REMAKING RIGHTS CONSTITUTIONALISM?

This account of the hold of classical liberal politics of definition in modern con­stitutionalism is necessarily impressionistic. Given the argument from internal legal pluralism, it should not be taken as suggesting that this is a monolithic picture, and as I will shortly discuss, there are important counterexamples in the doctrinal record.

[I]f fundamental rights come to operate in the private sphere, the critique which labels them as vacuous bourgeois tools of legitimization whose function is to deceive citizens into believing in the justness of the system begins to lose some of its force.^[714]

Here, I put this argument to the test, by discussing the two principal means pro­posed for moving to counterhegemonic constitutional adjudication. The first focuses on extending the reach of constitutional rights by redefining who or what counts as the state, the second on deepening their scope by redefining the nature of constitutional obligations. When we consider how these doctrinal positions have been employed in actual adjudication, this has not effected any significant shift in the parameters of constitutional discourse, and so confirms the difficulty in practice of transcending the classical liberal default.

The Application to State Institutions (ASI) Model

The first proposed reworking takes its bearings from North America where the operative idea is that constitutional rights only speak to the institutions of the state. As such, constitutional rights do not apply to conduct which is ‘funda­mentally a matter of private choice and not state action.’^[715] The central doctrinal issue of the Application to State Institutions (ASI) approach is, having fixed a putative constitutional violation onto a discrete act, to ask if that act can be attributed to the state. Analytically, this approach breaks down into four categories:

1. The alleged unconstitutional action is carried out by a body which, by definition, is manifestly a state institution, so the constitution applies.^[716]

2. The action is executed through a body whose ‘public’ status is more ambigu­ous, but which on closer examination can also be seen to be a state actor.^[717]

3. The violation is committed by a private entity, but, because there is sufficient implication by the state in the ‘private’ action, the constitution applies.^[718]

4. The violation is by a private entity, but here there is insufficient implication of the state to invoke the constitution.

It is the cases in the fourth category where classical liberal politics of definition are most visible, as under this rubric both Supreme Courts have shielded a significant proportion of social life from constitutional review. For example, each Court has held that where an action is carried out by a private body under statutory authority, this may not in itself be enough to attach constitutional interest. In Canada, where a university had been created by, and exercised its powers (including its employment policies) pursuant to, statute, the Charter was held not to apply to its mandatory retirement scheme as this was neither instituted under statutory compulsion nor was it ‘following the dictates of the government.’^[719] In Flagg Bros v Brooks,^[720] the US Supreme Court found that the granting of a lien under a legislative property code was not state action, as the state’s providing the background statutory framework was no more than ‘acquiescence in a private action.’^[721] The US Supreme Court has also held that the failure of state-employed social workers to prevent a boy from being beaten and seriously injured by his father was not state action as the Constitution did not require the state ‘to protect the life, liberty, and property of its citizens against invasion by private actors.’^[722] The question of the courts’ status as constitutional actors has also arisen, with the Supreme Court of Canada, in RWDSU v Dolphin Delivery,^[723] adopting the position that courts are not covered by the reference to legislatures and governments, and so the Charter does not apply directly to their decisions.^[724]

We can see a number of classical liberal hallmarks running through this case law, whether in the characterisation of legislation as necessarily coercive (so that the absence of direct coercion equates to the absence of the state), or in cast­ing courts as ‘neutral arbiters’^[725] providing the framework for disputes arising from the free interaction of private individuals.

Extending the Reach of Constitutional Rights?

There are three counter-tendencies in the jurisprudence, which underline that we are again dealing here with a situation of internal legal pluralism. The first refocuses on courts as constitutional actors, and how their decisions should not be treated as something apart from the constitution. In the US, for example, the Supreme Court has had no issue in making the common law of libel conform to the First Amendment in a dispute between two private litigants.^[727] In Canada, the Dolphin Delivery decision makes this more problematic in direct terms. However, McIntyre J’s dictum in that case, that the common law should be developed in a manner consistent with Charter values, has been seen as the basis for ‘connecting] constitutional rights to private law’^[728] In this way, the Charter

can indirectly influence private action: in this regard, the Court held in Dagenais v CBC^[729] that common law rules on publication bans in criminal trials had to be reformulated to accord with Charter values, and in Pepsi-Beverages (West) Ltd v RWDSU, Local 55 8^[730] that the right to free expression meant that the common law could not be interpreted as making secondary picketing illegal per se.

