In most continental European legal systems good faith is a central concept of private law. It has its origin in Roman law and has found its way into virtually all European codifications. English law, on the contrary, traditionally does not know a concept of good faith and many English lawyers even regard it with hostility. At first sight therefore the concept of good faith could be an important apple of discord between English law and the European legal systems on the continent. For a long time this differences between common law and civil systems in Europe, however interesting from an academic point of view, have not been of much practical relevance. However, today things are different. Since the last few years European lawyers (mainly academics) have seriously been considering the possibility of harmonising or even unifying private law in Europe. If there is one day going to be a European code or restatement of private law the question will inevitably arise whether such a code or restatement should contain a general good faith clause and what this European concept of good faith would mean. These questions are addressed in this book.

As a consequence of further-going European integration in the '90s the question arose whether private law in Europe should be unified and how such a common European private law should look like. Over the last decade these questions have become the object of a new fast growing field of legal science which has resulted in a variety of books, specific reviews, research projects and conferences. Until now those who are enthusiastic for a unification of private law in Europe have been mainly academics, more precisely scholars in comparative law. This is easy to explain because in the preparation of a common European private law their speciality would, play an important role: teleological comparative law. National private law scholars have shown much less interest and sometimes even hostility. This can also be explained: the idea of giving up their own little kingdoms in favour of an international scientific debate is not attractive to them. For similar reasons, the legal profession has shown little interest: the differences between the various legal systems in Europe are for them a source of earnings.

The most frequently heard argument in favour of harmonisation of private law in Europe is that it is necessary for the realisation of the common market. The various differences between the national legal systems, the argument goes, are just as many impediments to the four freedoms of the European Union. Another (very strong) argument inverts the question: it being clear that - even in the absence of a common market - in modern times with intense international contacts the difference between the various systems of private law are a nuisance both for ordinary citizens and for trade, what justifies these differences? A justification - and therefore an argument against a common European private law - sometimes forwarded is that private law is part of national culture. However, particularly in fields of law like contracts, torts and property this statement seems to be exaggerated. Another objection against unification

is that the channel between civil law and common law would be unbridgeable. But many of the asserted differences seem to be based on caricatures of common law and civil law.

Scholars have proposed many different roads towards a common European private law: preparation of a code by legal scholars, introduction of a uniform European legal education, a clear political decision of the EU to start harmonisation, comparative preparation of national legislation, inspiration of national courts by other courts, creation of a common case law of Europe by the European Court of Justice, the foundation of a European Law Institute (analogy with the American Law Institute) which would prepare restatements. The approach most likely to be successful, and since there is no authority orchestrating the creation of a common European private law, de facto the most likely to be adopted, seems to be the one in which all these efforts are combined and thus stimulate each other (synergy). It should be emphasised, however, that these combined efforts should not be regarded (in a neo-Savignian way) as preparatory work for a European civil code since it is doubtful whether codification is the proper way of dealing with private law in the next century.

Most European civil codes contain a general good faith provision. In addition, some codes contain specific rules in which reference is made to the concept of good faith. Moreover, many specific rules in the codes are said to be special applications of good faith.

Most systems make a distinction between subjective and objective good faith. Subjective good faith is usually defined as a state of mind: not knowing nor having to know of a certain fact or event. It is of relevance particularly in property law. Objective good faith, the concept that the general good faith clauses refer to, is usually regarded as a norm for the conduct of contracting parties: `acting in accordance with or contrary to good faith'. Some systems have even emphasised this distinction by introducing separate terminology for objective good faith (Treu und Glauben, correttezza, redelijkheid en billijkheid). In France, however, such a distinction is not usually made. In all systems objective good faith is usually regarded as a normative concept. Good faith is often said to be in some way connected with moral standards. On the one hand, it is said to be a moral standard itself, a legal-ethical principle: good faith means honesty, candour, loyalty et cetera. On the other hand good faith is said to be the gate through which moral values enter the law. Reference is thereby sometimes made to the Aristotelian concept of equity. Actually, some systems do not distinguish between equity and good faith; they regard them as the same objective standard. It is then said that abstract rules may lead to an unjust result in a specific case, and that good faith may provide the basis for an exception in that particular case. Finally, in some systems good faith is regarded - and actually used by the courts - as a means through which the values of the constitution enter into private law.

