ARTHUR S. HARTKAMP*

The private law of the Netherlands is firmly rooted in the "Civil-Law tradition," characterized by two features: its private law, notably the law of obligations, was mainly derived from Roman Law, and it is a codified System of law. The first codification, dating from 1809, came into being by the order of King Louis Napoléon, Brother of the Emperor. In 1811, after the incorporation of The Netherlands into the French Empire, it was replaced by the Code Napoléon itself. After the liberation from French rule in 1813, attempts were made to draft a national Code,1 which were crowned with success in 1838. Apart from a number of more or less important subjects in the law of persons and family law, the law of successions and the law of property, the Code of 1838 was largely, often literally, based upon the Code Napoléon.

By the middle of the 20th century both the Civil Code and the Commercial Code could be considered to be outdated an many points. Although it is true that the legislature had provided for several new statutes on subjects of social and economic importance, which were mostly incorporated into the Codes (such as juvenile law, lease and hire, labor contract, hire-purchase, company law), it had nearly entirely neglected to adapt the general parts of the law of property and the law of obligations to the needs of modern society. The ensuing gap in the development of private law was filled for the greater part by judge-made law. A major Innovation in this respect was the expansion of tort law: in the beginning of this century the relevant article of the Civil Code was construed in such a way that not only the violation of a right or an act violating a statutory duty constituted an "unlawful act," but also any act or omission violating a rule of unwritten law "pertaining to proper social conduct."2 On the basis of this new Interpretation the judiciary was 552 able to sanction all imaginable kinds of torts, ranging from traffic accidents to professional liability and unfair competition. The situation is similar to that in France where one general provision on tort provides the basis for a host of decided cases. Later on I will come back to tort law and to some other examples of innovative judgemade law.

As a result, legal opinion was sharply divided as to the road to be followed towards civil law reform. On the one hand, there were those who preferred to entrust the development of the law to the subtly differentiating and more concrete approach of the courts. On the other band, there were advocates of legislative change, in order to prevent the Dutch legal system in effect from slipping from a codified system of law - to a mixed system - such as those of several Scandinavian countries and eventually perhaps even to a System of case law.

In the end, the advocates of recodification carried the day. In 1947, Professor Meijers of Leyden University, who had long insisted upon the need for recodification, was entrusted with the mandate to draft a new Civil Code. This huge task, which was continued after his death in 1954, was carried on by several generations of successors. After some important intermediate steps - the coming into force of Book 1 (Law of persons and Family Law) in 1970, of Book 2 (Legal persons, including company law, associations and foundations) in 1976 and of Book 8 (Law of Transport) in 1991. 1992 has seen the coming into force of Books 3 (Patrimonial Law in General), 5 (Real Rights), and 7 (General Law of obligations), and several Special Contracts, including the law of sale.

I will not dwell here on the revision program as such, nor on general topics such as an outline of the system or the contents of the new Code.3 Suffice it to say that the new Dutch Civil Code from a 553 comparative point of view is unique as far as its system is concerned and also in the sense that it comprises nearly the whole of private law: civil law, commercial law (company law, insurance, all modes of transport law, etc.), consumer law, and labor law. The only important part of private law which has not been included as yet (although initially there is an Intention to do so) is the law of intellectual and industrial property. Moreover, apart from some exceptions, there is no private international law in the Code; there is a tendency, however, to codify that branch of the law as well. Finally, since the French era, procedural law forms the object of a separate Code.

What interests me is whether and, if so, to what extent, the two tendencies mentioned above - the codification approach and the case law approach - which (seem to be( at first glance seem to be opposite to if not incompatible with each other, have been combined in the new Body of codified law. The main objects of codification are clarity, certainty and predictability of the law, which are served by clear rules; and a convenient arrangement of legal rules, so that lawyers may know where to find the relevant rules for the matter at hand. On the other hand, our society is in full social, economic and technical development; given the reluctance (if not inability) of legislatures to supply new rules for new situations, the courts must have a certain leeway for innovative decisions and to act where legislation is absent - either for the time being or as a matter or principle. Portalis' "thousand unexpected questions" will not delay in presenting themselves again after the completion of the 1992 Code.

The Dutch legislature was perfectly aware of this dilemma, which of course is not surprising in the light of conditions prevailing at the time of the recodification effort and the lively debate between the two currents of legal thought referred to above. He solved it in a way which initially has caused some uneasiness to both of them but which recently has met with more approval.

It is outside the scope of this paper to deal extensively with the aspect of certainty and predictability of the law as set out in the new Civil Code. The Code as it now stands numbers some 3000 provi- 554 sions (articles), and this number will increase when the new law of succession (Book 4) is enacted and Book (...( (Special Contracts) is finished. Such a host of articles will undoubtedly provide many clear answers to numerous problems and questions which are already known to exist in practice or which the present-day legislator has been able to foresee.

The system of the Code, too, has been framed in such a way - noteworthy is its organization according to a strict pattern of general rules preceding more detailed rules, sometimes in various layers4 - that after some training it should not be too difficult to find one's way to the relevant section or chapter. It is more interesting to see how this huge body of substantive rules is balanced by open-ended concepts (general clauses, blanket formulas), intended to warrant the freedom of the courts in the sense mentioned before. So I will proceed to present a survey of the most important of those concepts, indicating briefly their purport and their function.

The concept of good faith permeates all branches of the Dutch law of obligations and contract law (and, as we will see below, other branches of the law as well).

Art. 6:1 para.5 provides that both parties to an obligation should behave in their relationship according to what is reasonable and equitable. Art. 6:248 para. 1 provides that contracts not only have the effects expressly agreed upon, but also those, which, according to the nature of the contract, result from the law, usage or the requirements of reasonableness and equity.