A second strategy has been to separate out the private nature of an entity from the public character of its actions.

The third, and potentially most far reaching jurisprudential innovation, deals with the constitutional consequences of state inaction. Over seventy years ago, the US Supreme Court, in the classic ‘apple v cedar trees’ controversy in Miller v Schoene,^[735] held that where a legislature elects to do nothing, this can be ‘none the less a choice’ for constitutional purposes. More recently, in Vriend v Alberta,^[736] the Canadian Supreme Court followed this reasoning when, in hearing a complaint by a teacher dismissed because he was gay from a private college, it considered the Alberta legislature’s omission of sexual orientation as a forbidden ground of discrimination in its human rights code. The Court decided that the Charter applied, stating that it was not only engaged through positive acts, but also where a statute’s under-inclusiveness failed to fulfil its enacting legislature’s positive constitutional obligations.

These doctrinal moves echo critiques of a restrictive constitutional public-private divide within the legal academy. The idea that the courts are not state actors has been rejected as this ‘necessarily transforms [the] judiciary into an elitist and anti-democratic branch of government.’^[737] Also, a ‘formulaic’ approach to labelling entities ‘public’ and ‘private’ has been attacked for ‘insu­lating from constitutional scrutiny behavior fairly attributable to the state’^[738]— instead it is suggested that focusing on whether an entity furthers ‘governmental objectives’ provides ‘more accurate state action determinations.’^[739] Further, the notion that positive state action is required to invoke constitutional review has been criticised because while ‘it imagines that the state can abstain from deci­sions involving the content of the law,’ in reality ‘the state cannot abstain.’^[740] Various benefits are assigned to extending the reach of constitutional applica­tion: for example, where it is properly guided by constitutional values, private law can play a part in ‘fulfilling...

The Politics of Adjudication: Pluralist v Classical Liberalism

Underlying these positions is the idea that (to varying degrees) a more expansive ASI model can prevent the abuse of constitutional rights by private power^[744]— what has been described as ‘human rights at their most vulnerable point.’^[745] This gives us the opportunity, in the context of the broader discussion, to consider the counterhegemonic potential of a reformulated constitutionalism. My point of departure is to question the assumption, implicit in all the above approaches, that we are dealing here with an interpretive issue, to which there are ‘better’ or ‘correct’ responses, and which, if implemented, can remove the problem of doctrinal inconsistency. Rather, in adjudication on the reach of constitutions, we are dealing with the clash of different political ideas—specifically, between visions of the minimal and maximal state—and that, a fortiori, we should expect to find that this jurisprudence, in the words of O’Connor J of the US Supreme Court, ‘ha[s] not been a model of consistency.’^[746]

We can approach the political context of constitutional application jurispru­dence by considering the seminal US case of Shelley v Kraemer.^[747] The Supreme Court held there that the Fourteenth Amendment applied as the actions of courts in enforcing racially motivated restrictive covenants in real property bore ‘the clear and unmistakable imprimatur of the state.’^[748] The case has attracted considerable interest, not least because of its potentially far-reaching impact, as ‘all private action ultimately rests on the state’s willingness to enforce the civil and criminal rules that facilitate that action.’^[749] However, the Supreme Court has not extended this rule to the enforcement of discriminatory wills,^[750] and for some, it is impossible to imagine it applying to court orders enforcing a racially- motivated ban on trespass.^[751] There have been various attempts to reconcile Shelley with other parts of the state action doctrine,^[752] to show either that it is an anomaly^[753] or contains a broader principle, but poorly expressed by the Court.^[754] However, these contradictions are further evidence of internal legal pluralism, and are more plausibly explained by pointing to the different polit­ical values at stake in each instance, ie that the Supreme Court was less troubled about the private disposition of testamentary property than it was about the public sale of property tainted by racist restrictions.^[755] (This point perhaps explains our intuitive sense that had the university and hospital, in McKinney and Stoffman respectively, discriminated on the basis of race not age, the Supreme Court of Canada may well have found that the Charter applied.)