At first sight, the theoretical status of good faith may seem quite unclear to an outsider since the terminology used by legal authors is far from unitary. Good faith is said to be a norm, a (very important) principle, a rule, a maxim, a duty, a rule or standard for conduct, a source of unwritten law, a general clause. To an English lawyer - often accused by his continental European colleagues of making inconsistent use of terminology - this may seem rather confusing. However, on closer inspection, the picture is less confused than it seems. It is generally agreed that a general good faith clause does not contain a rule, at least not one like most other rules in the code. It is not, like other

rules, susceptive of subsumption since neither the facts to which it applies nor the Iegal effect that it stipulates can be established a priori. Good faith is therefore usually said to be an open norm, a norm the content of which cannot be established in an abstract way but which depends on the circumstances of the case in which it must be ap plied, and which must be established through concretisation.

As said, the abstract standard of good faith must be concretised in order to be able to be applied. The court determines what good faith requires in the circumstances of the specific case (Einzelfallgerechtigkeit). However, the judge is not allowed to simply decide in the way which seems most equitable to himself. He has to determine the requirements of good faith in such an objective way as possible. In most systems, particularly in Germany, scholars both in private law and in jurisprudence have developed methods for rationalising and objectivating the decisions of the court. The purpose of these Methodenlehren is to render the application of the law in general, and of general clauses like good faith in particular, as rational and objective (and thereby predictable) as possible, instead of leaving it to the subjective judgement of the individual judge. The generally agreed method for rationalising is that of distinguishing functions and developing groups of cases in which good faith has been applied before (Fallgruppen). In doing so legal doctrine has developed an 'inner system' of good faith, which is regarded as the content of that norm. The result is a system of some times quite specific duties, prohibitions, (sub-)rules and doctrines which are all part of the content of good faith. It is said to have made decisions on the basis of the general good faith clause agreeably predictable (legal certainty) and rational. It is often emphasised, however, that concretisation will not, and indeed should not, lead to the fossilisation of good faith. First of all, it would be an illusion to think that concretisation wilt ever lead to a limited set of clearly distinguishable rules. But, it is said, more importantly good faith should remain an open norm in order to be able to continue to play its important role of making the law flexible. The inner system of good faith should not become a strait jacket.

In most systems good faith is said to have three functions: a function in interpretation, a supplementing function and a correcting function. In Germany in 1956 WIEACKER made an effort to specify the theoretical status of § 242 BGB. What he did was to assimilate the functions of good faith to those which Papinian had attributed to the praetorian law. Similar to the ius praetorium, WIEACKER said, good faith has three functions: adiuvare (officium iudicis), supplere (praeter legem), corrigere (contra legem). That trichotomy looks very similar to the distinctions made in most European systems: interpretation, supplementation, correction. Between the systems, and within each system, there is some difference of opinion about the exact definition of each of the functions. But it seems fair to say that in Europe usually three functions of good faith are distinguished, each of which correspond to one of the functions Papinian attributed to the ius honorarium with respect to the ius civile: 1) concretisa tion/interpretation; 2) supplementation (mainly of duties, e.g. duties of care, of loyalty, to co-operate, to inform); 3) correction/limitation (prohibition of abuse of right: ex ceptio doli specialis (praeteriti), tu quoque, venire contra factum proprium non valet, Verwirkung, dolo agit gui petit guod statim redditurus est, prohibition of excessive disproportion). Therefore, ignoring the subtleties, one could regard this trichotomy as the European common core.

Good faith has had great success in many European legal systems over this century. In most countries the number of cases where the good faith clause has been applied has grown explosively over the last few decades. Also the field of application has been growing considerably in many systems. In the various systems good faith has been applied in virtually all fields of general contract law (formation, invalidity, interpretation, 'implied' (i.e. heteronomous) obligations, privity, performance, non-performance and remedies, change of circumstances), with regard to many specific (nominate) contracts, and sometimes even far outside the field of contract.