The principle of bona fides or good faith (expressed by the complementary notions "reasonableness and equity")6 has three func- 

tions. First, all contracts must be interpreted according to good faith. Second, good faith has a "supplementing function": supplementary rights and duties, not expressly provided for in the agreement or in statute law, may arise between the parties. Third, it has a "derogating" or "restrictive" function, expressed in article 6:248 Paragraph 2, stating that a rule binding upon the parties does not apply to the extent that, in the given circumstances, this would be unacceptable according to criteria of reasonableness and equity. So good faith may, in a given case, extinguish rules prevailing between the parties or exclude their application. For a long time, it was sharply disputed in the Netherlands whether the requirements of good faith (reasonableness and equity) could lead to that result. Under the old Code, which like the French Code (article 1134 Paragraph 3) merely provided that "contracts must be performed in good faith," the Dutch Supreme Court (Hoge Raad) has long been reluctant to admit such a possibility, with the exception of cases in which a party may be estopped from asserting a contractual right because of his own conduct incompatible with his exercising that right. However, in more recent years the courts have construed the rule of the old code in the sense of the new one, such as by refusing a debtor the right to rely on an exemption clause which in the circumstances of the case, turned out to be unreasonably onerous ( "unconscionable") towards the creditor.

In the new Code, this latter case is to a large extent covered by the chapter on general conditions (article 6:231-247), which is in its basic structure and in a number of details modelled after the German "Gesetz zur Regelung des Rechts der allgemeinen Geschäftsbedingungen" [Standard Terms Act] of 1977. Article 6:233 provides that a stipulation in general conditions may be annulled (either through an informal declaration by a party to the contract addressed to the other party or by judgment) if, taking into consideration all the circumstances of the case, it is unreasonably onerous to the other party. But where that chapter is not applicable, e.g., because the unreasonable clause is not part of general conditions or because it is included in an international contract or in a labor contract to which that chapter does not apply (art. 6:245 and 247), the general rule of good faith regains its strength.

It is important to note that the "derogating" function of good faith is not limited to clauses of the contract, that is to say to the rights and obligations deriving from those clauses. The rule applies 556 to all rules binding upon the parties as a result of the contract. This includes all relevant statutory rules which govern the parties' relationship, irrespective of whether they are of a supplementary or of a mandatory character. So the courts may derogate from a mandatory rule of law, such as in the law of lease or consumer law, where the party protected by that rule would, in the circumstances of the case, act contrary to reason and equity by relying on it. In this sense all rules created by the parties or by statute have been brought under the control of the courts,-not, of course, to judge their intrinsic fairness, but to decide whether their application in a concrete case would lead to an unjust result.

The problem of a derogating function of bona fides has proved particularly urgent in cases of frustrated contracts7 Like other civil law systems, Dutch law provides for the discharge of debtors in cases of impossibility of performance. Apart from this, unforeseen circumstances may create major difficulties in the performance of the contract, especially by interfering with the intended equilibrium of the corresponding obligations or by defeating the purpose the parties had in mind. The classic situation is that in which goods sold can be delivered and paid for only years afterwards because of a war which temporarily prevented performance, but at the same time caused irreversible inflation. In cases decided after World War I, the Supreme Court took the view that even such events cannot alter the parties' contractual obligations. In recent years, however, the Court has changed its position, anticipating article 6:258. This article provides that the court, at the demand of one of the parties, may modify the effects of a contract or set it aside in whole or in part on the basis of unforeseen circumstances which are of such a nature that contracting party, according to criteria of reasonableness and equity, may not expect that the contract be maintained in an unmodified form. It must be noted that the expression "unforeseen circumstances" should not be taken in its literal sense. The test does not relate to what parties have foreseen or could foresee, but asks whether the contract makes sufficient provision for the supervening event.

Articles 6:248 and 258 apply directly only to contract. Article 6:2 applies to all obligations. The Code lacks a provision of such a general character as article 2 of the Swiss Civil Code according to which (among others) the provision on "good faith" is applicable to all 557 "civil law relationships." However, it would seem that in practice under the new Code the situation will be identical.

On the one hand this is due to statutory provisions which extend, either directly or per analogiam, the effects of the provisions discussed above to other branches of the law. Examples of these provisions are article 6:216 stating that the provisions an contracts apply mutatis mutandis to other multilateral patrimonial juridical acts (for example the creation of rights in rem); article 3:166 paragraph 3 (in the title on common property) stating that article 6:2 applies mutatis mutandis to the juridical relations between the partners; and article 2:8 with a comparable rule for the relations within legal persons (corporations, associations, etc.).

On the other hand, the Supreme Court under the old Code has gradually widened the scope of the old provision on good faith (see nr. 6 supra) and it may be safely assumed that, when appropriate, the Court will continue to handle in the same way the relevant provisions of the new Code.

To mention one interesting example: the Supreme Court has decided that parties when negotiating a contract, must act according to criteria of reasonableness and equity, which impose on each of hem the duty to take into account the others' reasonable interests. According to recent decisions, this general duty may imply that a party lacks the freedom to break off negotiations (in which case the court may grant an injunction to continue or to resume them), or that, when negotiations have been broken off, he is under an obligation to pay damages. These damages may even amount to the expectation loss, including the loss of profits, which the plaintiff would have made from the envisaged contract had it been concluded.8

The Interpretation of the concept of "unlawful act" referred to above (supra, nr. 1) has been laid down in the new Code (article 6:162 paragraph 2). According to that paragraph the following acts are unlawful: the violation of a right and an act or omission violating a statutory duty or "a rule of unwritten law pertaining to proper social conduct." This is by no means the only open-ended concept in the law of tort. It is up to the courts to decide whether an act under that paragraph really constitutes an unlawful act or whether instead there is "a ground of justification" for that act. Some of these grounds are set forth in statutes, but they may also flow from un- 558 written law. And under the old Code an unlawful act only gave rise to an obligation to pay damages if there was "fault" on the part of its author, that is to say if the act could be imputed to him. The new Code (article 6:162 paragraph 3) adds that an unlawful act may even in the absence of fault be imputed to its author if he is "answerable according to common opinion."9