Focusing on the underlying political contests clarifies that the fault lines of constitutional application doctrine can be traced to two competing visions of the role of the state.

is no longer seen as the means by which individual initiative and self-reliance will be axiomatically rewarded in a fair and just manner; pluralist liberalism acknowledges that if left on its own, the economic market will generate injustice and inequality.^[759]

In other words, pluralist liberalism is both more likely to see the state implicated in ‘private’ activity, and to regard violation of rights by private power as deserv­ing of constitutional remedy. These different political perspectives can be seen to animate the doctrinal incoherence in this area. For example, under classical liberalism, courts can be depicted as the enablers of free interaction between individuals,^[760] whereas under pluralist liberalism, the question of their prove­nance in state action cannot be avoided.^[761] For pluralist liberalism, the delega­tion of political power can be more readily attributed to the state,^[762] whereas classical liberalism can regard its constitutional significance as attenuated by the delegate’s autonomy.^[763] Questions of power are more to the fore of a pluralist liberal inquiry into the presence of state action, whereas matters of form tend to dominate a classical liberal one.^[764] The constant ebb and flow between these different political perspectives partly explains why different conclusions seem to apply to very similar situations. However, focusing on the nature of these dif­ferences also explains the limits of pluralist liberal approaches, by highlighting important points of overlap with classical liberalism. In doctrinal terms, this means that it operates within the same framework that produces classical liberal results, and so always includes the kernel of a more restrictive jurisprudence.

The Politics of Definition: The Enduring Hold of Classical Liberalism

If we take first the argument that courts are under a duty to interpret private law in accordance with constitutional values, this does not necessarily lead to a dif­ferent outcome than where courts are excluded from the definition of govern­ment. In the post-Dolphin Delivery case of Dagenais,^[765] we saw the Supreme Court hold that common law bans on publicity in criminal trials offended the Charter, seemingly going beyond its previous non-application stance.^[766] However, in the Hill case,^[767] it considered that the common law of defamation was consistent with constitutional freedom of expression, maintaining in place the classical liberal ‘fence of privacy’^[768] around the reputation of the individual. In Dolphin Delivery itself, the Court (in obiter dicta) cast the relationship between the union and employer in terms of a contract between formally equal parties,^[769] and held that an injunction against secondary picketing complied with the Charter as secondary employers could not make contractual conces- sions.^[770] While in Pepsi-Cola v RWDSU, Local 558, the Court held that protec­tion from economic harm had no pre-eminent status, and that here the presumption that the common law would uphold the right to free expression meant that an injunction restraining picketing could not stand, it kept open the possibility that this delicate balance would be struck in different ways in later cases, for example, where the economic harm to third parties was ‘undue’.^[771] Thus, to regard courts formally as constitutional actors does not guarantee that at a substantive level classical liberal ideas will not continue to inform the outcome of cases.^[772] These cases show that courts may tend to find constitutional rights consonant with the private law values that they themselves (particularly in common law systems) have helped to shape^[773]: we should therefore bear in mind that these have generally been in the service of maintain­ing classical liberal ideals of individual freedom,^[774] through, for example, privity of contract and the protection of private property. As the history of trade unions and minorities’^[775] engagement with the courts shows, such values have not tended to be rich sources of counterhegemonic constitutional politics.

