This is the last piece in a series of comments offered on John Bell’s and François Lichère’s book French Contemporary Administrative Law (CUP 2022 – available in open access by clicking on this link) at a hybrid workshop hosted by the British Association of Comparative law and the Constitutional and Administrative Justice Initiative at the Essex Law School on 16th March 2023. The workshop brought into discussions Professor John Bell (UK – Cambridge), Professor François Lichère (FR – Lyon 3), Professor Giacinto della Cananea (IT – Bocconi, his comment can be accessed here), Professor Peter Lindseth (USA – Connecticut University, his comment can be accessed here), Professor Anne Jacquemet-Gauché (FR – Clermont-Auvergne, her comment can be accessed here), Professor Rob Widdershoven (NL – Utrecht, his comment can be accessed here), Professor Karl-Peter Sommermann (DE – Speyer, his comment can be accessed here), all experts in comparative and/or European administrative law. Professor Bell introduced his book in the first piece published in this series (available here). The following pieces have been the comments kindly contributed by the speakers. John Bell now concludes this series by replying to the comments.
May all speakers and authors be warmly thanked for the constructive and highly stimulating exchanges – a genuine academic sharing at the highest level.
* * *
We are very grateful to colleagues for the serious attention they have given to our work and the insightful comments that they have provided. The reader, as well as ourselves, is much enriched by their succinct comments. The following reactions constitute a few illustrations of the amount we have learnt from that others have written.
SOMMERMANN
We would agree with his first point that administrative systems are learning systems an operate internationally. To the many good examples he gives, we would add the importance of the OECD (Organisation of European Cooperation and Development) which is based in Paris and whose work contributes substantially to the circulation of informed analysis of administrative developments around the world.
Sommermann helpfully highlights questions about research methods when studying another legal system which have repercussions beyond our book. The description of the “binocular” perspective is helpful. In the one eye, one has the typical perspectives, terminology and rules of the target legal system. In the other eye, one is looking from the perspective of the concerns, terminology and rules of the observer’s legal system. As he says, getting equidistance on these two perspectives is not easy. Furthermore, the importance of European influences shows the connectedness of legal systems.
One particular feature that he dwells on is the role of case-law. The place of case-law in French administrative law has not just been something noticed by common law scholars, such as Goodnow, Dicey and Wade. It has been the subject of discussion about French administrative lawyers from a very early period. (See notably, L-A Macarel, Des Tribunaux administratifs (Paris 1828), para. 31.) In significant part, it results from the way the law, particularly on judicial review (le recours pour excès de pouvoir) grew out of the practice of the courts, rather than from any code of procedure. Rather like the development of the common law from the practice of the courts in Westminster Hall, the law developed by the Conseil d’Etat was incremental and documented in the form of case reports. Legislation offered very specific regulations, but lacked a sense of overview (see my article ‘Path Dependence and French Administrative Law’, in K. Gromek-Broc, Public Law in a Troubled Era. A Tribute to Professor Patrick Birkinshaw (Kluwer 2023 forthcoming).
The other point is whether there are parallel developments, despite differences in terminology. We think this depends on whether you focus on broad, general trends or on institutional detail. In broad terms, there is a common desire to control the administration. But once this becomes more detailed, then the issues are more divergent. France organises its administration differently from the UK and controls it also differently. Looking at the rules of judicial review without looking at the institutional context in which they operate is unhelpful. The role of the prefect as supervisor (and earlier a controller) of local government and the role of the Recteur de l’académie in relation to local education provide a context for how administrative law operates. We might ask what problems come before the administrative courts and why. An English scholar looks at the subject matter in the annual report of the Conseil d’Etat and wonders why there is so much litigation on civil service employment, a matter which is not as common in England, because of different structures for processing complaints.
That said, there are some important similarities. Most topics governing the activity of the modern administration have a legislative origin – laws and decrees. You only have to think of social security, immigration, planning and the like to realise that case law is marginal in defining what the administrator does. But, in terms of setting the broad lines of the ethos of the administration, then case law in both systems has an important role to play, but is not exclusive. As Macarel pointed out, legislation is just too detailed.