As to general contract law, many systems recognise a general duty of pre-contractual good faith on the basis of which they usually recognise a pre-contractual duty to in form. Also they hold a party liable for breaking off negotiations in a manner contrary to good faith. Violation of the duty of good faith may lead to invalidity. For example, in many systems clauses (e.g. limitation clauses or standard terms or clauses in restraint of trade) are struck out on the basis of good faith. Further, violation of a pre contractual duty to inform, based on good faith (cf. above), may lead to invalidity for mistake or fraud. On the other hand, good faith may limit invalidity. The right to offer an adaptation in order to avoid invalidity is often based on good faith. And in some systems conversion of an invalid contract into a valid one is based on the general good faith clause. In most systems good faith also plays a role in interpretation. The importance of that role depends on how broad the concept of interpretation is defined. Usually the objective method of interpretation is based on good faith. In addition to that, in some systems there is a concept of supplementing interpretation (ergänzende Vertragsauslegung), also based on good faith, whereas in another system some kind of correcting interpretation seems to be adopted on the basis of good faith. Good faith has also been the basis of many heteronomous ('implied') obligations; in many systems good faith is regarded as one of the most important sources of heteronomous effects of the contract. The good faith duties most often incurred in the various legal systems are a duty of care, which protects the reliance interest, a duty of loyalty, which protects the expectation interest, a duty to co-operate, which was first proposed by DEMOGUE and was first 'codified' by the Principles, and a duty to inform, which does not only operate in the pre-contractual stage, but also during performance. In various countries good faith has also served as a basis for broadening the effects of a contract towards third parties. Cf., for example, in Germany the doctrine of Vertrag mit Schutzwirkung zugunsten Dritten, which considerably relaxed the privity of con tract by broadening the possibility of a third party to profit from a contract between two other parties. In other countries under certain conditions a party may invoke a limitation clause against a third party. In most systems good faith also plays a role with regard to performance. Actually, in most codes the wording of the good faith clause suggests that good faith is merely a standard for performance of the contract by the obligor. Furthermore, in many systems good faith also plays an important role in the field of non-performance and remedies. On the one hand, in many systems some of the remedies for non-performance are based on good faith. The rights to withhold performance (e.g. the exceptio non adimpleti contractus) form the most significant ex ample. In addition, the German doctrine of positive Vertragsverletzung is often said to be (ultimately) based on good faith. On the other hand, the exercise of a remedy may be limited by good faith. In many systems a party is not allowed to terminate the con tract or withhold its own performance for merely minor non-performances. Similarly, in France the invocation of a 'clause résolutoire' is limited by good faith. Also, the right to request specific performance may be limited by good faith. Finally, in the

course of this century a doctrine of change of circumstances has been adopted in most European systems, which is often based on good faith. Indeed, the problem of unforeseen circumstances is usually cited as the paradigm example of the operation of the concept of good faith and of the need for such a concept.

Good faith has not only been applied in the general part of contract law, but also with regard to specific (nominate) contracts. For example, in some systems the right (and its modalities) to end a contract concluded for an indeterminate period is based on good faith. Many other examples are to be found with regard to traditional con tracts like insurance contracts (which are usually regarded as contractus uberrimae fidei), labour contracts and lease contracts, but also with regard to more modern con tracts like solus agreements and on demand performance bonds.

It is often thought that good faith typically operates in contract law. Good faith is often seen, partly due to its collocation in codes of the French legal family, as a counterbalance to the rigour of the pacta sunt servanda rule. Although this may be an important role, in most systems the field of application of good faith is far from limited to contract law. First, many codes already themselves indicate that the obligor and obligee to any obligation have to meet the standard of good faith in their conduct. This means that a person who is liable in tort towards another person (e.g. because he has injured that person or damaged his property) or who has been unjustifiedly enriched at the expense of another person has to act in accordance with good faith. However, in addition to that, in some systems the field of application is even broader. It has been applied in, for example, property law, family law, company law, the law of civil pro cedure. In some systems good faith is said to operate in the whole of private law. In a few systems good faith even operates outside of private law, in administrative law, tax law and penal law. And also in international law good faith plays a role.