Anticipating that new rule, the Supreme Court has held the government and other public authorities liable for acts of legislation and for acts granting (or refusing to grant) public permissions and licenses which afterwards turned out to be annullable in administrative proceedings, irrespective of any fault on the part of the public authority. The same applies to cases in which a party to civil proceedings has obtained a judgment and has it executed, whereas later on it is annulled by a superior court. Dutch legislation does not contain specific provisions to deal with these matters; the courts have filled the gap through the application of the provisions on tort law. In principle, the same could happen - leaving aside the possible interference of EEC legislation - with modern liabilities, like product liability and liability for environmental pollution. So in effect, leaving aside specific legislative measures, the Code leaves it to the courts to decide when liability based on risk is to be accepted -a result which perhaps one would not expect to find in a system of codified law.10

I must immediately add, however, that the foregoing is only true insofar as the proper acts of the potentially liable person are concerned. As to the liability for other persons and for things, articles 6:169ff lay down a regulation of liability for persons (children, employees, independent contractors, agents) and things (defective movable things, defective constructions, dangerous substances, animals). This set of rules goes into quite some details; it does leave some leeway to judge-made law, but not to the extent as mentioned above.11 This, too, in a way follows the tradition under the old Code, in which the courts - unlike the French and Belgian judiciaries - have not felt free to create, in the absence of fault, liability for things and persons outside the fairly narrow limits imposed by the Code.12 Compared to the number of instances set out in the old 559 Code, these limits have expanded considerably, but the principle has remained the same in the sense that the legislator seems to consider it to be his prerogative to determine the limits of these cases of liability.

Since the old Code was nearly silent on the subject, the law of damages has been shaped by the courts, both in the fields of breach of contract and of tort. The new Code devotes a chapter to this subject (article 6:95-110) which for the greater part codifies the previous case law - including all the open-ended concepts peculiar to it. I will just mention a few examples.13 The chapter applies to both types of damages just mentioned as well as to other legal obligations to pay damages.

Although in principle the plaintiff has a right to claim compensation for the exact damages he suffered, the courts are free to assess the damage in a more abstract way, if that corresponds better to its nature (article 6:97). So if a car has been damaged, the amount to be paid is the amount which it would normally take to have it repaired; and that amount has to be paid irrespective of whether the owner has it repaired or not.

The obligation to pay compensation naturally presupposes that the non-performance or the unlawful act is the causa (cause) of the damage suffered by the plaintiff. On this point in Dutch law an interesting development has taken place. Between 1927 and 1970 the Supreme Court adhered to the theory of adequate causation; the court construed articles 1283 and 1284 of old Code (art. 1150 and 1151 of the French Code Civil) to require that the debtor could reasonably foresee the damage (including its extent) at the time of the conclusion of the contract - or in the case of a willful breach, at the time of that breach. The same test used to be applied in tort, in which case, of course, the relevant time was when the tort was committed. This approach was followed by the Draft Code of 1961.

In the 1960s a new current in legal doctrine criticized the test of foreseeability, contending that the degree of probability varied according to the nature of the liability and the nature of the damage, so that one should concentrate on those factors that make it reasonable (or not reasonable) to attribute the damage to the event for which the defendant is responsible, as its cause. Subsequently, the new Code adopted the new doctrine in article 6:98.

In practice the new doctrine has led to an extension of the obli- 560 gation to pay damages in the field of physical injury due to traffic or other accidents; in this field the concept of foreseeability hardly plays a role anymore. In other areas, however, notably the duty to pay compensation for pecuniary losses which are not caused by material or physical damage, the concept of foreseeability still is often used by the courts.

Article 6:101 adopts the concept of "comparative negligence" for the whole field of the law of damages. According to this provision the apportionment between plaintiff and defendant must be made "in Proportion to the degree in which the circumstances which can be imputed to each of them have contributed to the damage," that is to say in proportion to the causal relation (regardless of fault) between those circumstances and the damage. But it may be otherwise "if equity so requires due to the different degree of gravity of the faults committed or to any other circumstances of the case." Under the second brunch of the rule the Supreme Court has decided for instance to disregard the fault of a young child (below the age of fourteen) who is injured in an accident for which another person is liable.

Damages are normally paid in money. The plaintiff may also demand reparation in other form, but it is at the discretion of the court to grant it (article 6:103). Moreover, if the defendant has profited as a result of his non-Performance, the court may calculate the damage so as to include all or part of that profit (article 6:104).

Article 6:109 contains a rule which in its wide scope is new to Dutch law. It grants courts the power to reduce a legal obligation to repair damage on the ground that awarding full reparation "would lead to clearly unacceptable results." Among the factors to be taken into consideration are the nature of the liability (such as risk versus fault), the juridical relationship between the parties and their respective financial capacity. There is no power to reduce damages if the debtor's liability is or (by statutory duty) should have been covered by insurance.

Some legal scholars have harshly criticized this rule because they considered it to be a rule of social security law rather than a rule of private law. However, it was retained by parliament as a necessary complement of some new rules on liability without risk, especially the rule of article 6:165, which lays down that a person with a mental or a physical handicap is not immune from liability in tort. The result is comparable to § 829 of the German BCB, which places upon such a person a liability according to what is just and equitable. But it remains to be seen how the provision will be implemented in other cases of tortious liability and of liability in contract.

