulloch v Maryland decision of the U.S. Supreme Court.
The Engineers decision of 1920 involved an arbitral award concerning multiple employers operating in the entire Federal territory. In the State of West-Australia, three of these employers were government agencies or companies. The issue before the High Court was whether or not these public employers were subject to Federal (‘Commonwealth’) control exercised under the Federal legislative competence with respect to ‘industrial disputes’ [Section 51(xxxv) of the Australian Constitution]. The Court formulated the question as follows:
Has the Parliament of the Commonwealth power to make laws binding on the States with respect to conciliation and arbitration for the prevention and settlement of industrial disputes extending beyond the limits of one State?Footnote 8
The majority decision answered in the affirmative:
We therefore hold that States, and persons natural or artificial representing States, when parties to industrial disputes in fact, are subject to Commonwealth legislation under pl. XXXV. of sec. 51 of the Constitution, if such legislation on its true construction applies to them.Footnote 9
The Court’s reasoning in arriving at this conclusion was enormously influential in the later case-law governing all aspects of Federal-State relations. For immediate purposes, however, one need only note how the quoted extracts use concepts of power and liability that correspond to the Hohfeldian correlation. The challenged Federal power was the power to modify the rights, duties, privileges etc. of the States. Natural or artificial persons were not concerned except as agents or representatives of the States. And the States are subject to this power—i.e., they have a ‘liability’—as their legal relations may be modified by the Federal government exercising it. The legal situation of the Federation and the States in this respect can clearly be represented by Hohfeld’s ‘power—liability’ correlation.
The second extract from the judgment declares the existence of a State liability in general terms. In this case, the Federal-State relations in question revolved around an arbitral award; applying the Court’s decision involved the creation of many new rights, duties, privileges etc. A whole range of legal relations are subject to change. Further, the Court phrases its decision in general terms, creating a potential and undefined State liability; this liability will translate into specific concrete liabilities when other Federal laws that apply to the States come into force.
McCulloch v Maryland is another example where the Hohfeldian correlations are apparent. After determining that the Federal government was competent to establish a national bank, the Supreme Court had to decide whether or not the States could impose taxes on it. The answer was categorical:
The result is a conviction that the States have no power, by taxation or otherwise, to retard, impede, burden, or in any manner control the operations of the constitutional laws enacted by Congress to carry into execution the powers vested in the General Government. Footnote 10
The Hohfeldian analysis here is simple: the States do not have the power to tax the Federal bank—i.e., they have a disability—and the Federal bank has an immunity with respect to the States in the context of taxation. Since we identify the Federal bank with the Federal government itself, the ‘disability—immunity’ correlation in the Federal-State jural relation is evident. In the quoted extract, the ‘operations of the constitutional laws enacted by Congress’ are deemed to be immune from State taxation, but this should not confuse the underlying Hohfeldian relation: the State disability has as its correlative an immunity for the Federation, just as a State power would have had a Federal liability as its correlative.
Just as in the previous example, the Court’s wording—“retard, impede, burden or in any manner control”—is very general and goes beyond the actual legal situation at issue. It declares a wide Federal immunity that corresponds to a wide State disability, and this general and undefined correlation translates into many specific immunities and disabilities relating to specific legal acts at issue in different disputes. In this particular case, the Court determined a Federal immunity relating to the State act of imposing a tax on the printing of bank notes. In consequence, the States could not create a Federal duty to pay the tax, which in turn would have corresponded to a State right to claim the tax. The entire complex of Hohfeldian jural relations is present.
A. Intergovernmental immunities
These two canonical cases had great impact in their respective jurisdictions on the evolution of the doctrine of ‘intergovernmental immunities’ (‘immunity of instrumentalities’ in Australia), which is an important aspect of Federal-State constitutional litigation. In a modern federation, it is inevitable that “the legislative activity of one government may interfere directly with the activities undertaken by the other level of government, not as those activities affect individuals, but as they are carried on by the government.”Footnote 11 The legal issues involved concern only the two levels of government and no other parties, i.e., this is a directly bilateral legal relation.
What exactly are these ‘immunities’ that one level of government may claim (successfully or not) against the other? The first category of cases involve taxation, which is one of the main ways in which the exercise of governmental power can affect other governments. Since taxation is an incident of sovereignty, a system of government where there are two sovereignsFootnote 12 will always produce situations where a conflict arises due to one of them imposing a tax on the other.Footnote 13 As Marshall C.J. observed in the McCulloch decision, “An unlimited power to tax involves, necessarily, a power to destroy.”Footnote 14 Clearly, the Federal government imposing a high rate of taxation on a State enterprise—or vice-versa—can severely impede its functioning.
After McCulloch v Maryland declared Federal immunity from State taxation, the U.S. Supreme Court evolved a detailed and highly complex body of case-law on the subject. While it is not necessary to look at this case-law here, one may state that, generally speaking, the intergovernmental immunities in U.S. law are not balanced: Federal immunity with respect to the States—which correlates in Hohfeldian terms to State disabilities with respect to the Federation—remains much wider than State immunity.Footnote 15
One finds the same issues in the law of other federations, albeit to a lesser extent. Sometimes the constitutional texts themselves specify some intergovernmental immunities. Articles 285 and 289 of the Indian Constitution declare that the Federation (the ‘Union’) and the States have a reciprocal immunity with respect to taxation of government property, and that States have an immunity from taxation of their revenue.Footnote 16 Similarly, Section 114 of the Australian Constitution creates a reciprocal immunity from taxation on State and Commonwealth-owned property. An Australian scholar has however noted that there is a great ‘obscurity’ in Australian constitutional law regarding the question of immunity, both that of the Commonwealth vis-à-vis the States and vice-versa.Footnote 17
The second category of intergovernmental immunities is that of immunity from regulation. The power to tax is nothing but one aspect of legislative competence; intergovernmental taxation involves the legal power to impose the duty to pay a tax, while the wider category of ‘intergovernmental regulation’ involves the power to impose other duties, no-rights etc. through legislation. The fundamental jural relation of ‘power—liability’ is the same in both cases. A typical example of intergovernmental regulation arises when a public body engages in an activity that is regulated by another government. The Engineers decision involved precisely such a situation: the High Court held the States liable to Federal legislation, i.e., to the imposition of new duties, liabilities and so on. Before this case, just as it had done with respect to intergovernmental taxation, the High Court tended to favor the principle of a reciprocal immunity between the two levels of government. After the Engineers decision, by contrast, it moved to affirming the liability of the States to the exercise of Federal legislative power.