Immediately after the enactment of the codes in Europe it was thought that the only thing the courts had to do was to apply the law, which they could find in the codes. This led to the traditional civil law view that the courts do not create the law, but merely apply it. Today, however, it is broadly accepted that this traditional view may be in conformity with the ideals of the Enlightenment, but not with reality: also in civil law systems the courts do not only apply the law, but do also create it. As a matter of fact, in civil law as in common law systems, applying the law necessarily implies creating it. This follows from the simple fact that a court has to make a decision in every case that comes before it. In continental European legal systems the courts have to resolve cases by applying the rules from the code. This means that they have to apply abstract rules to concrete cases. A rule may be conceived as an abstract formula that links certain legal consequences to certain facts: if facts a, b, c occur the legal effect will be x. In applying a rule to a certain factual situation the judge may have to face three major problems. First, it may occur that he has some doubts as to whether the facts presented before him correspond to the facts as defined in the rule, or whether the remedy which the plaintiff requests corresponds to the legal effect as de fined in the rule. Secondly, it may be that not all the facts mentioned in the rule have occurred, but some others have, which he regards as equivalent, so that he finds it suitable that the same legal effect should follow in this case (or he regards the remedy requested as equivalent to the remedy mentioned in the rule). Thirdly, it may occur that all the facts mentioned in the rule have indeed occurred, but also some other facts which, in the eyes of the judge, make that the legal effect indicated in the rule should

not follow. Therefore the court may find it necessary to interpret (concretise). to supplement and to correct the abstract rule in order to reach an acceptable result. If a judge decides in a certain case to concretise, supplement or correct the law in a certain way, would he do justice if he refused doing so in a following similar case? The answer is no; in most systems it is thought that justice requires like cases to be treated alike. Therefore a court in civil law countries is bound to stare decisis just as much as a common law court. But if a court should decide in the same way as it decided in a precedent similar case, this means nothing less than that the court with its earlier decision has established a new rule, a rule which is a concretisation, a supplementation or a correction of an existing rule. Thus the concretisation, supplementation, and correction of the law has become part of the law. Therefore the conclusion must be that the application of the law necessarily implies creation of law. The same wish to treat like cases alike and different cases differently in accordance with the differences, also makes that the law - be it common law or civil law - is necessarily a system: rules are related to each other, according to the similitudes, and differences of the cases to which they apply. The difference between civil law and common law is that most civil law systems have (temporarily) fixed their system of rules in a code in which the sys tem is made more explicit; in most codes, particularly the German BGB, the Portuguese Código civil and the new Dutch BW, as a result of a higher degree of abstraction. The creation of new rules, (concretisations, supplementations, and corrections of existing rules) leads to new ramifications of the system. What has been said with regard to rules applies accordingly to the system. Cleariy the need to treat like cases alike may cause that as a result of the concretisation, supplementation, and correction of a rule other rules are adapted accordingly. Moreover, just like rules, the system as such (which is, as said, nothing more than the formalisation of the wish to treat like cases alike) may be concretised, supplemented and corrected. Again, this is equally true for civil law as it is for common law. The fact that in civil law systems the legislator at a certain point has fixed the system does not mean that it will remain unchanged and closed. Just as the legislator cannot make rules that will remain unchanged under application, he cannot make a system that does not change when it is applied either.

It is interesting to see that these three activities with regard to the law - concretising (or interpreting), supplementing, and correcting - which follow from the normal task of the judge (i.e. to apply the law), correspond to the three tasks that Papinian attributed to the praetor with regard to the written law in Rome. The praetor (and other magistrates) had to help (concretise), supplement and correct the ius civile, and the results of these three operations were regarded as new law which was called ius honorarium. As was said before, WIEACKER transplanted the Papinian trichotomy into modern German law. However, he did not do so in order to indicate the tasks of the judge with regard to the law in the code, or to show the relation between German judge-made law and the law in the code, but in order to indicate the functions of modern good faith. Good faith, WIEACKER said, concretises, supplements and corrects the law. And, as shown above, this view has not only been followed by most German legal scholars, it is also the common view in most European legal systems. But if the functions of good faith are the same as the tasks of the judge, i.e. interpreting, supplementing or correcting the law, does it then still make sense to regard good faith as a norm?