The law of damages is not composed entirely of blanket con- 561 cepts. There is a rather strict rule on the liability for non-pecuniary or moral damages (article 6:106). And in the case of physical injury or death, there is an exhaustive enumeration of persons who are entitled to claim compensation for losses which they have suffered as a result of the injury (article 6:107 and 108).

Apart from the broad good faith clauses discussed above, the new contract law is full of other general clauses, albeit of a less spectacular character. Without going into too many details I will offer some examples.

There is, of course, the traditional rule that contracts (in fact all juridical acts) contrary to good morals or public order are null (article 3:40 paragraph 1). This rule is not often applied anymore; in a country with few remaining taboos, it is hard to say what the concept of "good morals" implies. A more serious consideration is that many administrative acts directly lay down what is forbidden and what not; violation of such a rule entails nullity of the act, unless the court finds that another sanction (e.g., annulability) is a better sanction (article 40 paragraphs 2 and 3). In modern times other and more flexible concepts for the control of contracts and for the protection of a socially or economically weaker party has come into being (such as derogative function of good faith, the rules on general conditions, abuse of circumstances and economic duress).

Under the principle of consensualism, prevailing in the Netherlands, contracts are concluded by mutual consent (consensus) of the parties without being subject to any requirement as to form. As to the matter of consent, in legal theory there is a dispute as to the question which exactly is the basis of contractual liability: according to the will theory it is a party's will (concurring with that of the other party) which binds him; according to the theory of reliance it is the fact that the one party makes the other rely on his promise. The new Civil Code does not solve the problem in a theoretical way, bat contains provisions which enable legal practice to adopt a solution suitable to the circumstances of the case. On the one hand the Code provides that a juridical act requires an Intention to produce legal effect, which has manifested itself by a declaration (article 3:33). On the other, it protects a party's reliance on declarations or conduct of the other party, even if they do not conform to that other party's actual intentions (article 3:35). So the practical results are 562 close to the theory of reliance. A party may be bound beyond his will, if the other party reasonably misperceived his will. In such a case there is a valid contract, in contrast to the German solution which admits the annulment of the contract subject to the right of the relying party to seek compensation for its reliance interest.

The preceding remarks show that Dutch law favors a smooth and informal formation of contract. This tendency is confirmed by the absence of other requirements, well known to many legal systems, such as consideration and cause.

However, as to the doctrine of consideration, a trace of it may be found in the law of donation, which prescribes writing (even a notarial deed) for an enforceable promise. However, this provision has met with considerable resistance in legal practice and is soon to be abolished by the relevant chapter of Book 7 of the new Civil Code.

Moreover, a similar idea seems to play a part in the law on formation of contract in the case, set out above, that a party is bound beyond his intent. Under the old Code, which, apart from requiring "consent," was silent on the subject of formation, the courts were inclined to protect reliance only if the party relying on a promise (which did not correspond to the actual intent of the promisor), either had made a promise himself to the promisor or had detrimentally relied on the promise. Recently the Supreme Court changed its position under the influence of the new Code, which does not adopt this nuance: article 3:35 does not distinguish between gratuitous promises and promises made for value. However, the provision requires justifiable reliance which is justifiable in the circumstances, and this poses a duty of investigation on the promisee which may well be more extensive in the case of a gratuitous promise than in the case of a normal contract bestowing benefit an both parties.

And apart from this, even if according to article 3:35 a valid promise has come into existence, the promisee who has not given any benefit nor suffered detrimental reliance, may, through the operation of the provisions on good faith, be precluded from asserting his rights against the promisor. All this not only applies to promises, but also to other unilateral juridical acts which do not correspond to the real intent of the party making the declaration, for example, an employee resigning his job in a state of mental confusion. The Supreme Court in recent cases has adopted this live of reasoning, which concurs perfectly with the new Code so that it may be safely assumed that it will be maintained after its having entered into force an January 1, 1992. This again demonstrates the overwhelming importance of the provisions on good faith, which even play their role within the framework of such otherwise general clauses as those on the formation of contracts.

The reliance principle also plays a role outside the field of for- 563 mation of contract. First, as was mentioned above, contracts must be interpreted according to good faith. In this respect, the courts have adopted the same criterion as the one laid down in article 3:35: contract clauses should not be interpreted according to their literal sense, but according to that which the parties could reasonably attribute to it in the circumstances.

Second, in the law of agency the lack of sufficient authority may not be invoked by a person who himself gave the third party reasonable grounds to believe that the agent had such authority (article 3:61 paragraph 2). The same principle is expressed in a very general way in article 3:36: a third party is entitled to rely on the existence (or non-existence) of a legal relationship between two other persons, if by declarations or by other conduct they have given him reasonable grounds to believe that such a relationship did (or did not) exist. This provision covers the whole field of private law.14

Unlike the French Civil Code (followed by the old Code), the new Civil Code no longer requires a "cause" for a contract to be valid. This requirement has given rise to great uncertainty, but has only in a few cases proved to be useful. To one of these, the Code devotes a special rule: article 6:229 provides that a contract in furtherance of an already existing legal relationship between the parties can be annulled, if this relationship does not exist. In other cases, the solution must be drawn from general. provisions, like articles 3:33ff (intention and reliance of the parties in creating their relationship) and 6:74ff (rules on non-performance, for instance, if the thing sold does not exist or already belongs to the buyer).

The new Code contains several provisions on defective contracts (and other juridical acts). As in other countries, juridical acts may be defective in different ways. A fundamental distinction must be made between, (ipso iure) nullity and the possibility of annulment (avoidance) on the other. Ipso iure nullity - as in the case of illegality - operates automatically, just as the expression suggests. An act which may be annulled is valid until it is annulled by a subsequent act, which may be a judgment or - as is the normal case in the new Code - a declaration by the interested party to the other party. Performance rendered in the execution of the contract which is null or has been annulled must be undone; both parties may claim restitu- 564 tion of that performance or, if it can not be undone, they may be entitled to claim its value (article 6:203ff).