The Australian case-law with respect to the Commonwealth’s liability to State legislation is more ambiguous.Footnote 18 In some cases, the High Court has indeed refused to uphold the Commonwealth’s immunity from State legislation. In Pirrie v McFarlane,Footnote 19 it held that State traffic regulations could be enforced against a vehicle-driver for the Royal Australian Air Force. The State of Victoria’s prosecution of the driver, a Commonwealth employee, for drunk driving was upheld as valid by the Court’s majority decision. The Court here took the Engineers principle of non-immunity and declared it reciprocal.
The possibility of the Federal government claiming immunity from State legislation is well-established in the United States. According to Professor Tribe:
[i]n those rare cases where Congress has expressly granted regulatory or tax immunity to federal instrumentalities, agents, or contractors, the invalidity of any conflicting state law is definitively settled. But when Congress has not spoken, or has incompletely expressed its purpose, a residual immunity may still be inferred from the plan of the Constitution.Footnote 20
The Supreme Court has in general affirmed this Federal immunity. In a matter quite similar to the fact situation in Pirrie v McFarlane, the Court reached a conclusion very different to that reached by the Australian High Court. The case was Johnson v Maryland,Footnote 21 where the Court held that the State of Maryland could not arrest a U.S. Postal Service chauffeur for driving without a license. Since the chauffeur was performing his official functions, it was to be assumed that he had been considered capable of doing so (i.e., of driving), and so the principle of Federal immunity meant that the State could not apply its own regulations in this respect.
The Supreme Court is far more restrained in allowing State immunity from Federal powers. In National League of Cities v Usury,Footnote 22 it declared that the Federation could not exercise its powers so as to prevent the States from handling matters fundamental to their sovereignty. However, it later reversed this judgment in Garcia v San Antonio Metropolitan Transit Authority,Footnote 23 refusing to infer a doctrine of State immunity from the federal structure. It held that there was no restriction on the Federal government exercising its commerce power by applying its labor laws to local and State governments.
The Supreme Court of Canada has also had to consider how Federal norms apply to Provincial government activities. In Her Majesty in right of the Province of Alberta v Canadian Transport Commission,Footnote 24 the dispute concerned the applicability of Federal regulations to the acquisition by the Province of Alberta of a majority holding in an airline. Under the Federal laws existing at the time, the transfer of control of a commercial airline required approval from the Canadian Transport Commission; Alberta claimed immunity from this rule. The Court held in favor of the Province, relying on common law principles of Crown immunity from the operation of statute. The judgment did not expressly analyse Federal-Provincial relations, but it clearly upheld Provincial immunity from the Federal regulatory framework.
B. A Hohfeldian analysis of intergovernmental immunities
These examples of cases where intergovernmental immunities were at stake can all be analysed using Hohfeld’s table of correlatives and opposites.Footnote 25 In these matters, each level of government relates to the other in exactly the same manner as two individuals do to each other, i.e., each right, privilege, power or immunity corresponds to a duty, no-right, liability or disability. Thinking of these cases in terms of Hohfeld’s concepts shows that the ‘immunities’ at stake in fact involve a whole range of different legal situations, each involving a separate set of jural relations.
In Garcia, for example, the Supreme Court held that the Fair Labor Standards Act of 1938 (FLSA) would apply to local and State authorities. The San Antonio Metropolitan Transit Authority was thus bound to respect its provisions. Applying Hohfeld’s concepts allows us to see that this situation involves a multiplicity of legal relations, each of which may be represented on the table of correlations. Firstly, the fact that the FLSA applies to State authorities implies a jural relation of ‘power—liability’: a Federal power to impose labor regulations on the States correlating to a State liability to Federal regulation.
This secondary jural relation is related to several primary relations involving the first four Hohfeldian concepts, i.e., right, duty, privilege and no-right. And so, by virtue of its liability to the exercise of the Federal power, the SAMTA finds itself burdened with many new duties and no-rights with respect to its employees. For instance, it now has to pay a certain minimum salary and to compensate extra working hours; these duties correspond to the rights that employees now have to claim these benefits. It would also be possible to identify many secondary relations in this legal situation—i.e., those involving powers, immunities, liabilities and disabilities—such as the new liability for the States to respect the decisions of Federal authorities, to respect sanctions, etc. Hohfeld’s theory is useful here because it provides a flexible and yet rigorous conceptual framework in which each of these different legal situations may be analysed; flexible, because the analysis itself will depend on the specific legal aspect that one wishes to analyse; rigorous, because the concepts themselves and the relationships between them are strictly and systematically defined.
In the Canadian Transport Commission case, the Supreme Court of Canada determined that the Provinces had an immunity with respect to Federal rules regulating the transfer of the majority holdings in airlines. This meant that the Federation had a no-right when it came to demanding that the Province of Alberta seek the Commission’s permission for the transfer, which corresponded to the Province’s privilege to proceed without doing so.
In Pirrie v McFarlane, the driver (a Federal employee) was held liable to the operation of State laws, i.e., to State powers. Once we impute the Federal employee’s action to the Federation itself, the Hohfeldian relationships are clearly identifiable. The essential correlation is of ‘power—liability’, which then conditions various primary relations. The driver was held susceptible to the regulatory scheme of the State of Victoria, and was under the duty to have a proper driving license. In Hohfeldian terms, he did not have the privilege of driving without a permit, and so had the duty to obtain a permit. Similarly, it will be possible to analyse through this prism the powers, rights etc. of State agents arising by virtue of this legal situation, such as the power to arrest the driver, to forbid him from continuing to drive, and so on. Again, what is significant is that whatever the legal act chosen, Hohfeld’s correlatives will apply.
The above examples show that the expression ‘intergovernmental immunities’ as commonly used may refer in a given case to an underlying correlation of ‘immunity—disability’, but that it may sometimes be more helpful to think in terms of the opposite relation, i.e., ‘power—liability’. Further, and more importantly, using Hohfeld’s analysis reveals that these generalized, secondary relations actually translate to a multitude of specific relations, both primary and secondary. The study of the Federal-State legal relationship here would benefit from Hohfeld’s theory in exactly the same way as the study of private law relations between individuals does.
C. Intergovernmental ‘immunities’ in the broader sense: the limits of Hohfeldian analysis
But the expression ‘intergovernmental immunities’ is not always used in precisely this sense. In some contexts, we come across it in an altogether different situation, where the two levels of government do not behave toward each other in the same manner as individuals do. It is here used to describe a general non-interference in the exercise of one level’s powers by the exercise of the other level’s powers. To take the example of State immunities, the essential difference is that, here, the term ‘immunity’ refers not to the relationship the States have to the Federal government’s power/disability to impose duties, no-rights etc. on the States themselves, but to the protection of their own regulatory powers from the operation of Federal regulatory powers. Here is power/disability against power/disability respecting a third party. Unlike with the Hohfeldian ‘immunity—disability’ or ‘power—liability’ correlations, this relationship between the States and the Federation in their capacity as regulators is not a ‘direct’ legal relation but rather a function of each level’s legal situation with respect to third parties, i.e., citizens.