Good faith is usually said to be an open rule or open norm (see supra). From the use of the term 'open norm' one could conclude that there must be a category of open norms which can be distinguished from other norms which are not open. However, such a conclusion would not be correct: all norms are more or less open. Every norm

could be placed on a scale which ranges from totally open to totally closed: a norm is more open when less situations are excluded from its applicability. The closer a norm gets to one of both extremes, the less it makes sense to speak of a rule: a totally closed rule (all cases but one are excluded) is an order, a totally open rule (no case excluded) is equivalent to the (potential) law. Although all norms are more or less open there are some norms which are usually explicitly called 'open', like 'gute Sitten' in Germany, the 'zorgvuldigheidsnorm' in the Netherlands, 'a reasonable time' in the UNIDROIT Principles, and 'unfair term' in the European Directive. Often the legislator has deliberately formulated them in an open manner, usually because he was not yet able to determine which cases should be covered by the rule. The content of more open (or abstract) rules becomes clearer by their concretisation. Through application, sub-rules are developed which are interconnected. Thus a sub-system of the rule is developed. Therefore as a result of its application, every rule will become the top of a sub-system. Thus through a process of concretisation it becomes more and more clear what the content of the norm is. However, the particularity of the general good faith clause as it has developed in many European legal systems is that - unlike other 'general clauses' - it does not contain (or no longer contains) a rule, because it is completely open: with regard to the premises (the norm) as well as with regard to the effects it is totally open. Part 3 has given an impression of the variety of norms and legal effects that have been based on good faith by the courts. Statements, sometimes heard, that a general good faith clause makes the whole of contract law superfluous, or that the concept of equity could replace the whole law, are correct to the extent that in its field of application (cf. below) good faith may be applied in any factual situation to establish any legal effect. In other words: any rule could be based on it. There is no difference between saying 'good faith requires' and justice requires' or 'the law requires'. The cause of good faith. having become a completely open norm, and therefore having ceased to be a norm at all, probably lies in the fact that the content of good faith is said to consist of three functions which in reality are the tasks of the judge with regard to the law. Had the courts given real content to the good faith norm (e.g. leaving it a norm for the way in which an obligor has to perform), instead of calling the three main tasks of the judge its content, it would have been a real (open) norm. However, the assimilation of the functions of good faith with the tasks of the judge has resulted in good faith be coming a completely open norm. In every system where the content of good faith is said to consist of these three functions good faith is a completely open norm, i.e. no norm at all. Therefore, as it seems, because of the way in which good faith has developed in most European legal systems (see above), it can no longer be regarded as a norm. In reality, good faith is not a norm but a mouthpiece (a 'porte parole') for new rules (see for a few examples of new rules part 3). However, if the general good faith clause is not a rule, it does not make sense to say that a decision is based on it. In reality such a decision is based on the new rule. Therefore the court should mention the new rule, and formulate it as explicitly as possible.

One may ask why the courts mention the good faith clause when they develop new rules and doctrines. Why did the courts say - and are still saying - that they apply good faith instead of openly saying that they are interpreting, supplementing and correcting the law? Several reasons can be given. They all show that good faith has not only been a mouthpiece but also a cover. Judges in continental European systems have felt uncomfortable with their role as creators of law. First, there is the traditional civil law view with regard to the role of the judge: the judge has to apply the law (trias politica). Therefore the courts may have felt uncomfortable in creating law instead of merely applying it, and they were encouraged to feel this way by legal doctrine. Secondly, civil law judges may have felt somewhat at fault when they changed or created

new rules because the law created by them was not democratically legitimised. There fore, it was much easier for them, instead of openly declaring that they changed the law, to state that they merely applied it, invoking the general good faith clause in the code which had been adopted by the democratically elected legislator, and attributing to it the functions of concretising, supplementing and correcting the law. A third reason why the judge felt uncomfortable, and probably the most significant one, lies in the sanctity of contract. In many of the cases where the courts have `applied' good faith (particularly those in the first period) they have more or less radically interfered with the binding force of what parties had freely agreed to (the most typical example: unforeseen circumstances). They therefore preferred to say that the highest norm of contract law required that in the given case the contractual right could not be exercised rather than to bluntly state that in their view it would be just to accept an exception to the abstract rule pacta sunt servanda. Then there is the argument of convenience. It is much easier for a judge to just mention the general good faith clause as the rule he applies than to formulate the new rule (Die Flucht in die Generalklauseln). Finally, it should not be forgotten that many judges actually believed that good faith is a truly existing supreme norm, a belief that fitted perfectly in a European tradition of Aristotelian and Christian belief in natural law.