We can see the further hold of classical liberalism on judicial thought if we turn to the second strategy for extending constitutional application, ie to focus on the public nature of the acts of private entities. We can illustrate the limita­tions of this approach by returning to Eldridge in more detail. It will be recalled that this concerned whether a hospital’s denial of sign interpreters infringed deaf patients’ equality rights. As the Hospital Insurance Act^[776] under which medical care was provided did not preclude such services, the Supreme Court found category 1 of the ASI model inapplicable, as it was ‘not the impugned leg­islation that potentially infringes the Charter.’^[777] It also found the hospital did not fall into category 2, as (repeating its reasoning in Stoffman) it was not sufficient that it carried out ‘what may be loosely termed a “public func­tion.” ’^[778] It was therefore under category 3 that the Charter applied, as the definitionally private hospital, in ‘providing medically necessary services,’ was carrying out ‘a specific governmental objective.’^[779]

This approach has been favourably contrasted with Stoffman (showing why constitutional rights should apply equally to the providers as well as the recip­ients of publicly-funded hospital services).^[780] However, my argument is that both cases fall within the same broad framework, and so the Eldridge approach is of limited potential in doctrinal terms. At the heart of the act-entity distinc­tion is the idea that the state is only complicitous in some private acts: a public-private divide is still in place, the difference in Eldridge being that the Court draws the line closer to the pluralist liberal end of the spectrum. However, a future court following Eldridge would gain little guidance on how to answer questions such as how we differentiate between generic state func­tions, like health and education, where the Charter does not automatically apply, and specific governmental objectives, like providing medically necessary services, where it does. Such questions are left open by the Eldridge approach, and are postponed rather than answered. How they are answered, in immediate doctrinal terms, depends on the adjudicative contest between classical and plu­ralist liberalism: in Eldridge, the latter wins out (perhaps suggesting that the Court regards disability discrimination as more offensive to formal equality than age discrimination). This though leaves in place a framework which enables courts to draw the line in a different place, and the record shows that they frequently do in a manner that reflects classical liberal ideas of carving out a broader realm of individual autonomy.^[781]

However, even were a pluralist liberal approach to prevail, there is a strong sense, as in the response to Shelley, of inbuilt limits to its effectiveness in reach­ing private power. These limits were articulated by La Forest J in Eldridge when he discussed the constitutional position of corporations: while they are ‘entirely creatures of statute,’ the Charter does not apply to them because legislatures ‘have not entrusted them to implement specific governmental policies.’^[782] In its own terms, this is not self-evident—as discussed in chapter two, in the age of the global economy, state and corporate objectives are often closely intertwined. However, it does make sense when placed against the overarching hegemonic politics of constitutional definition, which reveals important points of com­monality between classical and pluralist liberalism. Both are premised on the primacy of the constitutional protection of individual freedom,^[783] but divide on the state’s orientation thereto. This dichotomy falls squarely within the frame­work of the state-civil society divide, which, as Santos notes, encompasses:

both the idea of a minimum and a maximum state, to the same extent that state action [is] simultaneously conceived as a potential enemy of individual freedom and as the condition of its exercise.^[784]

Thus, social life is divided between the state—whose conduct determines the condition of liberty^[785]—and everything else, which is placed in the realm of civil society. Crucially, this also includes economic activity, which is thereby separ­ated from the political, and assumed to pose no inherent threat to individual freedom. Under pluralist liberalism, the state will sometimes be regarded as enhancing freedom, but such action is seen as necessary to correct malfunctions in civil society, which is then returned to its natural state of equilibrium. In this way, classical and pluralist liberalism agree that economic power should not be subjected to the same direct constitutional scrutiny as state power.

We can now see that focusing on courts as constitutional actors, or imputing governmental objectives to the acts of private entities (as defined by the courts), does not necessarily lead to a more expansive constitutional application jurisprudence. This conceptual doubt is corroborated by the record of courts following these tests, which as often as not rely on a formal public-private divide to hold the constitution inapplicable to private relations. We can thus enter the interim conclusion that the hold of classical liberal politics of definition remains strong in the judicial mindset. What though of the third strat­egy for reaching private action outlined above, ie, applying the constitution to legislative and governmental omissions to protect rights? As previously stated, this seems to hold out the greatest prospect for counterhegemonic adjudication, as it has the potential to impose positive constitutional obligations on states to protect rights, whether or not these were initially infringed by non-state actors (as in the private community college in Vriend). In fact, it seems to mark a con­ceptual shift, in moving beyond asking ‘who did it?’ to ‘what was done?’ As such, it regards the constitution as applying to inconsistent laws rather than the actions of (constitutionally responsible) actors.