This becomes clear when we look at a concrete example. One may contrast two situations. In situation A, the Federation wishes to tax a State enterprise. The State claims an immunity from the Federal power of taxation. This is the situation in the cases discussed above: there is a direct legal relation between the two levels of government. In situation B, a State is keen to preserve its power of taxation with respect to certain activities, and claims that Federal taxation of those activities would constitute an interference with this power. Situation B is very different from situation A, because the ‘immunity’ claimed here is not with respect to the Federal government’s power to tax the State itself; it is not a direct legal relation that solely—or at least primarily—involves only the two governments. The State seeks rather to assert its own powers of taxation with respect to its subjects. To do so, it claims that the Federation has a disability, not just to tax the State but also to tax its own citizens. The legal relationship between the Federation and the States in situation B is far from evident, and not explained by Hohfeldian terms.
In fact, this configuration of legal relations between Federal government, State governments and subjects or citizens appears in the majority of federalism issues that come up in constitutional litigation. As will be shown in the next Section of this paper, Hohfeld’s theory is not helpful in these situations. A Federal power does not necessarily correspond to a State liability: the fact that the Federation could exercise its power with respect to its subjects and that this could affect—both practically and normatively—the exercise of State powers with respect to those same subjects—which happens all the time in any federation—cannot be clearly represented using the Hohfeldian correlation of ‘power—liability’. Even if it were possible to attempt to shoehorn this situation into the ‘power—liability’ relation, one would have gone far enough away from Hohfeld’s original conception to make it almost unrecognizable. Most importantly, it would certainly not help the clarity of the analysis, which after all has been the goal all along. It is thus necessary to devise a new theory to describe this kind of Federal-State relation.
Part II. The Foundations of a New Approach
The first step is to further contrast the two situations described above, i.e., the direct Federal-State relationship and the indirect relationship that arises between them in their capacity as regulators. The conceptual distinction here needs a terminological distinction: it is convenient to speak of the first situation—where Hohfeld’s concepts are directly applicable—as involving ‘Hohfeldian’ Federal-State relations, and of the second as involving ‘tertiary’ Federal-State relations. The reasons for the choice of the word ‘tertiary’ will become clear later. The second step would be to identify the concepts necessary for a proper theory of tertiary relations; simply adopting all of Hohfeld’s definitions and assumptions will not work. The third step would be to answer the question ‘why?’; it will be shown that this theorization is not an abstract exercise, but an approach that helps analyse several actual problems of constitutional law. It is particularly useful for comparative analyses, where the frames of reference, the evolution of case-law and even the terms used in different legal systems and cultures may make comparison difficult if not impossible. In this respect, the present article stays true to—indeed, is inspired by—Hohfeld’s instrumentalist philosophy, i.e., the idea that legal theories and concepts should have an application to real-life legal problems.
1. Why Hohfeld’s correlations aren’t enough
The reason why the correlations of ‘power—liability’ and ‘immunity—disability’ don’t seem useful in situations where both levels of government in a federation act in their capacity as regulators is that, in general, regulatory powers aren’t ‘aimed’ at other governments, they are aimed at the private individuals, companies, etc. subject to those powers. Other governments are included in their scope when they carry out activities similar to those carried out by the private subjects; as discussed earlier, these situations involve purely Hohfeldian Federal-State relations. For instance, a Federal law imposing a minimum salary requirement on all employers operating in federal territory might apply in equal measure to the State governments qua employers; this corresponds to the legal situation in Garcia v SAMTA, discussed above. In this case, the States find themselves in the same position as private subjects with respect to Federal legislative powers, i.e., they are liable to having their own rights, powers etc. modified by the exercise of Federal power.
But in most cases, regulatory powers aren’t directed at other governments qua governments. The powers and immunities involved are not ‘intergovernmental’ in any real sense. At the same time, there does seem to be some kind of jural relation here involving the normative situation of both governments with respect to their subjects. Federal and State governments are affected not just politically but legally by the extent of each other’s powers and how they are exercised concretely. A large number of Federal-State disputes in many different federations concern precisely this issue. A whole host of complex constitutional issues—the scope of Federal and State enumerated and residual powers, the interpretation of specific heads of power, the problem of implied powers, the meaning of supremacy—arise in this context, and a majority of the cases in which these questions have been addressed by constitutional courts involve this indirect legal relationship between the Federal government and the States (‘indirect’ because it depends on their relations with third parties.)
Let us look at a hypothetical example. Imagine a Federal power to fix the minimum price at which a certain product, say multivitamins, may be sold. Anyone involved in the sale of the vitamins—the manufacturer itself, wholesalers, retail sellers—is subject to the Federal power. Under the specific provisions of any Federal law adopted under this power, a whole range of persons may find their legal rights, duties, privileges etc. modified or new ones imposed—in the Hohfeldian way.
Here we have the crux of the problem: if the States are not themselves involved (through government agents or companies) in the commercial sale of the product in question, can we speak of a legal relationship between the Federation and the States in this context, and if yes, how? The Federal power to fix minimum vitamin prices doesn’t affect the State governments in precisely the same manner that it affects private subjects involved in that commercial activity of marketing vitamins (except to the extent that the States are also participants in the activity). We are now concerned with a situation where the States are not to be considered in the same category as these private subjects, but in a separate category in their capacity as regulators. The two roles are not mutually exclusive, but the legal relationships involved are very distinct. In the latter situation, one is no longer in the presence of a ‘directly’ bilateral relationship similar to the Hohfeldian relationships. If there is a strict jural correlation even in the absence of State participation in the regulated activity, it must necessarily come into being ‘through’ other relations, i.e., the relations between both governments and their subjects. The structure of these indirect jural relations is very different from that of purely Hohfeldian relations
2. The tertiary nature of legal relations between regulators
Using the adjective ‘tertiary’ helps bring out the specificity of these legal relationships with respect to the ‘primary’ and ‘secondary’ relationships represented in Hohfeld’s table, where the former consist of correlations between the concepts right, duty, privilege and no-right, and the latter of correlations between the concepts power, liability, immunity and disability. As discussed earlier, this second group of four concepts can be considered ‘secondary’ because they have as their content the modification or creation of other jural relations: a power is the legal capacity to modify other relations, a disability the lack of this capacity, and so on. The first four concepts are primary because they have certain human actions as their content, i.e., what legal actors are allowed or not allowed to do.
Keeping this distinction between primary and secondary relations in mind is useful when it comes to characterizing jural relations between the Federal and State governments in their capacity as regulators of citizen behaviour understood in a Hohfeldian fashion. These relations may be called ‘tertiary’ because they have secondary relations as their content, or, more accurately, because they depend on secondary relations for their existence. The specific nature of this complex of legal relations arises from their ‘triangularity’, which is illustrated below.