On the other hand, the question arises why the courts did not invoke the general good faith clause in all the cases where they interpreted, supplemented and corrected the law. Firstly, the courts only invoked good faith when they thought that the new rule could not realistically be based on another provision in the code. That is why good faith interpretation of the law is fairly rare. The courts usually only invoke good faith when they want to supplement or correct it. Secondly, if one looks at the (often explosive) expansion of the field of application and the number of applications of the general good faith clause over the last few decades it seems that the courts themselves have been asking the question: why not use good faith as a foundation for change in fields of the law other than contracts?

It is clear that good faith, as is the case with the judge, should fulfil these three tasks (interpretation, supplementation and correction) with regard to the whole of private law, not only with regard to contract law. The assimilation of the functions of good faith with the normal tasks of the judge therefore explains why it has proved to be impossible in most countries to limit the field of application of good faith. The idea that the field of application of good faith should be limited follows from the conception of good faith as a standard for conduct which is higher than the one prevailing between strangers, and which should therefore be limited to contractual and similar situations. If, however, concretisation, supplementation, and correction of the law are regarded as the functions of good faith, there is no reason why good faith should operate only in a limited field of the law. Any rule is subject to interpretation, supplementation, correction, or, to put it in terms of rights and duties: rights and duties are never abstract, but always subject to concretisation, supplementation and limitation. To give an example, it is clear that not only a contractual right but any right should not be abused (i.e. any abstract right has certain limits).

Part 3 has shown that the courts have developed many new rules and doctrines on the basis of good faith, which for an outsider at first sight do not seem to have much in common. Part 2 has shown that all these rules are usually regarded as sub-rules of good faith, as part of the content of the good faith norm. However, if, as maintained here, good faith is not a norm, and if the functions of good faith in reality are the normal tasks of the judge, it does not make any sense to regard these new rules as the content of the good faith norm. This also means that it does not make much sense to develop a sub-system of good faith. As said, the content of more open rules becomes

clearer by their concretisation into a sub-system of rules et cetera. However, the sub system of a completely open rule inevitably turns out to be a parallel system, which covers the whole field of law in which it applies, bot in an alternative manner. The result of the concretisation of gold faith would be just another system of law, parallel to the system of the code, but not self-sufficient. 'Good faith rules' have nothing more in common that distinguishes them from other rules than that the courts when adopting them have mentioned the general good faith clause as their legal basis. In particular, good faith rules are not more', just or more equitable or more moral than other rules. Therefore, the sum of good faith rules and doctrines has no inner coherence. In this respect 'the content of good faith' is very similar to Equity in old English law and ius honorarium in Roman law. Legal doctrine, instead of bringing the new rules developed on the basis of good faith in connection with each other, and becoming involved in discussions on whether a decision may be based on good faith and, if yes, on which good faith article or on which function et cetera, should relate them to the rules or doctrines that are concretised, supplemented or corrected by them. In Germany, the present commentator on § 242 BGB in the prestigious STAUDINGER commentary has come to the same conclusion. The 11 th edition (1961) of that commentary, by WEBER, counted 1553 pages. J. SCHMIDT, the present commentator, rightly says that ideally the commentary to § 242 BGB should be limited to a few phrases explaining that good faith is not a norm, and then for the discussion of rules which the courts have adopted mentioning § 242 BGB, refer , to the commentary to the roles that were changed by them. In addition to these the4retical objections, in many countries there will Boon be a practical need for abolishing the inner system of good faith. As a result of the enormous number of cases 'based , on good faith' it will no longer be manageable. How ever, this number will inevitably continue to rise in all systems where the tasks of the judge with regard to the law are regarded as the functions of good faith. More and more lawyers wilt question the sense of a distinction between the rules of the code and the rules that are said to be the content of the good faith norm. Therefore it seems likely (and indeed desirable) that the same will happen to good faith as happened to Equity and ius honorarium: when the distinction is no longer held to be justified and in addition wilt prove to be impractical, it will be abolished. German law may already have reached that point.