Under the old Code, a problem in this respect was created by the fact that there were no intermediary solutions, even in cases in which the consequences of nullity or annulment were unsatisfactory. Under the new Code the courts have been granted considerable freedom to deal with those consequences appropriately. For example, in some instances the contract may be replaced by a substitute (article 3:41). Moreover, according to article 3:58 a null juridical act may become valid ex post facto if a legal condition for its validity is fulfilled after the execution of the act. Under article 3:53 paragraph 2 the court may refuse to give effect to an annulment in whole or in part, if the act has already produced effects which can only be undone with difficulty and it may order that a party who is prejudiced by this decision be compensated by a party who unjustly benefits from it. In some cases of annulability (misrepresentation and abuse of circumstances, art. 3:54 and 6:228ff) the court may even, instead of pronouncing the annulment, modify the effects of the contract in order to remove the prejudice suffered by the abused party. These provisions may, if appropriate, be applied per analogiam to cases of nullity. Finally it must be borne in mind that there also exists the possibility of modifying the effect of the statutory rules through the operation of the general clause an good faith.15

Contracts may in various ways affect the position of third parties, for example, in the sense that a third party who induces a debtor to violate his obligations as against the creditor or who consciously derives profit from such a violation, may be liable in tort towards the creditor. The general rules on tort are broad enough to encompass cases of this kind. But as a rule, third parties will not be bound by contracts concluded inter alios. However, this principle has important exceptions in Dutch law as elsewhere. Suffice it to mention the case of a collective labor contract; the case in which the property in a leased object passes from the lessor to a new owner, who will take the place of the lessor as contracting party and for the future assumes his rights and duties; and the stipulation for the benefit of a third party.

Apart from these cases, which have a long history and are regulated in the Civil Code or in separate statutes, an interesting new development is taking place in Dutch case law, supported but not fully 565 covered by some provisions in the new Code. I refer to the effects of exoneration clauses as against a person who is not a party to the contract. Suppose A hands over a thing to B for reparation and the reparation is carried out by B's employee who causes damage to the thing. According to Dutch law the fact that A may claim compensation from B in itself does not prevent him from bringing an action in tort based on negligence against C. However, if by means of a "Himalaya clause" B has exempted himself from liability (which may be either contractual liability on account of non-performance of his obligation, article 6:77, or vicarious liability in tort, article 6:170) the new Code allows the employee, too, to invoke the exemption clause (article 6:257). Rules of the same type exist in the law of deposit (article 7:608) and in the law of transport (article 8:361ff), in which they have a much wider scope.

On the other hand, it is possible that a party who has exempted himself from liability may invoke the clause not only against the other party to the contract, but also against a third person (e.g., the owner of the thing). Such rules can be found in the articles of Books 7 and 8 just mentioned, but not in Book 6, the law of obligations in general. However, the Supreme Court, under the old Code, has applied that principle in a case where the owner A who handed over goods to B, granting B full discretion to conclude a contract for work with a third party C. When the goods were damaged by C, A brought an action based on tort against C; but C was allowed to invoke the exception clause he had agreed with B. For the scope of this paper it is interesting to note that the decision was not based on the doctrine of representation,16 so that the court created an altogether new doctrine of unwritten law. Under the new Code, the courts have deliberately been granted the freedom to develop the law in this area, since the old article (corresponding with article 1165 of the French Civil Code), providing that contracts only have effect as between the parties, has been deleted.

In Dutch law it is generally understood that any source of obligation (as opposed to the more general concept of "duty," for instance in the law of torts (see nr.11)) must find its basis in statute law. In the new Code this concept is laid down in article 6:1: obligations can only arise if they result from the law. However, already under the old Code, which contained a similar provision, this was liberally construed by the Supreme Court, which took a position 566 similar as that of article 1173 of the Italian Civil Code: in situations not expressly regulated by a statutory provision, the court should adopt the solution which fits into the statutory legal system and is in live with rules already laid down for similar situations. It is expected that the courts will take a similar position under the new provision.

But it is not certain whether and to what extent they will feel induced to use that freedom in the near future, since the new Code has filled some gaps which were felt to be cumbersome under the old law. The most important of those is the general action on unjust enrichment (article 6:212), which the courts have not felt free to allow as such in the past.

The old Code was silent on the subject of abuse of rights. Case law nevertheless developed the doctrine, which is now set forth in article 3:13 of the new Code. The provision begins by stating that the holder of a right may not exercise it to the extent that it is abused. And it goes on by presenting the most important examples of such abuse, for instance the exercise of a right with the sole intention of harming another or for a purpose other than that for which it was granted; or the exercise of a right even though its holder could not reasonably have decided to exercise it, given the disproportion between the interest in exercising the right and the harm caused thereby. The three cases mentioned in the provisions concur with the existing case law; since in the new provision they are listed as examples only, the courts are free to further develop the concept of abuse of right.

As to its scope, I may add that the concept covers the whole field of private law. At first glance (judging from its position in Book 3), one would think that it is meant to deal with abuse of right in the context of patrimonial law only, but article 3:15 states that the article also applies to areas other than patrimonial law. In fact, the concept has already been applied by the courts in the law of civil procedure and in family law. On the other hand, within the realm of patrimonial law the doctrine of abuse of right - unlike the situation in French and Belgaen law - does not play any part in the law of obligations and in the law of contracts, because there its function is completely covered by the provisions on good faith.