In the following diagram, F represents the Federation, S the States, and C the citizens (chosen over the word ‘subjects’ to avoid confusion).
In addition to all the limitations of a two-dimensional image, the representation of the subjects or citizens is obviously very simplified. But it is sufficient for present purposes, as we are concerned more specifically with the relationship between the two levels of government inter se.
In the diagram, the Federation (F) and the States (S) are regulators with respect to their common subjects/citizens (C). If the Federation has a power, for instance the power to impose a minimum commercial price for vitamins sold by some citizens, the relation F—C is a Hohfeldian relation of ‘power—liability’: the presence of a power for F corresponds to the presence of a liability for C, i.e., the liability to have its own legal situation modified by the exercise of F’s power. Similarly, if the States had the power to fix minimum prices for the sale of vitamins, the relation S—C would be a Hohfeldian ‘power—liability’ relation.
But the object of the present study is the relation F—S. The existence of this non-Hohfeldian jural relation derives from the fact that the relations F—C and S—C are not mutually independent; they have direct and systematic consequences for each other. They are interdependent because they have the same content, i.e., the legislative power—with respect to C—to fix the minimum price for the sale of a certain product, viz., those very vitamins. The relation F—S exists because of this interdependence of the relations F—C and S—C.
It is a ‘tertiary’ relation because its content derives from secondary relations each party has with others, i.e., the relation between each level of government and the subjects or citizens. The point is that when we speak of federalism, we are speaking of two governments, which implies someone who is governed. This is why it is essential to think of the triangle. However, the Federal-State relation F—S is still a strictly bilateral relation; the citizens are not themselves a part of this relation, even though it takes shape thanks to their relationship with the Federation and with the States. There would not have been any reason to draw a triangle if there were not three corners to it, but each side is still a line between only two of them. F—S is as strictly bilateral a relation as F—C and S—C are; it just cannot be analysed using Hohfeld’s correlations the way they can. Citizens are captured at the tip of the triangle as ‘power-liable’ to both governmental levels (with respect to commerce in those vitamins) in a way that F and S are not, which in fact entails that those two governmental levels themselves are in a non-Hohfeldian jural relationship with respect to their interdependent exercise of their own separate regulatory powers. The bilateral character of the F—S relationship is the reason why, as mentioned at the beginning of this article, the term ‘Federal-State relations’ is very appropriate for the study of federalism.
3. The exclusive nature of Federal-State relations
I have mentioned repeatedly that the relations between each level of government and its subjects are interdependent and that they have systematic consequences for each other. Indeed, the entire analysis undertaken here is based on this premise. But the premise itself needs to be justified more than it has been so far. Why exactly does this interdependence arise?
The answer is linked to what I would like to call the ‘exclusive’ nature of Federal-State relations. In private law, nothing prevents many different people from having similar powers. Several different bodies or individuals may have coexisting powers to nominate someone to an association or professional body. A may enter into three different contracts with B, C and D respectively, giving some power to each of them at the same time. These contracts are themselves created by the exercise of a legal power held by A and by the other contracting party. Everybody has the power to give a loan, to write a will, to get married. Infinite examples and varieties of legal powers can be imagined.
However, in the non-Hohfeldian tertiary jural relations, legal powers are exclusive to the two levels of government; the powers in question are regulatory powers, and thus are held only by regulators. Further, we are tracing these regulatory powers to the original legislative powers held by the State and Federal governments themselves, established by the federal constitution. All legislative power belongs by definition either to the Federation or to the States, or to both of them concurrently. Only the two levels of government can promulgate laws. Indeed, the exclusivity of legislative powers is an essential aspect of the way we conceive of the legal order—by consequence, the legal relations involving these powers are also exclusive, and they exist in their own sphere.Footnote 26 This is why the presence or absence—and not just the exercise—of legislative powers or disabilities for one level of government has a legal correlation with the presence or absence of powers and disabilities for the other level, and this correlation is precise and systematic enough to be rigorously analysed.
4. Essential concepts
Hohfeld’s eight fundamental conceptions are interdependent. One cannot define any one of the concepts ‘right’, ‘duty’, ‘privilege’ and ‘no-right’ without using the others; the same goes for ‘power’, ‘liability’, ‘immunity’ and ‘disability’. The inter-relationship between these fundamental conceptions is an essential part of their definitions.Footnote 27 Since the theory itself postulates these definitions, one cannot seek other definitions outside it. Further, it has been pointed out that Hohfeld’s notion of correlation does not really involve the association of two distinct things, but rather two different ways of saying the same thing, as in the use of the active and passive voice to describe an action.Footnote 28 A’s right to claim $100 from B and B’s duty to pay $100 to A are not two linked propositions, they are one and the same.Footnote 29
If the two legal attributes in a Hohfeldian correlation are nothing but two different expressions of the same thing, each attribute can have only one correlative. ‘Right’ is always the correlative of ‘duty’, and this correlation is part of the very definition of these concepts. A ‘power’, i.e., the legal capacity to alter legal relations, cannot exist without a correlative ‘liability’ for another person to have their relations altered.
But this creates some difficulties for the present study. If Hohfeld’s theory imposes its own definitions for its concepts, how can we move away from the theory without moving away from its definitions? If, as I have shown earlier, Hohfeld’s correlations are not applicable in all federalism-related situations, the concepts defined in terms of those correlations cannot themselves be adopted without careful redefinition. The most important of these concepts are those of ‘power’ and ‘disability’.
A. ‘Power’
The concept of power is obviously fundamental to any study of the division of powers in federal systems. Generally speaking, however, the term is not employed with a precisely outlined definition. Along the lines of a practice quite common in the theoretical literature, it is sometimes used as a synonym of ‘competence’,Footnote 30 which is itself a word omnipresent in the context of federalism. There are of course many variations of usage, and—importantly—linguistic differences and difficulties of translation.Footnote 31 But this situation is not ideal if the object is to develop a clear and coherent theory of Federal-State jural relations; a regular and precise terminology is required.
In the language of federalism, and especially in the context of the division of powers, one uses the words ‘power’ and ‘competence’ with two distinct meanings: a) the capacity or ability to do something, i.e., to carry out a legal act or bring about a legal situation by means of a law (or one or more provisions in one or more laws); and b) a matter/field/domain with respect to which the legislator has this capacity. This distinction between the material aspect of legislative capacity and the functional aspect is fundamental for a systematic classification of Federal-State relations, and indeed for any clear analysis of the case-law relating to the division of powers. For our purposes, it is convenient to retain the usage under which ‘competence’ indicates the field or domain and ‘power’ refers to specific legislative acts associated with a certain competence.