The Application to Law (LAW) Model

The LAW model has its roots in European traditions of constitutionalism, and regards the central application issue as whether the positive law of the state fully respects constitutional rights. Here, drawing on the jurisprudence of the European Court of Human Rights^[786] and the Bundesverfassungsgerichts, I out­line the principal features of the LAW model which some commentators see as providing the means of imposing basic constitutional limitations on private power. While, in terms of the outcome of cases, the LAW model can often be commended over the ASI approach, I conclude that it shares many of the latter’s shortcomings, limiting its potential for advancing a counterhegemonic form of constitutional politics.

Expanding the Scope of Constitutional Rights?

We can illustrate the major point of difference between the LAW and ASI approaches by considering the case of Costello-Roberts v United Kingdom.^[787] This involved a European Convention challenge under Articles 3 and 8 to cor­poral punishment administered to a seven-year-old boy at an independent school, which was financed from private tuition fees and received no direct government funding. The issue was whether and how this engaged the UK’s responsibility under a system of rights protection that speaks only to states. The European Court of Human Rights stated its general approach as follows:

The Court has consistently held that the responsibility of a State is engaged if a viola­tion of one of the rights and freedoms defined in the Convention is the result of non-observance by that State of its obligation under Article 1 to secure those rights and freedoms in its domestic law to everyone within its jurisdiction.^[788]

The Court thus concluded that the school’s actions could engage the UK’s responsibility under the Convention. Thus, what was crucial was not the qual­ity of the institutional or functional link of the private school in relation to the state^[789] (as, for example, in McKinney), but rather that the Convention placed a positive obligation on the state to secure, through law, the protection of its rights. Article 11 jurisprudence gives further examples of this approach, with the Court applying the Convention to the dismissal of workers under a ‘closed shop’ agreement, made lawful by domestic law, irrespective of the public or private nature of the employer.^[790] In Germany, the Bundesverfassungsgerichts, in the seminal Luth case,^[791] has echoed this broad approach:

... far from being a value-free system, the Constitution erects an objective system of values in its section on basic rights, and thus expresses and reinforces the validity of the basic rights. This system of values, centring on the freedom of the human being to develop in society, must apply as a constitutional axiom throughout the whole legal system: it must direct and inform legislation, administration, and judicial decision. It naturally influences private law as well: no rule of private law may conflict with it, and all such rules must be construed in accordance with its spirit.^[792]

This focus on how rights constrain the operation of all (state) laws has two important practical effects which seem to extend the scope of the application of rights significantly beyond the ASI model. First, the position of courts is much less contentious as attention centres on the law they interpret and apply, rather than their status as constitutional actors. In the Sunday Times^[793] case, there was an Article 10 challenge that an injunction against publication issued on the basis of the common law of contempt of court violated the newspaper’s freedom of expression. For the European Court of Human Rights, the English court’s insti­tutional status was irrelevant; rather, it regarded the issue before it as ‘whether the rules of contempt of court as applied in the decision of the House of Lords granting the injunction are a ground justifying the restriction under Article 10 § 2.’^[794] Similarly, in Germany the Constitution is said to have a ‘radiating effect’^[795] on private law, so that a judge ‘is constitutionally bound to ascertain whether the applicable rules of private law have been influenced by basic rights... : if so he must construe and apply the rules as so modified.’^[796]

The second, and potentially more far-reaching, consequence of this approach, is its implication of positive obligations on the state to protect fundamental rights, whether or not they are being infringed by public or private actors. For example, in Plattform ‘Artze fur das Leben’ v Austria,^[797] the European Court of Human Rights held that the Article 11 right to freedom of association required Austria to amend its positive law to ensure that private individuals could hold demonstrations without fear of physical attack from other private individuals.^[798] Such positive obligations can apply even in the ‘pri­vate’ sphere of family relations, with the European Court stating that, for exam­ple, the Convention’s Article 8 right to respect for family life could require ‘the adoption of measures designed to secure respect for private life even in the sphere of relations of individuals between themselves.’^[799] On this basis, the Court has found that a wife who had been the victim of domestic violence was deprived of respect for her family and private life when the Irish legal system denied her legal aid to petition for judicial separation.^[800]