In the Netherlands there exist two separate jurisdictions court systems: the ordinary (which is competent for cases of private law and criminal law) and the administrative. The latter deals only 567 (roughly speaking) with the legality and validity of unilateral juridical acts of a public law nature (such as public permissions). Where government or lower public bodies and agencies take Part in legal intercourse by concluding contracts, by committing torts, or by exercising property rights, possible conflicts resulting from those acts are brought before the ordinary courts. For this reason the ordinary courts for a long time have been familiar with the assessment of the nature and effect of public interest in private law cases. I already mentioned in this respect the examples of the liability in tort of public agencies for the damage caused by the fast that public permissions afterwards are annulled in administrative proceedings.

The new Code does not alter this dual nature of our judiciary system17 nor does it contain general provisions for the private law relationships between citizens or companies on the one band and public authorities on the other. But the open-ended concepts discussed in this paper enable the courts to take into consideration the specific aspects of those relationships and especially the public interest which the administration intends to serve with them. In some cases this will mean that public authorities are judged in a more severe way than a private party would have been. In other cases, a private party is obliged to tolerate an infringement of his rights from such authorities if he would have to suffer if it emanated from a private counterpart. An example of this in the law of tort is offered by article 6:168, stating that the court may reject an action to obtain an order prohibiting unlawful conduct on the ground that such conduct must be tolerated for important societal interests; this does not preclude, however, the victim's right to compensation. This doctrine has been applied in a case in which a local authority withdrew water from the ground of a private owner in the interest of the public drinking-water supply.

The public interest may also play a role in other instances, even between private parties. This is true in the law of community (Joint ownership), where the courts, if there is a dispute as to the use of common property or its partition, are called upon to take into account both the interest of the parties and the general interest (articles 3:168 and 185). The general interest may be a factor to be taken into consideration when deciding upon the modification or even rescission of a contract or of an act constituting a servitude (articles 6:259 and 5:78). And even in applying the concept of good faith in general, on which this paper has already touched several times, the courts are reminded that they should take into account, among 568 other things, "the particular societal and private interests involved" (article 3:12).

A last provision which is worthy of note in this connection is articles 3:14: A right which a person has pursuant to private law may not be exercised contrary to the written or unwritten rules of public law. Among other things, this provision enables the courts -.again in conformity with existing case law - to measure the private law acts of public authorities by the same standards employed by the administrative courts. Those standards, although phrased in public law terminology, conform to a large extent to the test of good faith in private law. In conclusion, through the operation of the open-ended concepts of unwritten law, the two bodies (public and private) of law have much more in common than the system of separate court systems suggests.

The recodification process in the Netherlands, now coming to an end, has led to a product which is remarkable in several respects. The first goal of codification18 has been reached to a considerable extent. Nearly the entire private law - civil law in the strict sense of the word, commercial law, labor law, consumer law, specific legislation of a private law character - has been unified in a single Body of codified law. Its clean systematic structure and its uniform terminology seem to guarantee considerable certainty, clarity and predictability of the law.

On closer examination, however, this certainty and predictability may be called into question. As the previous paragraphs have shown, the Dutch legislature has granted a large amount of discretion to the courts. As we have seen, this discretion has two aspects. The courts are free to further develop the law where its provisions are silent (for example, the general clause on tort and the provisions an good faith in its suppletive function). On the other, the courts have the possibility of derogating from specific provisions of the law or of a contract to avoid an unjust result in the specific circumstances of the case (e.g., the provisions an good faith in its restrictive function and the rule on abuse of right).

In addition to the examples given above (a survey which is far from being an exhaustive enumeration of open-ended concepts in the new Civil Code), I should add that the Code refrains from laying down a specific hierarchy among the various sources of law (statute, custom, equity). Initially, following the old Code, there were draft provisions to that effect. But these provisions were deleted at a later 569 stage and the issue was left to the discretion of the courts. This met with no resistance among legal scholars, since the Supreme Court in a decision of 1972 had ruled (in spite of a provision to the contrary in the old law) that in exceptional circumstances a statutory provision of a mandatory character could in fact fall into disuse and be neglected.

There are not many legal scholars in codified law systems who would be opposed to open-ended concepts as a matter of principle. Indeed, the advantages of flexible rules are obvious. Only by means of such a flexibility can Codes survive the many years they are destined to exist. Only this flexibility allows the court to do justice to the circumstances of each particular case, which often do not fit into the abstract provisions of the Code (nor, for that matter, into the rigidity of judicial precedent). "In the long run, the real test of a Code is a measure of flexibility which it has allowed the law.19

Moreover, open-ended concepts have the important function of bringing internal cohesion and unity to a system of law seen as a whole:20 through them legal Innovation in one area of the law may be exported to other areas, in order that unity is brought about wherever possible and differences may be allowed for wherever necessary. In the past, the general clauses in the civil law have allowed it to "commercialize" - that is, to accept influences from the Commercial Code-branches of the law set out in the Civil Code, but not substantially different from commercial law. They have in the same vein enhanced the social aspect of civil and commercial law where that was felt to be necessary but the pertinent legislation (rules protecting workers, tenants, consumers, commercial agents) had not yet come into existence. The analogy clauses make it possible to transplant to a certain extent the suppleness of the law of contracts to the law of property and to bring private law and public law more into line.

Few lawyers would question these advantages of flexible rules. But, of course, the important issue is to find the correct relation between certainty and flexibility in the law. Although some Dutch scholars have questioned the wisdom of the legislature in this respect, I think that the new Code fully reflects the present-day Situation of Dutch law and the attitude of the courts; and there have been more authors fearful of a second21 wave of legalistic interpretation than authors critical of the general clauses.