B. “Disability”
The concept ‘disability’ requires much less exegesis. We can simply affirm that the theory of tertiary relations retains the opposition of the two concepts ‘power’ and ‘disability’. Although we are moving beyond Hohfeld’s framework, his notion of legal ‘opposites’ is useful: indeed, the opposition of ‘power’ and ‘disability’ is particularly appropriate in the context of Federal-State relations. A disability is the negation of a power and vice-versa: the presence of a power with respect to a certain legislative act implies the absence of a ‘disability’ with respect to that same act. It must be emphasized however that the correlations of ‘power—liability’ and ‘immunity—disability’ are not useful for the understanding of tertiary Federal-State relations; only the opposition of power and disability is retained. This will become clear later.
C. Legislative ‘acts’
For Hohfeld, a legal power is the capacity or aptitude to effect a change in jural relations. Powers are omnipresent in legal relationships: the creation of contractual obligations, the delegation of rights or of other powers, the nomination of a person to a certain position, the possibility of making a gift and of revoking it, all involve the exercise of legal powers. As for each of Hohfeld’s concepts,Footnote 32 the content of a power consists of an act; it is the capacity to do something that has these legal consequences.Footnote 33
The definition of ‘power’ adopted here echoes the Hohfeldian emphasis on legal acts: a power is thus the capacity to do something that effects a change in legal relations. This emphasis on the act as an essential element of a power, i.e., to its exercise, is pertinent for the conceptualization of Federal-State legal relations because it allows one to clearly distinguish between the material and functional aspects of legislative capacity as described above, here defined as competence and power respectively.
Identifying a legal act is not particularly difficult in private law, where we are concerned with the specific actions of individuals or private companies, etc. But if one is speaking of legal relations involving legislative acts, the identification of the ‘act’ in question may appear problematic. Given that statutory provisions typically create or modify the legal relations of an indefinite number of persons in a multiplicity of different situations, what exactly is a legislative ‘act’? In other words, what principles or techniques can we use for defining and delimiting legislative acts for the purposes of analysing legal relations between governments, i.e., between legislators?
A formalist approach would seek to delimit an ‘act’ in terms of the formal statute itself or in terms of a discrete provision or set of provisions. For example, this approach could involve considering a single statute as a single legislative act, or a single clause of a statute as a single legislative act, and so on. Such an approach is obviously inadequate, since statutes and their clauses often create multiple norms addressing multiple situations. Further, a single ‘rule’ may actually be identified by reading two or more provisions in two or more statutes together.
But if one rejects a simple formalist approach, what is the alternative? The fact that one cannot define an ‘act’ in formal terms means that there is no universal rule of construction that can be applied strictly: if one cannot always identify a precise legal provision as containing or incorporating a legislative act, how is one to proceed in an analysis that depends on identifying these acts?
The solution is found in the terms of the problem itself: the legal acts forming the content of specific legislative powers (which are associated with specific legislative competences) need to be defined keeping in mind the specific situation at hand and the particular legal problem raised in a dispute. There can be no fixed guideline for how exactly one goes about this. The amount of detail and precision required is determined by the context.
Let us take an example. In Wickard v Filburn Footnote 34, the United States Supreme Court declared that “the power to regulate commerce includes the power to regulate the prices at which commodities in that commerce are dealt in and practices affecting such prices.” The competence with respect to commerce is attributed using the verb ‘to regulate’, which is in line with Article I Section 8 of the U.S. Constitution; from a purely grammatical point of view, to regulate is to perform an act. It would however be quite pointless to regroup all possible laws or rules relating to commerce within the outlines of a single legislative ‘act’. The formulation “to regulate commerce” is so generalized that it makes more sense to consider it not as a specific power relating to a specific act but as a legislative competence with respect to the field ‘commerce’.
However, the delimitation of legislative powers and competences is often less clear when they are described in narrower terms. In the same sentence from Wickard, the Court also refers to the power to “regulate the prices at which commodities … are dealt in.” This could be conceived of as a specific power associated with the competence with respect to commerce. But the power to regulate prices itself includes many different associated powers, concerning for example the establishment of rules or standards relating to the quality or the quantity of products, the creation of fines or other sanctions for infringement, the creation of bodies authorized to enforce them, and so on. It would also make sense to consider the power to regulate prices not as a specific power but as a legislative competence with respect to the field ‘prices’.
This shows how the definition of the legislative act in question in a given case is not fixed.Footnote 35 While analysing any given power, it may be convenient to discuss a large ensemble of legislative acts under one rubric, such as the power to regulate prices; in other cases, one may need a narrower delimitation of legislative acts (e.g., if we are dealing with concurrent Federal and State fields).
We are thus using an approach that emphasizes the relativism of legal definitions: the identification of something as an ‘act’ is not absolute but is done according to the needs of the moment. Surprisingly, there is a direct precedent for this not in the works of any modern legal scholar (to my knowledge) but in the writings of Bentham:
It has been every now and then made a question, what it is in such a case that constitutes one act: where one act has ended, and another act has begun: whether what has happened has been one act or many. These questions, it is now evident, may frequently be answered, with equal propriety, in opposite ways: and if there be any occasions on which they can be answered only in one way, the answer will depend upon the nature of the occasion, and the purpose for which the question is proposed.Footnote 36
Bentham distinguishes between simple and complex acts, with complex acts being made up of several simple acts that derive a sort of unity from their common purpose.Footnote 37 Legislative acts are very clearly complex acts, where several norms are laid down in order to achieve a certain aim. Once we adopt this idea of a legislative act and combine it with the flexible approach to identification discussed above, analysing legal relations formed around specific legislative acts does not encounter any conceptual or methodological obstacles.
Part III. Correlations Present in Tertiary Federal-State Relations
Two major correlations can be identified to describe the situation where the two levels of government act in their role as regulators: the correlation of exclusivity and the correlation of concurrence.
1. The correlation of exclusivity
In the hypothetical example of the possibility of fixing minimum prices, there are three conceivable situations: (1) only one level of government—Federal or State—holds the power; (2) both levels hold the power simultaneously; and (3) neither government holds the power. Situation (1) involves a correlation of exclusivity.
The language of ‘exclusivity’ and ‘concurrence’ is often used in the literature of federalism in conjunction with both ‘power’ and ‘competence’. It is important to re-emphasize the distinction made here between the latter two concepts: the exclusive character of a certain competence attributed by a text (i.e., capacity relating to a field or matter) does not necessarily imply the exclusivity of all powers (i.e., capacities to effect specific acts) apparently associated with that competence. The present analysis will focus on the exclusivity of legislative powers, i.e., the capacity to do something—which involves altering legal relations—by means of passing a law (or laws).