JACQUEMET-GAUCHE
Anne Jacquemet-Gauché’s comment on our book raises the important question of how far there is a dialogue between the French and the European judges. She asks whether the picture we present is overly optimistic. Lichère has done a lot of work on the subject of dialogue in the past. Bell has published a fuller analysis of the issue in K. Gromek-Broc (ed.), Public Law in a Troubled Era: A Tribute to Patrick Birkinshaw (Kluwer 2023), ch. 10. The key point we would want to make is that there is now considerably more engagement by French administrative judges with European Union law and the European Convention than there was when Bell started working on French administrative law in the 1980s. The existence of greater engagement does not mean that there is subservience or the denial of a distinctive French point of view. Conflicts and divergences are inevitable, even in good working relationships. So what we see in France is the recognition of the importance of the European dimension in French public law, but also attention to the primary function of French administrative law as regulating the operation of the French state and its relations with those resident in France, be they French nationals or others.
First, there is a much greater interaction between French judges and European judges now than forty years ago. It is helped by the fact that many of the French-appointed judges of Strasbourg and Luxembourg come from the Conseil d’Etat. This permits an important level of informal interaction, not only between members of the Conseil d’Etat and these judges, but also with their cabinets. There is also an intelligence unit within the Conseil d’Etat, the délégation au droit européen, that has formalised much of the understanding of the case law of the European courts for a French audience. The French Conseil d’Etat often collaborates in formal and informal ways with other administrative courts, particularly in the EU. The Strasbourg and Luxembourg courts are making more efforts to invite national judges to meet them in an informal way, giving an opportunity to exchange points of view outside a litigation setting.
Secondly, the current generation of senior judges has grown up with European law as a natural dimension of French law. It is not so alien as it might have appeared in the 1970s and 1980s. The French judges are literate in reading European materials. Indeed, many of them write commentaries in French law journals and give lectures to academic audiences. If you look at the speeches of the Vice-President of the Conseil which are on the Conseil d’Etat website, they are replete with references to European law.
Thirdly, being influenced by European law does not mean that they always agree. A big difference between common law judges and those in continental Europe is their attitude to precedent. The work Bell did with Neil MacCormick and Bob Summers in the 1990s made it clear that continental judges have a commitment to getting the right answer as a priority over following precedent (see D. Neil MacCormick and Robert S. Summers, Interpreting Precedents: A Comparative Study (Aldershot: Dartmouth, 1997). That means that lower courts are willing to refuse to follow higher courts – respectful disobedience is not unusual. So, it is perfectly normal in such a system that French judges will disagree with their European colleagues. That is a natural part of the system. The question that Jacquemet-Gauché puts is whether the French have a temperament of resistance to things that do not fit easily with the French way of doing things. We think there is a difference between a temperament of resistance and a resentment that the supranational court has not looked carefully enough at the national situation and the national way of achieving justice before it starts criticising. That is how we read the comment of Bruno Genevois, ‘Conserver l’apport du commissaire du gouvernement tout en prenant compte de la jurisprudence européenne’ AJDA 2006, 900.
Jacquemet-Gauché raises an important point in relation to enforcement. She cites the JMB litigation (ECHR 30 January 2020, JMB v. France, Appl. no. 9671/15, AJDA 2020, 1064 note Avvenire) in which France was condemned for the ineffectiveness of the remedies available for prison overcrowding which infringed human dignity under art. 3 of the Convention. Injunctions issued in 2012 requiring remedial work in prisons did not produce results. As a result, art. 13 of the Convention was breached. The defence of the administration was that they had to take all the prisoners they were sent and so could not, in practice, comply with the court orders to improve conditions at the same time. In response to the decision of the Strasbourg Court, the Conseil d’Etat did issue further injunctions on the administration to produce specific improvements: to provide covers for the exercise yards, privacy for toilets in cells with multiple occupants, to repair windows and improve lighting. But the Conseil did not immediately enjoin the use of containers to hold prisoners or require the provision of toilets in all exercise yards. When the prison authorities had not carried out its injunctions after two years, the Conseil did impose an “astreinte”, a daily penalty to enforce compliance. But it accepted that some work in relation to replacing containers might take until 2025, i.e. five years from the original decision. (See CE 19 October 2020, Garde des sceaux c Section française de l’Observatoire internationale des prisons, no. 439372, AJDA 2021, 694 with critical note by J. Schmitz ; CE 18 November 2020, Section française de l’Observatoire internationale des prisons, no. 439444 ; CE 11 February 2022, Section française de l’Observatoire internationale des prisons, no. 452354.) The problem continues to be litigated : see CE ord., 15 May 2023, Section française de l’Observatoire international des prisons, n° 472994. The problem in this series of cases is not that the French courts are particularly reluctant to enforce European law. It is rather that they have problems with enforcement generally. As we noted on p. 125 of our book, the number of astreintes issued by the administrative courts is very small in comparison with the complaints of non-enforcement registered each year. If there is a problem (as there is with prisons), this is not really a problem specific to European law, but a more general problem of ensuring the rights of citizens are enforced. Here, the claims of rights clash with the limits of public spending and that is the real problem at issue.
