Search Menu


Issues list


Law as a Means of Normative Communication

ISBN 9985-870-21-2

Cover image



On Legislative Style and Structure

Should a legislator write in legal language that gives precision to his intentions but which the average citizen may have difficulties in understanding? Or should he instead use simple and popular language, which the citizen can understand but is less precise?

Issues of style and structure are closely related. In the structure of a statute or code a legislator may use legal criteria, as some legislators do when they separate the law of obligations from the law of property, and address the property aspects of the sale of goods in one book and the mutual obligations of the parties to a sales contract in another. That enables treatment of issues relating to property and to obligations together, but the citizen then has to find material on these two aspects of a sale in different places. It would be easier for him if all the rules relating to the sale of goods were located together.

Should the code or statute be exhaustive and cover all of the issues that one can imagine being touched upon by a subject, or should it cover only some important issues, leaving the others to be developed by the courts? Should the statute provide broad principles, letting the courts then fill them in, or detailed rules, which give the courts little freedom of discretion?

In the following discussion, I will address the ways in which English, German, and French legislators have dealt with these problems. The focus will be on the statutory provisions relating to obligations. In addition, I will discuss the style and structure of the UN Convention on Contracts for the International Sale of Goods (CISG); the Principles of European Contract Law *1   (PECL), which were prepared by the Commission on European Contract Law; and the UNIDROIT Principles of International Commercial Contracts *2   (UPICC), which were the fruit of UNIDROIT work.

1. England

From the British one learns to be cautious when drafting black-letter rules, and not to draft rules that are too broad and general. The members of the two groups, the Commission on European Contract Law and the UNIDROIT Working Group, realised that their English colleagues were walking dictionaries for their cases, who could often point to English decisions showing the adverse consequences of a proposed rule.

However, neither the Commission on European Contract Law nor the UNIDROIT Working Group adopted the elaborate style of the English legislature. In Great Britain, most statutes were originally enacted to counteract some mischief created by the case law, and the English courts did not like them. They regarded statutes “as an evil, a necessary evil no doubt, which disturbed the lovely harmony of the Common Law” *3 . So when they interpreted the statutes, they applied them only to the precise situations that they unquestionably covered, applying a narrow and pedantic construction of statutes.

This led the English legislators to go into great detail in order to force the courts to give effect to their intentions. Hein Kötz gives an example of this. He recalls the way in which article 6 of the European Directive on Liability for Defective Products is implemented in the British Consumer Protection Act 1987. One of the factors in determining whether a product is safe is ‘the presentation of the product’. In implementing this directive, the British act succeeds in expanding these five words into 45 without adding anything of substance. The language on the presentation of the product becomes “the manner in which and the purpose for which, the product has been marketed, its get-up, the use of any mark in relation to the product and any instructions for, or warning with respect to doing or refraining from doing anything with or in relation to the product”. *4  

One wonders whether it is still necessary for the British legislation to go into such detail. In the last few decades, the English courts have moved, in Lord Diplock’s words, from the “purely literal towards the purposive construction of statutory provisions”. *5  

It seems, however, that the traditional style has got into the bloodstream of British lawyers. We find traces of it with the Study Group on a European Civil Code, which is in the process of amending the PECL. Article 2:102 (1) of the PECL lays down the general rule that “each party must act in accordance with good faith and fair dealing”. Under British influence, this paragraph has been amended to read: “A person has a duty to act in accordance with good faith and fair dealing in negotiating or concluding a contract or other juridical act and in exercising a right or performing an obligation.” Here, 12 words have become 34 words. *6  

2. Germany

The Germans — most notably, the Pandectists of the 19th century — made great efforts to refine the legal language. They set the imprint of precision and consistency on the German Civil Code (BGB), which came into force in the year 1900. Its draftsmen strove to find the most clear and precise expressions, to use words in the same sense wherever they appeared in a text, to utilise one expression universally for the same phenomenon, to avoid unnecessary repetition, and to place the more general rules before the more special ones. These techniques were later adopted by many of the world’s legislators.

However, the structure of the BGB is not only very carefully thought through but also very complicated. The idea of placing the more general rules before the more specific was carried into action with great consistency. Book 1 covers the rules that apply to the obligations, property, family matters, and matters of succession that are addressed by Books 2, 3, 4, and 5, respectively. This book provides, among other things, rules on the elements of agreements, the authority of agents to bind their principals, and prescription.