An exclusive legislative power is a legislative power that is held solely by one level of government in the federation. Because it is exclusive, it cannot exist simultaneously for the Federal and State governments (that would correspond to situation no. 2); if one level has a power, the other will have a disability; this disability is not created by the exercise of the first level’s power, but already exists in the legal situation of the second, e.g., when a certain legislative power is constitutionally attributed only to one level and not to the other. The fundamental correlation is thus one of ‘power—disability’. This is unique to tertiary Federal-State relations: in Hohfeld’s table, the only relationship between power and disability was one of opposition, i.e., a logical contradiction or negation when speaking of one person’s legal situation; here, ‘power—disability’ also becomes a correlation between the legal situation of two different actors. This allows us to highlight the unique nature of these relations, where the presence of an exclusive power for one level of the federation means that the other level has not a liability but a disability—the absence of a power. An exclusive power for the States implies a corresponding disability for the Federation and vice-versa. If the States have an exclusive power to fix minimum commercial prices, the Federal legislature has no possibility of doing so; if on the other hand the Federation holds this power, the States find themselves with a disability. This is precisely what we mean when we speak of exclusive powers in the constitutional framework of federalism.
The ‘power—disability’ correlation may be considered static, in that the actual exercise or non-exercise of the power has no effect on it. If the Federation has an exclusive power to fix minimum commercial prices for a certain product, the States have a disability, and this situation will not change even if the Federation does not exercise its power. Similarly, the Federal power will not be extinguished by its exercise: the Federation will retain the power to modify the minimum price levels once they have been fixed. External factors—such as judicial decisions or constitutional amendments—may replace or alter the ‘power—disability’ relation with respect to a certain legislative act, but there is no evolution within the terms of the correlation itself: if there is an exclusive power, there is necessarily a corresponding disability.
The ‘power—disability’ correlation is tertiary in nature because the power and the disability in question involve the legal relationship of each level of government not with the other level but with the persons subject to the power. The triangular image used earlier makes this clear. If we are dealing with a correlation of exclusivity, the structure of the legal relationships involved can be represented as follows:
We see that the tertiary relation F—S and the two Hohfeldian relations F—C and S—C are interdependent.Footnote 38 The legal act involved in each case—in the given example, the setting of a minimum price—is the same.
Listing out all the relations involved in this manner shows clearly that although ‘disability’ becomes a correlative of ‘power’ in the tertiary relation, it remains its opposite, since the presence of a legislative power for one level of government is the negation of a disability for that same level. The fact that disability and power are both correlatives and opposites shows the essential ‘either/or’ dichotomy of Federal-State relations involving exclusive powers. Either the States or the Federation hold the power in question, and in either case the other level doesn’t hold it, i.e., it has a disability.
It is important to repeat that the F—S relation cannot be explained in terms of a Hohfeldian ‘power—liability’ or ‘immunity—disability’ relation. If we are speaking, say, of exclusive powers for the States (as in the second set of relations listed next to the triangle, with F—S as ‘disability—power’), the Federation is not ‘liable’ to having its legal situation changed by the exercise of State power; it simply has a disability which exists even if the States don’t exercise their power; this disability is created by the federal constitution which denies a certain power to the Federal legislature. At the same time, the citizens are clearly liable to State power in the Hohfeldian sense (and immune from Federal power in the Hohfeldian sense). The nature of the F—S relation is thus different from that of the F—C and S—C relations.
2. The correlation of exclusivity in judicial reasoning
Due to the fact that the Indian and Canadian constitutions contain lists of both Federal and State or Provincial exclusive powers, express references to the idea of exclusivity are much more common in these two federations. The arguments are usually formulated in terms of the exclusivity of competences and not powers; one finds a succinct statement in Schneider v The Queen Footnote 39:
Once a matter has fallen under exclusive federal jurisdiction there is no room left for the province to legislate with respect to that subject matter.Footnote 40
The specific, exclusive legislative powers associated with an exclusive competence are to be identified in the particular context of each case. In the following passage, the Indian Supreme Court clearly identifies an exclusive power linked to an exclusive competence:
There can be little doubt that under Entry 45 of List I, it is the Parliament alone which can enact a law with regard to the conduct of business by the banks. … Setting up of an adjudicatory body like the Banking Tribunal relating to transactions in which banks and financial institutions are concerned would clearly fall under entry 45 of List I giving the Parliament specific power to legislate in relation thereto.Footnote 41
The identification of the field of exclusive competence—banking activities—is followed by the identification of a specific legislative power—the creation of the tribunal—associated with it.
The correlation of exclusivity is also present in the constitutional jurisprudence of other federations, albeit less clearly. In Gibbons v Ogden Footnote 42, the U.S. Supreme Court remarked that:
The genius and character of the whole government seem to be, that its action is to be applied to all the external concerns of the nation, and to those internal concerns which affect the States generally; but not to those which are completely within a particular State, which do not affect other States, and with which it is not necessary to interfere, for the purpose of executing some of the general powers of the government. The completely internal commerce of a State, then, may be considered as reserved for the State itself.Footnote 43
The delimitation of a sphere of internal commerce creates an exclusive State competence. Further, despite this delimitation in terms of a field of competence, the above passage admits the need for a delimitation in terms of specific powers: when the Court excepts concerns with respect to which ‘it is not necessary to interfere, for the purpose of executing some of the general powers of the government’ from the States’ exclusive sphere, it clearly envisages certain specific Federal powers relating to the field of commerce.
The static nature of the exclusivity correlation is also expressly recognized by constitutional judges, especially in Canada. The exclusivity of a competence translates to the exclusivity of a whole range of specific legislative powers associated with it, implying a whole range of corresponding disabilities. The Supreme Court of Canada observed in Johannesson v Municipality of West St. Paul Footnote 44 that:
The Judicial Committee having decided that legislation in relation to aeronautics is within the exclusive jurisdiction of the Dominion, it follows that the province cannot legislate in relation thereto, whether the precise subject matter of the provincial legislation has, or has not already been covered by the Dominion legislation.Footnote 45
The exclusive nature of the legislative competence relating to the field ‘aeronautics’ implies the existence of an indefinite number of exclusive legislative powers associated with it. The expression “the precise subject matter of the provincial legislation” can be understood as referring to a specific legislative act. Thus, the presence or absence of Federal legislation—i.e., the exercise or non-exercise of Federal powers associated with the competence relating to the field of aeronautics—has no relevance; the Provincial disability is static and is not modified by Federal action or the lack of it (as mentioned above, this is also why it is not helpful to think in terms of a ‘liability’ here).