The new Netherlands Civil Code certainly has gone much fur- 570 ther than its French model;22 and also further than its German and Italian counterparts, since although it is true that in those systems good faith plays a role, these Codes do not contain as many general clauses as does the new Dutch Civil Code. On the other hand, in non-codified systems or in systems with very old codifications, the leeway for the courts is much greater than in the Dutch system. It would seem that in this respect, as in several others, Dutch law has steered a middle course.23

Perhaps the Dutch approach to codification could be compared to the Swiss.24 It is true that the system of the Dutch Code is different and also that the Dutch Code is more comprehensive; moreover, it employs a more technical language than the Swiss Code. But the relation between codified and judge-made law, and notably the overt acceptance of the courts' task in interpreting and further developing the law, in both systems are comparable. The well known and wise provisions on that subject to be found in the Introductory Title of the Swiss Civil Code correspond to a remarkable degree to the actual situation in the Netherlands. That Situation has now been im- 571 plicitly (but explicitly as far as the Intention of the legislator is concerned25) codified. As one scholar aptly put it: judicial discretion and creativity are now part of our codified law.

Personally, I do not belong to the group of lawyers who nourish misgivings about the outcome of the legislative process. First, I think that the Code corresponds perfectly to the actual situation in the Netherlands. Had the legislature taken a more conservative position as to the strictness of statutory provisions and, correspondingly, to the freedom of the courts, that certainly would have met with severe criticism from a majority of legal scholars.

Second, I agree with Professor Sauveplanne: "In the last resort the certainty of the law does not depend on the provisions of a Code or the strictness of a rule of precedent, but on the wisdom of the judges."26 In the Netherlands the course followed by the Supreme Court during the last decades has been widely acclaimed, even to the extent that after Meijers' death Supreme Court Justices have been among the Codes most influential drafters. That course has left its mark on the codification developments. As yet, I see no sign that things will change after the enactment of the new Civil Code.