However, the rejection of an inner system of good faith does not mean that all the efforts made by legal doctrine over the last century have been useless. First of all, it has been of great importance that scholars have formulated the rules or even doctrines which were adopted in cases where the general good faith clause was 'applied'. Sec ondly, many parts of the inner system of good faith may be transferred directly into the system of the code, particularly into its general provisions, and, for contract law, into the chapter on 'heteronomous (implied) obligations' (in the Principles Galled 'Content'). That is not only true for rules, bot also for entire doctrines. This is shown where countries have adopted a new code: many doctrines that were adopted on the basis of good faith under the old code have been given their proper place in the new code, e.g. e.n.a.c, hardship, Schutzpflicht, duty to co-operate. Incidentally, if good faith really were a norm it would have been much more logical for the systems to have placed all the good faith sub-rules in one section under a heading like 'Good faith concretisations'.

The discussion as to whether good faith has a supplementing and correcting function has served as a facade for the question whether the judge may supplement and correct the law (first: the contract). In other words, the emancipation of the judge has taken place under the cover of emancipation of good faith and equity. The judge as a creator of new rules which supplement or correct the law has entered the scene cov-

ered by the mask of good faith. Today, all European code systems have accepted a supplementative and a corrective function of good faith; the pretensions of the En lightenment have been overcome. It seems that in this post-modern age the time has come to recognise that what the judge is doing is nothing to be ashamed of and that he can remove his good faith mask. Rightly, in some systems it has been pointed out that if there had been no general good faith clause the courts would have adopted the same new rules, although possibly somewhat later. Indeed, most of these new rules have been adopted in other systems without the use of the doctrine of good faith. On the other hand, still other systems with a general good faith clause did not adopt the same rule. This shows that the real question is whether the new rule should be adopted. In resolving this question the general good faith clause is of no assistance.

The main conclusion must be that if the role of the judge as a creator of rules is fully recognised, there is no need for a general good faith clause in a code or restatement of European private law. It may even do harm because it gives the courts an excuse for not formulating the rule which they apply. If, however, there is still some doubt as to the power of the courts, a good faith clause could be useful in order to assure that the judge may create new rules. This may be of particular importance for a new code for Europe where the European court may need extensive powers right from the beginning. It would then be logical not to put the article in the chapter on contract law, but right at the beginning as one of the first preliminary provisions of the code, just like in Switzerland. The wording would not matter much; experience shows that any phrase containing the words 'good faith' will do. However, if good faith were to have only such an Ermächtigungsfunktion, it could be argued that it would be more straightforward instead of using good faith terminology to provide expressly that the courts may interpret, supplement and correct the code where necessary. It may be argued that for tradition's sake the term 'good faith' should be used. However, since this term may lead to hostile reactions from common law lawyers (however unjustified) 'equity' may be an acceptable compromise, since it is part of both the civil law and the common law tradition. It is submitted, however, that this term has the disadvantage of having a strong natural law connotation.

Secondly, this paper shows that the concept of good faith in itself should not keep common law lawyers and civil law lawyers divided. On the one hand, common law lawyers should not fear the concept of good faith. The adoption of a general good faith clause in itself does not say anything about which rules will speak through its mouth. Good faith does not differ much from what the English lawyers have experienced with Equity. The real question is whether the rules adopted by the courts mentioning good faith should be included in a European code or restatement. It does not make any more sense for a common lawyer to fight the concept of good faith than it would have been to fight the whole of Equity. Rather, good faith serves as a guarantee against the rigidity that the English fear from a code. On the other hand, civil lawyers should not insist too much on including a good faith provision in a code or restatement of European private law. If they fully recognise what the courts do when they 'apply' good faith they should acknowledge that it should not be necessary that a court mentions the words 'good faith' when it creates a new rule which supplements or corrects the law1.

Common law lawyers do not believe that the law should be exclusively made by the legislator, nor do they consider it necessary that all the law should be democratically legitimated. This may explain why English law has not needed the concept of good faith.

Finally, it has become clear that it is not possible to say anything on the 'content' of European good faith without knowing the system that it will be operating in. Ideally, at the beginning it should be empty. All the rules mentioned in part 3, for example, if accepted, could (and should) be given their proper place in a code or restatement of European private law.

1The same is true for a (separate) doctrine of abuse of right: if the courts are only prepared to limit a right if they can say that the right is set aside by a higher norm (the prohibition of abuse), such a doctrine should be maintained. If, however, a court is prepared to say that the right given in abstract terms by the code is limited in certain situations, such a doctrine is not necessary.