This aspect is equally clear in the following extract from the Privy Council’s decision in AttorneyGeneral for Alberta v AttorneyGeneral for Canada Footnote 46, referring to legislation ultra vires Provincial legislative competence:
In such a case it is immaterial whether the Dominion has or has not dealt with the subject by legislation, or to use other well-known words, whether the legislative field has or has not been occupied by the legislation of the Dominion Parliament.Footnote 47
3. The correlation of concurrence
The correlation of exclusivity is not in itself sufficient to give an account of Federal-State jural relations. A significant part of these relations involves a relation of concurrence, which is a distinct correlation still within the tertiary relations discussed here but not involving the ‘power—disability’ relationship. Since Federal and State territories and subjects coincide, the coincidence or concurrence of Federal and State legislative powers is inevitable. Certain federal constitutions expressly provide for concurrent competences with respect to which both levels of government may legislate; in others, this concurrency is implicit in the constitutional texts or admitted in the case-law. Sometimes, however, the existence and the extent of concurrent fields remains unclear.
The absence of concurrent competences does not imply the absence of concurrent powers; it is quite possible that both levels of government are authorized to perform a certain legislative act even when there is no constitutional list of concurrent competences. On the other hand, the presence of such a list necessarily implies the existence of certain concurrent powers. For example, a shared competence with respect to the sale of a certain class of products implies a shared power to fix minimum commercial prices for the sale of a product belonging to that class. Since both Federal and State legislatures have the legal capacity to effect this legislative act, the ‘power—disability’ correlation is inapplicable; the presence of a Federal or State power does not correspond to that of a State or Federal disability.
But these situations still fit within the framework of tertiary Federal-State relations as represented by the triangle Federation-States-citizens. Putting two ‘power—liability’ relations into the diagram—because both Federation and States have a Hohfeldian power corresponding to a Hohfeldian liability for the citizens—it becomes quite clear that we are now dealing with a new tertiary correlation of two powers:
The ‘power—power’ correlation is the correlation of a Federal power with a State power respective to the exact same legislative act. We will see that this correlation is particularly useful for the study of many different doctrinal issues in the jurisprudence of federalism. However, apart from the different legal relationship involved, the correlation of concurrence is significantly different from that of exclusivity in that it cannot be considered static. Indeed, the dynamic nature of the ‘power—power’ relation is a fundamental aspect.
We have seen above that the relation of exclusivity does not depend on the actual exercise or non-exercise of the power in question. But this is not the case for the relation of concurrence, which contains within itself the possibility of the legal situation evolving; this evolution can take place not just due to an external event but due to the inherent configuration of the ‘power—power’ correlation. This will become clear when one considers the notion of the primacy or supremacy of the laws of one level of government—usually the Federation—over those of the other. The simultaneous existence of the exact same power for two different legislatures can create obvious problems if there is no mechanism to resolve conflicts between the legal norms they lay down; the doctrine of supremacy addresses this. For present purposes I will assume that Federal laws have supremacy over State laws.Footnote 48
The need to avoid conflict implies that the exercise of a concurrent power by one level of the federation can change not just the legal situation of the subjects but also that of the other level. Let us imagine that the power to fix minimum prices for the sale of a certain product is a concurrent power; the Federal-State relation involved is one of ‘power—power’. Now, if the Federation exercises its power and fixes a certain price, the States have lost their power to do so.Footnote 49 The Federal power continues to exist, since the Federation may modify its own rule and impose a different price. But the States have lost their power to legislate on the matter, or rather to carry out the specific legislative act of fixing a price. In other words, the State power has become a disability, while the Federal power remains. This ‘power—disability’ relation is of course a correlation of exclusivity.Footnote 50 The performance of the legislative act by the legislature having supremacy transforms the correlation of two powers into the correlation of a power and a disability, i.e., into a correlation of exclusivity.
Of course, in an actual constitutional litigation, the extent and nature of the legal power exercised by the Federation—and what the States are thus excluded from doing—will not always be clear. One will need to carefully delimit the ‘act’ that has been preempted by the exercise of Federal power. This is not something that any general theoretical study can do; it will depend on the peculiar facts and circumstances of each case. What the present study aims to do now is show the underlying legal mechanism that affects Federal-State relations when concurrent legislative powers are at stake. Like Hohfeld’s concepts, these correlations are universal and are necessarily implicated in the way we conceive of these legal relationships.
The particular character of the correlation of concurrence further shows itself in the disparity between the two levels of government. Only the level enjoying supremacy can effect a legislative act that has as a consequence the suppression of a power held by the other level. If Federal laws have supremacy over State laws, the movement from the correlation of two powers towards the correlation of one power and one disability happens only in the direction favoring the Federation, i.e., towards the correlation ‘Federal power—State disability’. The lack of parity becomes obvious if one considers a situation where the States act first: the exercise of State power does not create a Federal disability with respect to the legislative act in question. When the State level acts to fix commercial prices for a product in exercise of a concurrent power, the Federal government does not lose its own power to do so.
4. The correlation of concurrence in judicial reasoning
One finds a very clear expression of the correlation of concurrence in the following extract from the High Court of Australia’s decision in Clyde Engineering Co Ltd v Cowburn Footnote 51:
Where, therefore, a Federal dispute exists, no existing State law, whatever its terms, can indirectly or to any extent be regarded as presenting a legal bar to the full exercise of the Federal arbitral power. And equally, when in the absence of a State law on the subject a Federal award has been made, no State law can disturb or vary or affect the Federal adjustment of the dispute.Footnote 52
Both the asymmetric and the dynamic nature of the correlation of concurrence are evident: a State act cannot prevent a Federal act, but a Federal act may create a State disability.
Analysing situations of concurrence shows the usefulness of the scheme of tertiary relations presented here. It displays the dynamic of potential exclusivity inherent in concurrent powers thanks to the principle of supremacy, and thus provides a way to explain one of the most significant centralizing mechanisms in federal systems (given that supremacy is usually accorded to the Federal level). Further, thinking in terms of powers and acts allows one to appreciate the existence of issues of concurrence even when the existence of concurrent competences is uncertain. This facilitates an analysis of different areas of constitutional jurisprudence in a single coherent framework that can explain the common legal relations involved. To demonstrate this, one may take a look at some of these areas in the light of the proposed correlations.
A. Preemption and the occupied field
The issue of preemption and the notion of the ‘occupied field’ is the clearest example of the correlation of concurrence and its latent movement towards a relation of exclusivity. Indeed, as explained above, this movement is itself predicated on the doctrine of the supremacy of the laws adopted by one level of government in a federation. Preemption in all its formsFootnote 53 depends on supremacy and implies a prior concurrence of powers: the Federal and State levels are both in principle competent to carry out a certain legislative act, but when the Federal level exercises its power the States lose theirs—in other words, they find themselves with a disability vis-à-vis that act. The movement from a relation of concurrence to one of exclusivity is manifest.
The occupied field doctrine—or ‘field preemption’ in the United States—is nothing but an extension of this movement, in the sense that a Federal act is here considered to have created not just one State disability but several disabilities vis-à-vis several different legislative acts. The ‘field’ covers a multitude of different legislative acts for which the States earlier had powers but which are now reserved exclusively for the Federal legislature.