In her final section, Jacquemet-Gauché raises the question of whether the French Conseil d’Etat’s focus on the centrality of parliamentary law and the importance of the effectiveness of administration offers a contrast to the more Germanic Rechtsstaat and fundamental rights priorities which Strasbourg and Luxembourg enforce. First, we think that, whatever the pretensions of Brussels and Luxembourg, much of European Union law involves the coordination of national legislation. Member States vote for directives and then have to implement them through legislation on very practical matters that affect the operation of the Single Market. The primary issue is how the coordination of legislation is agreed and how much the national parliament retains an influence on the process. With regard to Strasbourg, the margin of appreciation allows national legislatures to continue to diverge in major areas. The diversity in Europe on the wearing of the veil is a good example. Members of the Conseil d’Etat who have been judges in Strasbourg do not give the impression that it is a difficult climate within which they have to operate.
WIDDERSHOVEN
Rob Widdershoven helpfully draws attention to the tension that exists in any legal system between (internal) path dependence and (external) influences such as Europeanisation. No legal system is an island in either time or geography. As Bell suggested elsewhere: ‘Legal development is explained not simply by the effect of social and economic pressures operating on the law from the outside at the current time, but also by the internal dynamic of the law, the pressure of established ways of dealing with issues.’ (J. Bell, ‘Path Dependence and Legal Development’ (2013) 87 Tulane Law Review 787) In a forthcoming article, Bell explains that the path dependence of French law lies not only in its doctrinal ideas developed by the Conseil d’Etat, but also in its court institutions, its personnel and its procedure. There is a body of established ways of doing things which shape the approach of the system to new problems. In one sense, Europeanisation has been a new feature relative to the established ways of regulating the administration which go back to the early nineteenth century. But, from the perspective of 2023, Europeanisation is now part of the contemporary established way of doing things. Though people who grew up in the 1960s and 1970s might remember a time when Europe was rather new and external, it has been a normal and ever-present dimension of the working life of almost everyone now active in the French administration and administrative courts. It has become part of the current path dependence. So it is often difficult to identify a single strand of influence.
The topic Widdershoven raises presents us with a tension in EU law. On the one hand, in the absence of specific EU remedies, courts of Member States are supposed to adopt the Marleasing approach to interpretation, namely that national law is read so as to make it compatible, as far as possible, with EU law (see Case C 106-89, Marleasing v La Comercial International de Alimentacion SA, EU:C:1990:4135). On the other hand, the principle of effectiveness requires that national remedies for breach of EU law be no less favourable than remedies for breach of domestic law (33/76, Rewe-Zentralfinanz eG and Rewe-Zentral AG v Landwirtschaftskammer für das Saarland, EU:C:1976:188, para 5). The former encourages alignment of national law with EU law. The latter sets national law as a baseline. What happens when EU law is less favourable to the claimant than national law?
As Widdershoven points out in section 2 of his comment, EU law has set its face against no-fault liability. He notes that the EU focuses on the protection of an individual in cases such as liability for the environment. As Widdershoven notes, our book emphasises that French law is focused on the fault of institutions, not of individuals, and that there is no need for a correlative duty (p. 240). As a result, illegality is treated as fault rather than breach of a duty of care or a duty of protection (pp. 248-50).