*ARTHUR S. HARTKAMP is Advocate-General at the Supreme Court of the Netherlands; Professor of Private Law at Utrecht University.
1ln fact more than one, viz a Civil Code, a Commercial Code and a Code of Civil Procedure.
2Quotation from the English translation of the New Netherlands Civil Code, Patrimonial Law 298, article 6:162 (P.P.C. Haanappel trans., 1990). This translation uses English civilian terminology as opposed to the translation of Book 6 The Law of obligations, Text and Commentary (1977), which adopted the common law terminology. The 1977 translation must be considered as outdated because of the many changes which were made in the original Draft of Book 6. The 1990 translation not only covers Book 6 (General Part of the Law of obligations), but also Book 3 (Patrimonial Law in General), Book 5 (Real Rights) and parts of Book 7 (Special Contracts).
3I refer the interested reader to some articles I published earlier an those subjects: "Civil Code Revision in the Netherlands: A Survey of its System and Contents, and its Influence an Dutch Legal Practice," La. L. Rev. 1059-90 (1975); "Vers un nouveau Code Civil Neérlandais," Revue Internationale de Droit Comparé 319-36 (1982); "Civil Code Revision in the Netherlands 1947-1992," in New Netherlands Civil Code. Supra n. 2, at xiii-xxvii (with a select bibliography of literature in foreign languages on the new Dutch Civil Code at page xliii). See also Archiv für die Civilistische Praxis 191 (1991); Ewoud Hondïus, Das neue Niederländische Zivilgesetzbuch 378-395; A.S. Hartkamp, Einführung in das neue Niederländische Schuldrecht, Teil I 396-410 and J.B.M. Vranken, Einführung in das neue Niederländische Schuldrecht, Teil II 411-432. These articles have also been published (in a slightly modified version, and together with other contributions to a conference held in Salzburg, 1990) in Franz Bydlinski, Theo Mayer-Maly & Johannes W. Pichler (Hrsg.), Renaissance der Idee der Kodifikation (1991). In this paper I will refer nearly exclusively to publications on the new Code which have appeared in languages other than Dutch; and I will occasionally use some passages from my earlier publications without explicitly referring to them.
4See for more details the literature mentioned in note 3.
5Articles of the new Civil Code are cited by the number of the Book followed by the number of the article within that Book.
6American lawyers are well aware of the fact that the term "bona fides" or "good faith" is used in two different senses. See Farnsworth, Contracts 550-51 (1990); O'Connor, Good Faith in English Law 20ff (1990). In contract law, acting in good faith refers to the observance of reasonable commercial standards of fair dealing, or as the Dutch legislature has put it, acting in accordance with reasonableness and equity. This is a purely objective test: if a party acts in an unreasonable and inequitable way, it will not be a defense to say that he honestly believed his conduct to be reasonable and equitable. In the other sense, good faith refers to a test that originally was purely subjective, indicating a state of mind (lack of notice) such as a requirement for the acquisition of movable property where the transferor is not the owner of the thing. Later on, this test was developed in such a way as to include also an important objective element. In the new Dutch Code, this results from article 3:11, where it is stated that good faith in this sense not only requires that the party concerned did not know the relevant state of affairs, but also that he should not have known it; this implies that he may be under a duty to investigate. In order to prevent the confusion resulting from the two concepts of good faith, the Dutch legislature has decided to use the term "good faith" only in the second sense, and to describe the term in its first sense by the concept of "reasonableness and equity." In this paper, however, I will use good faith (or bona fides) in this first sense only, which is shorter, more elegant and more familiar to foreign scholars (the other concept will not be discussed).
7This section and the next one, as well as Segments of this paper are taken from An Introduction to Dutch Law for Foreign Lawyers, Ch. 9 (D.C. Fokkema, J.M.J. Chorus, E.H. Hondius, & E.Ch. Lisser, eds., 1992). In the first edition of the book (1978) this chapter was written by D.C. Fokkema; for the second edition, it was revised and adapted to the new Code by the present author.
8See J.M. van Dunné, "The Prelude to Contract, The Threshold of Tort, The Law an Precontractual Dealings in the Netherlands," in Netherlands Reports to the Thirteenth International Congress of Comparative Law 71-85 (E.H. Hondius & G.J.W. Steenhoff, eds., 1990).
9The same criterion applies to liability for breach of contract (art. 6:75). In this and other respects the borderline between the law of tort and the law of contract has become less sharp; see for the law of damages infra, n. 13.
10See W.Th. Braams and E.H. Hondius, Wege zu einem neuen, Europäischen Haftungsrecht - der Beitrag der Niederlands (1990); Vranken, supra n. 3.
11See generally, Arthur Hartkamp, "Responsabilité délictuelle sans faute," in Actes du Congrès Codification: valerus et language 625-46 (1990).
12According to Tunc, "Codification: The French Experience," 69, paragraph 63ff (Stoljar ed. 1977) Problems of Codification, the French judge-made law in this area is obscure and it would have been better if the legislature had enacted a law for compensation in traffic accidents (which was passed eventually in 1985).
13See for an overview, Bloembergen, "Schadenersatz im neuen niederländischen Zivilgesetzbuch," Bersicherungsrecht paragraph 4ff. (1983); Vranken, supra n. 3.
14See article 3:24ff., 86 and 88. In the law of property several more detailed provisions have been included, protecting third parties who have obtained property from a non-owner, which can not be dealt with here.
15Article 6:211 in the chapter an undue payment offers an example, but should by no means be considered as exhaustive.
16Which in Dutch law, as in other civil law systems, requires that the agent disclose his capacity to the third party.
17But at the moment legislation is pending to merge the judiciaries. The short remarks of Koopmans in the International Encyclopedia of Comparative Law, Vol. 1, p. N-13, on the administrative judiciary are outdated.
18See Bydlinski, "Civil Law Codification and Special Legislation," in Questions of Civil Law Codification 25ff (1990).
19Paton, cited by Tunc, A Textbock of Jurisprudence 255 (1972).
20See Hedemann, Die Flucht in die Generalklauseln 59 (1933).
21Like the one the Netherlands witnessed in the second half of the 19th century following the codification of 1838.
22See Talfon, "La codification en matière de droit du contrat," in Questions of Civil Law Codification 168ff (1990). Of course, this does not mean that the development of French law has been frozen by the codification. On the contrary, as both Tunc, cited above, and Talfon show French courts have been very hold in adjusting the law to contemporary thought. My point, however, is the extent to which the Code itself explicitly allows the courts to do so.
23The new Dutch Civil Code, notably the law of obligations, has been thoroughly influenced by comparative law research and by uniform law. Although the study of the extent of this influence has yet to begin, there is a wide-spread belief that Dutch law has floated away from its French origin into the direction of the German law family. Personally, I think that this assumption is correct in respect of some systematic modern features (although a general part in the German sense has been rejected) and of a number of rather technical provisions; moreover, there is German influence on some important subjects, such as the provisions on good faith and on general conditions. At the same time, however, there are strong influences, both ancient and modern, from French and Belgian law (for example, misrepresentation, undue influence, anticipatory breach) and from uniform law (the 1964 and 1980 Conventions on the International Sale of Goods have exerted their influence not only in the law of sales, but also in the general provisions on formation of contract and on non-performance). In the law of property there are notable differences between German and Dutch law in the law of transfer of things and in the law of security of movables. More important perhaps is that Dutch law is distinctly less "dogmatic" or "conceptual" than German law. For a brief survey of foreign influences an the new Code, see A.S. Hartkamp, "International Unification and National Codification (and Recodification) of Civil Law: The Dutch Experience," in Questions of Civil Law Codification 67ff (1990). See also F.J.A. van der Velden and N.A. Florijn, "Harmonization of Private Law Rules Between Civil and Common Law Jurisdictions," in Netherlands Reports to the Thirteenth International Congress of Comparative Law 43-57 (E.H. Hondius & G.J.W. Steenhoff eds., 1990).
24See Von Overbeck, "The Role of the Judge under the Swiss Civil Code," in Stoljar o.c., p. 135ff; Kötz & Zweigert, Einführung in die Rechtsvergleichung I (1984), p. 196ff. I agree with the interesting remark of Kötz & Zweigert at page 207 that the Swiss model of codification should become the orientation for a future European Civil Code.
25See the quotations in Hartkamp, Wetsuitleg and rechtstoepassiny na de invoering van het nieuwe Burgerlijk Wetboek 39-49 (1991). I will mention just a few examples. The Minister of Justice reminds us of the fact "that, under the new Code, as well as under the existing law, the court, when called upon to decide a particular case, will have to take Info consideration both the applicable provisions of the Code (if any) and unwritten law by means of which the scope and the effect of those provisions often must be further determined." And: "The handling of the written law (its interpretation and application) must be entirely left to the courts, guided by legal doctrine. It is not desirable (and the legislator does not wish) to steer developments in this area by means of general rules." Finally: "It is not the task of the court to modify or to abolish a written provision of the law; it is his task, however, to decide how it is to be applied to the case at hand and to determine the relation between such a provision and unwritten law which is of relevance (custom, reasonableness and equity, rules of unwritten law pertaining to proper social conduct, abuse of right, etc.)." These quotations refer to the relation between the provisions of the Code and unwritten law; it is understood that in the case of "lacunae" the courts are free to further develop the law on the basis of analogy and of unwritten law, taking Into consideration the opinions of legal authors and developments in other legal systems as the court may see fit.
26J.G. Sauveplanne, "Codified and Judge Made Law, The Role of courts and Legislators in Civil and Common Law Systems," in Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letterkunde, Nieuwe Reeks, p. 95ff., p. 119 (1982).