B. Implied powers
The term ‘implied powers’ may refer to several different legal phenomena, especially in the U.S. context.Footnote 54 The most common usage however is with respect to powers that are not expressly attributed but may be implied or inferred from the existence of certain powers that are expressly mentioned in a text. They are thus ‘ancillary’ or ‘incidental’ to the express powers. The ‘Necessary and Proper” clause in the U.S. Constitution refers to this kind of implied power, and several other federations have similar concepts in their constitutions or evolved by case-law.
The concept of implied powers illustrates how a simple two-dimensional vision of the division of legislative competence between the two levels of government is not workable, and how a conception of specific legislative acts as subjects of Federal-State relations is necessary. If one does not adopt this perspective, the idea of an implied power poses some conceptual problems. When a power is considered to be exclusive and one infers the existence of other powers necessary for it to be effectively exercised, are these implied powers also exclusive? If the answer is yes, what happens to the legislative capacity of the other level of government—since in some federations the doctrine of implied powers may be invoked in favor of either level—if it already held that power? And what happens if this power was held by virtue of a competence that was itself supposedly exclusive?
One cannot avoid these complications without insisting on the distinction between fields of competence and specific legislative powers with respect to specific acts. The notion that a law may not appear to touch directly upon a legislative competence but may nevertheless be justifiable as an exercise of an incidental power, does not create any conceptual difficulties. Indeed, it is the only way to explain this type of implied power without completely upsetting the overall scheme of the division of powers.
In this perspective, the use of the notion of an incidental power often implies an underlying correlation of concurrence with respect to the legislative act in question. For instance, the creation of penal sanctions in order to enforce a certain regulation may be considered ‘ancillary’ to the field in which the regulation operates, but it may also be considered an exercise of a power relating to the field of criminal law. If the legal competence for the two fields is attributed to different levels of government, it is easy to imagine situations where there is in effect a concurrence of powers, i.e., where both levels of government are competent to enact those penal sanctions. Thanks to the principle of supremacy, this may transform into a correlation of exclusivity when the power is exercised (by the legislature having supremacy).
Once again, one cannot usefully represent and explain this dynamic without stressing the idea of specific powers as distinct from fields of competence. Using the correlation of concurrence as an analytical tool shows the legal relationships involved in this situation, which indeed arises very often in practice.
C. The ‘double aspect’ doctrine
According to the doctrine of ‘double aspect’—very significant in Canadian and Indian constitutional law and present in somewhat different form in Austria as well—a certain legislative measure may be considered as pertaining simultaneously to fields of both Federal and State legislative competence because it has more than one ‘aspect’. It may be founded on a Federal competence under one aspect and on a State competence under another. This necessarily implies the concurrent existence of Federal and State powers even in the absence of concurrent fields of competence.Footnote 55 The power/competence distinction allows us to reconcile the exclusivity of competences with the concurrence of powers, i.e., with the possibility of one legislative act being validly adopted by either of the two levels of government.
The doctrine of double aspect can be invoked to justify both Federal and State laws. In the latter case, the exercise of a concurrent power so justified does not create any evolution in the underlying correlation of concurrence. In contrast, when the doctrine is used to justify a Federal law, it is in effect an exercise of a Federal power that transforms the correlation of concurrence into one of exclusivity: the State legislature loses its capacity to carry out that particular legislative act. Despite being apparently neutral in that it may help affirm the validity of laws enacted by either legislature, the doctrine of double aspect has a clear centralizing effect in a larger perspective of Federal-State relations. This has not been sufficiently remarked upon in the existing literature.
5. The correlation of two disabilities
The correlations of concurrence and exclusivity are the fundamental concepts necessary for an understanding of the constitutional jurisprudence of federal countries, and they underlie a vast majority of Federal-State disputes over competence. However, they are not exhaustive. The enumeration of possible correlations would be incomplete without reference to the possibility of two disabilities being correlated to each other, i.e., of neither legislature having the legal capacity to carry out a particular legislative act. The clearest example arises in the context of fundamental rights, where the legal situation of individuals is associated with the legal attributes of the entire federal State. The correlation stems from the common limits to governmental action, whether it be the Federal or the State government. Strictly speaking, the correlation does not involve Federal-State relations as they arise in the case-law, because here the two levels of government become indistinguishable and the division of competences is not directly in question. Without looking too deeply into these conceptual issues, identifying this correlative is useful because it further illustrates how the relationship between powers and disabilities in tertiary relations is multivalent. Plus, these are still legislative powers at stake, significant for our larger perspective of the federal system; there is of course a large body of literature on the relationship between the evolution of fundamental rights and federalism.Footnote 56
Conclusion
The above concepts provide us with a comprehensive view of the possible relationships involved in Federal-State relations. Each legislative act implies either a correlation of two powers, or of one power and one disability, or of two disabilities. Even in terms of simple logic, no other possibility can be envisaged. When these tertiary relationships are considered in conjunction with the Hohfeldian relationships described earlier, it becomes clear that each and every legislative act effected by the Federation and the States is part of a complex web of correlations between the two governments and the citizens. To say that each and every legislative act implies one of the given correlatives simply means, of course, that each and every legislative act may be represented using one of them; they are essentially instrumental concepts.
This paper has been based on the premise that technical questions arising in the positive law, especially in the interpretive approaches adopted by constitutional judges, are important by virtue of their impact on the balance of power between the Federal and State governments. The technical analysis of legal issues presented here is not abstract but has a direct political significance. Although, as with Hohfeld’s theory, the proposed correlations do not correspond to any empirical ‘reality’, they do help us understand such a reality, i.e., the kind of legislative acts that the Federation and the States are actually authorized to undertake in the constitutional practice of federalism. If the fundamental question in this practice is ‘Who does what?’, then the concepts of power and disability and their interrelationship are key to understanding the federal balance. They are useful not just in explaining the specific legal problems arising in specific cases but also in thinking more generally about how the federal legal order contains within itself the potential for evolution.
This systemic explanation of the federal structure has as a major component the correlation of concurrence. As shown, this correlation explains how the exercise of power by one level of government may gradually erode the powers of the other level. State powers are displaced by Federal powers, and over time this dynamic confirms the gradual centralization of almost all federal systems. The correlation of concurrence explains this inherent centralizing tendency. But it should be repeated here that this ‘internal’ evolution of Federal-State relations is not the whole picture: these relations may also be modified ‘externally’: constitutional amendments, evolving case-law, economic development, scientific innovations—there are many factors and events that may in some way or another impact the range of possible legislative action open to Federal and State governments. It is only when these aspects are looked at in conjunction with the interaction between the different roles played by these governments—regulators, participants in commerce, entities holding rights and subject to duties—that we may obtain a coherent and comprehensive view of the entire federal legal system.