The problem of liability for courts is shown by the requirement in EU law that there is a ‘manifest’ breach of EU law, something which was not found in the case of the Conseil d’Etat in the Lectalis case (CE 9 October 2020, Lectalis, no. 414423). The retention of faute lourde for this area runs counter to the trend in national law (see our pp. 245-248) and arguably bolsters the wrong self-defence approaches of judges.
Similarly, the Francovich approach contrasts with the normal French administrative law view that the state is liable for imposing unequal burdens as the result of legislation (lawful or unlawful), a principle established back in 1938 in La Fleurette (see pp. 259-61). In 2019, the Conseil d’Etat established this kind of no-fault liability for unconstitutional laws. Logically, the principle of effectiveness in EU law would require the same remedy for laws passed in breach of EU law. Indeed, in Société Métropole Télévision (M6) CE 22 October 2014, no. 361464, para. 11, the Conseil d’Etat did seem to open the possibility of bringing a claim for liability for the effect of legislation in respect of EU law. Widdershoven is right in suggesting that there is a tension here with EU law, only avoided by the fortuitous findings of fact in each case that did not give rise to a decision in favour of the claimant.
Widdershoven is right that part of the problem is that we need to look at the package of conditions for liability together, rather than separately. Duty, fault, causation and damage are parts of a picture and different systems make use of them differently. The French (in public law and private law) tend to make more use of causation as a limiting factor, German and English law focus more on a duty of protection as establishing the scope of liability (see J. Bell, ‘The Reform of Delict in the Civil Code and Liability in Administrative Law’ in J.S Borghetti and S. Whittaker (eds.), French Civil Liability in Comparative Perspective (Hart, 2019), ch. 20, p. 432). It may be a question of looking at the result, more than the means by which the result was achieved. In a sense, national law should be judged by the comparable effectiveness of its results, rather than the exact comparability of its liability rules.
All the same, Widdershoven raises the question of how far a national judge should seek to be faithful to EU law to the detriment of national litigants. The French faced this with the product liability directive and were obliged to reduce the scope of normal national liability rules under the Civil Code so as to give effect to a maximising harmonisation under that directive. The Rewe principle is not a maximising harmonisation, but sets a baseline: the remedy for breach of EU law should be no less favourable than in national law. That does not prevent better treatment for litigants before French courts than for litigants before Dutch courts. But it does require an attenuation of the Marleasing approach to interpretation – EU law should not always be the benchmark of the best result.
LINDSETH
The major point that Peter Lindseth makes in his paper is whether ‘supremacy’, rather than ‘primacy’ is the best English term to describe the French approach to the laws of the European Union and the European Convention.
The starting point for the French is art. 55 of the Constitution, which initially regulated the status of both EU law and the European Convention. It provides: ‘Treaties or agreements duly ratified or approved shall, upon their publication, have an authority superior to that of statutes, subject, for each separate agreement or treaty, to reciprocal application by the other party.’ The position was clear that French constitutional law was the foundation for the authority of treaties under art. 55. By contrast, art. 88-1 of the Constitution, introduced in 1992, entails a specific approach to EU law as opposed to other international sources.
The terminology used in French scholarly and judicial writings does not make the marked distinction which Lindseth proposes between ‘supremacy’ and ‘priority’. Indeed, art. 55 uses another term. The unspecific nature of French terminology can be illustrated by a former member of the Conseil constitutionnel and of the Conseil d’Etat, Dutheillet de Lamothe (O. Dutheillet de Lamothe, ‘Le Conseil constitutionnel et le droit européen’ (2004) 57 Revue française de droit constitutionnel 23, 27). He described the position under art. 55 as creating a form of constitutional review operated by the ordinary courts:
In not ensuring itself the primacy of a treaty over a loi, but leaving it to the ordinary courts to ensure this supremacy, the Conseil constitutionnel has opened for these courts a new form of constitutional review.