There are limits, though, to how far the European Court of Human Rights is prepared to go in imposing positive obligations on states to remedy violations of rights committed by private parties. In Hatton v United Kingdom,^[801] the Grand Chamber rejected the applicants’ claim that their Article 8 rights to respect for their private and family life were infringed by aircraft landing at night at Heathrow Airport, and so disturbing their sleep. At issue was the UK Government’s scheme which permitted airlines to land aircraft at night provided a certain noise quota was not exceeded. While the Court reaffirmed the principle that the Convention may impose obligations on states to take mea­sures to protect citizens’ Article 8 rights where the immediate violation is by a private party, in the instant case it found that these rights could be restricted, inter alia, in the interests of the economic well-being of the country. Similarly, in Appleby v United Kingdom,^[802] the Court held that the United Kingdom did not fail to fulfil its positive obligations to protect the applicants’ rights to free­dom of expression where the positive law enabled a privately owned shopping centre to refuse access to a group collecting signatures for a petition against the development of nearby public playing fields.

However, notwithstanding these reverses, the general framework of the LAW model can be seen to go considerably beyond the approach of the North American courts. As one commentator depicts the contrast between US and German constitutionalism, the former is based on ‘the withdrawal of the Constitution from society,’ while the latter rests on a ‘general acknowledgement of affirmative constitutional obligations affecting society.’^[803] This more expan­sive approach, perhaps reflecting the broader social theory of European political traditions,^[804] has been favourably contrasted with the ASI model: for example, David Beatty has stated that under the former ‘few interests... are put beyond the reach of [courts’] powers of review,’^[805] thus extending the range of laws that have to be justified against standards of rationality and proportionality.^[806] The LAW model has also been commended as the better approach to the question of the application of rights in new constitutionalist jurisdictions, including the UK^[807] and South Africa.^[808] Others have emphasised how the European approach can meet the critical views that rights are empty and formal by shift­ing the focus from the state to the victims of human rights abuses,^[809] and so deal with the threat to human rights that ‘also lies in wait from the conduct of other private parties.’^[810] One writer sums up the social policy underlying the LAW jurisprudence as resting on ‘the protection of human rights of weaker individuals against infringement by private parties who wield systemic socio­economic power over them.’^[811]

The Counterhegemonic Limits of the LAW Model

While the LAW model transcends some aspects of the public-private divide at a formal level, it is more similar to the ASI approach than the arguments outlined in the preceding paragraph would indicate. The LAW model also operates within the framework of the state-civil society divide, and so in a number of important respects also takes its bearings from classical liberal constitutional politics of definition. Accordingly, this limits its doctrinal scope for engaging with private power in substantive terms; moreover, this suggests that symboli­cally, it is more likely to leave in place, rather than disturb, political attitudes that private actors are not a source of political authority. The first point of sim­ilarity between the ASI and LAW models is that the latter still sees the protec­tion of constitutional rights primarily in negative terms. If we return to the case of Plattform ‘Artze’, discussed above: while this seems to extend association rights beyond a strictly vertical relationship with the state, it is important to note the similarities of approach with the ASI model. Society is still viewed in terms of liberal social theory as a ‘competition among individuals and groups seeking to further their interests and conceptions of the good,’^[812] and the benefits conferred by rights are to protect such groups from interference in their private sphere, here manifested in their expression of political views. A fortiori this is the case where the European Court of Human Rights has applied Article 8, for example, to require the state to provide redress for a mentally retarded woman who had been sexually assaulted in a private nursing home.^[813] Again, here the Court’s application of the Convention to private relationships is principally concerned with enhancing individuals’ private spheres.