The style and main structure of the Civil Code was retained when the law of obligations was revised in 2001. Today, Book 2 (on obligations) has eight parts. Part 1 deals mainly with the contents of the obligation. In this and in parts 4–7, all obligations are handled via the same provisions, whether they are contract, tort, or restitution obligations. This has made it necessary to lift them to a high degree of abstraction and disembodiment. In § 241 BGB, which begins Book 2 and applies to all obligations, it is provided in paragraph 1 that the debtor is obliged to tender the performance he owes. This, as some writers note, is obvious. *7  

Among the new provisions of the revised edition of the BGB, some deal with culpa in contrahendo, fault in contracting. This subject was not treated in the BGB; rather, rules had been developed by the courts. *8   In § 241 (2) it is now provided that the debtor must also pay regard to the rights, ‘legal goods’, and interests of the other party. The debtor has what the Germans call Schutzpflichten, duties of care, toward the other party. But a party who negotiates a contract is not yet a debtor, as there is not yet a contract. The BGB’s § 313 (1), which is in Part 3 (dealing with obligations arising out of a contract), takes care of that. It specifies that the duties of care provided for in § 241 (2) may also arise out of contract negotiations. Also, § 280 (1) BGB, which is in Part 1, provides that a party may claim damages for the loss he suffers when the other party violates duties arising out of a ‘debt relationship’. Thus it is that the rules on culpa in contrahendo are to be found in a conjunction of §§ 241, 313, and 280. A general duty of care is established in § 241, with § 313 (1) extending that duty to negotiations and § 280 imposing liability for violating the duty. This is logical and carefully considered. However, in order to understand it, one must be a well-informed German lawyer. And the rules state only that a liability exists; there are no indications as to the situations to which they apply. In order to know that, one would have to consult the case law.

This style differs from the one used in the PECL’s article 2:301 (2) *9 and in UPICC article 2.1.15 (2) on negotiations in bad faith. These provide that a party who has negotiated, or broken off negotiations, in bad faith is liable for the losses caused to the other party. And they both give examples illustrating what constitutes behaviour in bad faith. *10  

The BGB’s § 242 provides that the debtor must perform his duty in accordance with good faith and fair dealing, having regard for commercial practices. Originally this rule was not meant to play a very significant role. But, due to the turbulent history of Germany and to a growing need for social justice, § 242 now has paramount importance. It has operated as a sort of ‘super-provision’ — a king of the Code, as it were — that has modified other statutory provisions, and it has been applied to change the rigorous individualism of the original contract law of the BGB, so as to adapt the law to the changed social and moral attitudes of the society. *11   On the basis of § 242, the German courts have created a number of obligations (called institutions), which ensure the loyal behaviour of the parties. They have, for instance, set aside unfair contract terms, most notably in standard-form contracts. This section of the BGB has been so widely invoked both in private and in public law that its application extends much further than its language can carry.

Some of the obligations that the courts created under § 242 have now been laid down in the revised BGB — for instance, the rules on standard-form contracts and the above‑mentioned provisions on culpa in contrahendo — but most of the institutions established by the courts still remain under § 242. The new BGB retains § 242 but does not tell us where the king rules.

In § 276, we find laid down the main rule on liability, contractual and non‑contractual. *12 The debtor is liable only for his deliberate and negligent acts, unless the existence of a stricter or more lenient degree of liability is specified or to be inferred from the other subject matter of the obligation — in particular, the assumption of a guarantee or an acquisition risk.

Thus German law has retained the rule that the remedy of damages is fault-dependent. In this respect, it differs from the Common Law, CISG, PECL, and UPICC terms, which set forth strict liability as the main principle. Some prominent German academics have emphasised the ‘fairness’ and the ‘obvious ethical superiority’ of the rule *13 , while others are less enthusiastic. *14 However, there are important exceptions. A stricter liability may be determined or may follow from the contents of the obligation — in particular, through the assumption of a guarantee or an acquisition risk.