Of course, from the strictly legal perspective, it is not constitutional review but review of compatibility with a treaty, and it does not have the same consequences. It cannot lead to the censure of a loi preventing it being promulgated. It results simply in the fact that the judge, in specific litigation, sets aside the national provision from loi or règlement which it considers contrary to European law. (author’s translation)
Here the terms ‘primacy’ and ‘supremacy’ are used interchangeably within the same sentence. As Lindseth notes, the CC decision no. 2004-505 DC used the term ‘primauté’ to discuss the status of EU law (see paras. 12 and 13). The same term was also used by the Etudes et Documents du Conseil d’Etat to describe the relationship with French domestic law (EDCE no. 58: Rapport Public 2007, p 284). But the Grands Arrêts de la Jurisprudence Administrative (21st edn 2019), p 598) heads its discussion of Nicolo with the label ‘Suprématie’, but then talks of ‘primauté’ of the treaty over a loi (p 600 and p 604, also 690, 804). It uses the label ‘primauté’ to denote the status of the Constitution in internal French law (p 684). ‘Suprématie’ gets used for both treaties and EU law on other pages: see pp 608, 623 and 684. Lindseth’s argument is prescriptive, not descriptive of French usage.
The key question Lindseth raises is what do the French public lawyers mean? ‘Primacy’, ‘supremacy’, ‘priority’ and ‘superiority’ are all used to denote that a treaty such as the European Convention or European Union law operate as what Raz describes as an ‘exclusionary reason’, a reason which allows the decision maker to decide by ignoring strong arguments that would otherwise have prevailed under the balance of reasons (J. Raz, Practical Reason and Norms, 2002, pp. 37-43, 190-194). These are not absolute reasons that prevail in all circumstances, but ones which normally exclude consideration of, in this case, the primacy of loi in French law. Whereas art. 55 of the Constitution clearly does not lead to the unconstitutionality of lois, art. 88-1 does at least require lois in conflict pro tanto to be disapplied. Furthermore, as Lindseth points out, CC decision no. 2022-841DC of 13 August 2022 declares that the correct transposition of an EU directive or the implementation of an EU Regulation is a constitutional requirement. The failure to comply with this will lead to a loi being sanctioned by the Conseil constitutionnel (paras. 4 and 5 of the decision). But this is not an absolute requirement. As Lindseth points out, the Conseil reserves the situation where a rule or principle affecting the ‘constitutional identity of France’. But in other cases, a loi which correctly applies a directive or regulation cannot be challenged. So the correct application of EU law is a strong exclusionary reason for ignoring the normal superiority of loi in French domestic law.
The French give as the source of this exclusionary reason the Constitution, especially since the 2005 referendum. They do not seek to give it an independent status as the result of some constitutional revolution. By contrast, in 1975, Procureur général Touffait preferred to base his argument not on art. 55 of the Constitution but “on the very nature of the legal order instituted by the Treaty of Rome”. The Court itself, whilst explaining that the Treaty had effect under art. 55, then argued that the Treaty inaugurated a distinct legal order within the legal order of Member States (Cass. Ch. Mixte 24 May 1975, Société des cafés Jacques Vabre D. 1975, 497 conclusions Touffait). To that extent, the French courts align themselves to the UK Supreme Court in Miller (no.1) in viewing the status of EU law as something the national constitution can change in the exercise of its sovereignty. Article 54 of the Constitution makes this clear. It provides that, if the Conseil constitutionnel has ruled that a provision of a treaty is incompatible with the Constitution, the treaty can only be ratified after the Constitution has been amended. It is by virtue of the constitutional amendment of 1992 (arts. 88-1 to 88-4) that EU law has any special status in French domestic law.
Lindseth rightly remarks ‘French public law (both administrative and constitutional) adopts a primacy approach. It reserves ultimate supremacy to the French Constitution, even as it recognizes that national apex courts should give strong deference to the doctrines and role of the CJEU in the interest of furthering the integration project.’ We would not doubt that. But the French do not take the concept of ‘supremacy’ in the same sense as the Americans. A country which lives with multiple supreme courts is different from a country with a unitary supreme court. Each court is supreme in its own sphere of competence. So the French approach to EU law and to the European Convention is really to ask first about the sphere of competence in which the problem is situated. Then it decides who has the supreme authority. The sphere of competence approach is neatly shown in the 2022 decision which came out after our book. Para. 5 states ‘Toutefois, la transposition d’une directive ou l’adaptation du droit interne à un règlement ne sauraient aller à l’encontre d’une règle ou d’un principe inhérent à l’identité constitutionnelle de la France, sauf à ce que le constituant y ait consenti.’ (CC Decision n° 2022-841 DC of 13 August 2022).