This is not to deny that the LAW model can bring important benefits to the parties in individual cases, but rather to argue that the ‘deep grammar’^[814] of judicial thought which accompanies it is not indicative of a significant change of approach from the ASI model. We can further highlight these similarities by considering some of the German cases on freedom of expression in private law cases. In Luth, an injunction had been issued against the plaintiff’s call for a boycott of the films of an anti-Semitic director on the basis of the damage to the latter’s reputation. The Constitutional Court applied Article 5 of the Basic Law, protecting freedom of speech, to rescind the order, referring to both the importance of ‘intellectual exchange and the contest among opinions,’^[815] and the collective interest in breaking with Germany’s Nazi past.^[816] In Blinkfuer,^[817] a newspaper company threatened to withhold its products from vendors who stocked the plaintiff’s pro-communist weekly publication. The federal High Court found the boycott protected by Article 5, but the Constitutional Court reversed this holding, stating that the use of the company’s economic power could deprive people ‘of their ability to draw their conclusions freely.’^[818]

Both decisions are strong examples of how German courts, in adjudicating actions between private parties raised under the Civil Code, will expressly weigh private law rights against constitutional values. However, while on the surface this can be contrasted with the formal exclusion of private disputes under the ASI model, at a deeper level there are various aspects common to both approaches. Johan van der Walt characterises the distinction between Luth and Blinkfuer as follows:

The boycott in Luth won the favour of the court because the political manner in which the politically motivated economic boycott was conducted could be reconciled with the court’s subordination of the economic to the political. The boycott in Blinkfuer failed to win the court’s favour because the economic manner in which the polit­ically motivated boycott was conducted, constituted an economic distortion of the political.^[819]

However, he questions whether these terms have quite the singular meaning implied by the Constitutional Court,^[820] and concludes that these cases reinforce ‘the instability of the distinction between the public and the private, the polit­ical and the economic.’^[821] This highlights that the dualisms which marked ASI jurisprudence have not disappeared under the LAW model, but have simply been relocated. For example, in Luth, we can see that the Court is choosing between pluralist liberal accounts of societal interdependence, so that some greater value may be served by enhancing Germany’s reputation abroad, and the classical liberal view of the primacy of the individual, and his or her right to a reputation. Also, in Blinkfuer, the Court is choosing between classical liberal ideas that economic power is beyond the reach of state regulation and the pluralist liberal notion that where this is used to excess, the state may step in. Thus, saying that the Constitution applies to all state law does not answer how these dilemmas are to be resolved. As with the ASI model, it leaves open the pos­sibility that courts may take a less expansive approach, and indeed German jurisprudence on expression later took a restrictive turn,^[822] with the courts more prepared to uphold private law rights against constitutional challenge.^[823]

Highlighting the common features between the ASI and LAW models returns us to the question of the broad parameters of constitutional debate that pertain to the latter. Of significance here is the LAW model’s view of the state, which, it is suggested, still embodies a dichotomous view of the latter’s relationship to individual freedom. While the state is seen as the potential guardian of rights when implementing its constitutional obligations, it is also seen as the ultimate perpetrator of rights violations, given its responsibility for the condition of the positive law. Thus, rights do not apply in private disputes because of the threat of a boycott by powerful economic actors (or the admission of beatings by a private school), but because the state permitted such breaches of individual freedom to occur. Linked to this is the important point that the LAW model rests on a distinction between, as the Supreme Court of Canada put it in Vriend, ‘“private activity” and “laws that regulate private activity.” ’^[824] Only the latter is of constitutional significance, and this provides us with a clue as to the sorts of obligations we can expect to be imposed on the state: these will relate to mat­ters within its authority, but not to matters which are deemed to fall in the free realm of civil society. Thus, while the state provides education, or health services, the LAW model can require it to ensure that the positive law treats everyone with equal dignity. However, as economic activity takes place in civil society, this is not a matter of state law, and this is why we do not see, and do not expect to see, the European courts imposing obligations on states to protect their citizens from the disabling effects of global capitalism (although we can expect to see them continuing to uphold corporate rights).