The reference to the ‘constituant’ reflects the approach of the Conseil constitutionnel to the Maastricht treaty where para. 43 of the initial decision on the constitutionality of that Treaty held that, in authorising a single currency, the Treaty deprived a Member State of the exercise of an essential condition for the exercise of national sovereignty (see CC decision 92-308 DC of 9 April 1992.) The solution was then for the French Constitution to be revised to authorise such a transfer of competence as permitted by a new art. 88-2 of the Constitution (see CC decision 92-312 DC of 2 September 1992, para. 35). European Union law is only competent in the areas in which the French constituent has transferred sovereignty. No absolute sovereignty exists, for even some matters cannot be changed within the regime of the current French constitution, notably its status as a republic.
So, whether we use the term ‘primacy’ in Lindseth’s sense or other words, we are denoting that EU law and treaties have a status above loi within their relative sphere of competence. Within the various supranational laws, EU law is special. Article 88-1 has been interpreted so as to allow a different approach than the one resulting from article 54 (that you could quote). In short, the Constitution shall prevail upon treaties, but EU law shall prevail upon the Constitution unless it harms the identité constitutionnelle, which has never occurred so far. Of course, who has the final say on whether a Member State has transferred competence is a matter of different views from the nation states and the CJEU. But, in this, France is not unusual.
DELLA CANANEA
The very clear and helpful comment by Giancinto della Cananea draws attention to the way in which there are similarities between English and French administrative laws. The original grounds of review (error of law and abuse of power) existed in eighteenth century English law and also were ideas taken over in the early years of the Conseil d’Etat from roots which existed in eighteenth century French law. Indeed, as David Ibbetson has pointed out, abuse of right is also found in French private law as early as the 1820s. We have ideas here which have their roots in notions of legality which pre-date the French Revolution or modern ideas of democracy (see P. Murray, ‘Judicial review for jurisdictional error of law in nineteenth-century certiorari and prohibition proceedings’ (Cambridge thesis 2013).
The more specific ideas of a right to be heard and proportionality are also old, but have taken on new meanings in the twentieth century. If the broad categories have long been recognised, their application has been contested. If the right to be heard in the deprival of rights or property has long been recognised, this has been less true of the failure to meet expectations or, indeed, within disciplinary contexts. Proportionality competes as a ground of review with narrower approaches in both systems. Both systems have had to accommodate ideas found in supranational law as they widened their own grounds of review. In that way, neither legal system is entirely generated internally.
Della Cananea also notes that French law has increasingly become legislative in terms of general principles contained in codes, such as the Code des relations entre le public et l’administration. English law tends to have specific statutes which may copy each other, but do not have visible commonality.
A final observation rightly points to the ebb and flow of the influence that French administrative law has had. If it had been in the mid-nineteenth century really the only show in Europe, by the mid-twentieth century, it had competition from at least the US and Germany. Furthermore, the supranational systems provided alternative sources of ideas, as well as norms. Ways of governing were not peculiarly national in many respects. There were common ambitions exhibited by the growth of UN agencies and bodies like the OECD where policy ideas were debated and national models from all over the world put under scrutiny. Inevitably, France has some good ideas and others will take an interest in these.
CONCLUSION
As the final observation of della Cananea makes clear, there continue to be good reasons why French administrative law continues to be worth studying by foreigners. The benefit of comparative law is to set national debates in a broader context. We are not searching for perfect models. But we are seeking to broaden our understanding of the questions which national administrative systems face and the factors which may shape their response.
The discussion between the comments from colleagues and our responses represent but a small illustration of the kind of comparative reflection on contemporary French administrative law which is desirable. It is for others to take on the issues and for their research to enrich us with their insights. We are grateful to Professor Yseult Marique for having hosted the discussion and for encouraging us to publish the results. This is but a small example of her contribution to comparative law in the UK and continental Europe.
Posted by John Bell (Emeritus Professor, University of Cambridge)
* * *
Suggested citation: J Bell, “Conclusions”, BACL blog available at https://wp.me/p80U0W-1tD.
British Association of Comparative Law 