What is it to assume an acquisition risk? In order to know that, one must be familiar with the German case law on the subject. The seller of generic goods is the main example of a debtor who has assumed such a risk, but also other suppliers may carry it, and there are cases where a seller of generic goods has not assumed it. *15 Additionally, what only the best-informed lawyer knows is that the rule is also a gateway to a politically desirable tightening up of liability, bringing this aspect of the German law of obligations closer to the international level. *16  

It is with some regret that I saw the German legislators of 2001 stick to the style and structure of the original BGB of 1900. They claimed — without any remorse — that this was their proud tradition. *17  

3. France

France produced the Code Napoleon, which gained world renown and became the greatest legal export article ever seen in modern times. This code reflected the sprit of the Enlightenment and the newly gained freedom and equality; its pithy language was that of great writers. It had, and still has, a remarkable vitality.

Still, the Code Napoleon is old. Several of its articles are now obsolete, but they are still there, and to remedy that the French courts have given them a meaning different from the one they had when the code was promulgated in 1804. This happens to old codes and was also envisaged by the authors of the code. It was expressed in famous words on legislative drafting. The task of the legislator”, Jean-Étienne-Marie Portaliswrote, “is to lay down the general maxims of the law, to establish principles which are rich in implications, and not to go down into a detailed regulation of all the issues that may arise in every matter”. *18 However, even broad principles cannot resist the ravages of time.

In addition, several provisions of the code are incomplete, some terms are ambiguous, and some lack the necessary terminological exactitude.

For instance, in article 544 et seq., the word ‘propriété’refers to ownership of movable and immovable things, and in the title of Book 3, which deals with the acquisition of propriété, it means wealth of all kinds. The code mentions cause as a requirement for the formation and validity of a contract, but it does not explain what exactly cause is, and there are many conflicting explanations. *19 Some Frenchmen make fun of it. According to one of them, if somebody thinks that he understands what cause is, this is because it hasbeen badly explained to him. But the French courts still invoke cause in several contexts.

To cleanse the portion of the code that deals with obligations of its ambiguities and obscurities and to render it up to date, a group of some 30 French academics under the chairmanship of Professor Pierre Catala has prepared an Avant projet de réforme du droit des obligations et du droit de la préscription, a draft reform of the law of obligations and of the law of prescription (PrC). The draft was submitted to the Garde des Sceaux, the Minister of Justice, in September 2005. *20  

Many of the rules of the PrC are clear and easily understood, but several of them are so loaded with implications, so fécondes en conséquences, and held in such abstract terms that the lay citizen could not understand their true scope. Cause is still there, and the style and main structure of the existing code have been preserved.

The Civil Code of 1804 had no rules on formation of contracts. It was left to the courts to establish the rules. The Cour de Cassation considers several aspects of the formation to be questions of fact, which are left to the sovereign appreciation of the lower courts. *21 Since their decisions are incoherent, there is some uncertainty as to which rules apply. The courts seem to agree that the offer is revocable. However, in cases where the offeree had reason to believe that the offer had been made irrevocable, the offeror, though not bound by the contract, is held liable in relation to damages, if he nevertheless revokes his offer.

The PrC still makes it possible for the offeror to withdraw his offer until it is accepted. The PrC does not say this, but it may be inferred from article 1105 (4) *22 , where it is provided that if an offer addressed to a specific person contains an engagement to uphold it for a certain period, an untimely revocation does not prevent the contract from coming into existence. The same rule is to be found in PECL article 2:202 (3) (b), CISG article 16 (2) (a), and UPICC article 2.1 4 (2) (a).

One may ask whether this is the only exception to the rule on revocability.The PrC does not mention the other exceptions provided for in the PECL’s article 2:202 (3) *23 — namely that “a) the offer indicates that it is irrevocable and c) it was reasonable for the offeree to rely on the offer as being irrevocable, and the offeree has acted in reliance of the offer”; see also CISG article 16 (2) b and UPICC article 2.1 4 (2) b. It may not be necessary to provide the former exception; however, we do not know whether the courts may establish the latter rule on reliance or whether they can hold the offeror liable for damages if he withdraws the offer.

Another question is whether the PrC upholds the existing French case law under which the parties may be bound by the contract even if there is a minor disagreement between them as to the terms *24 or whether the PrC imposes an absolute mirror image rule, which does not tolerate any disagreement.

Article 1105-5 provides:

– An acceptance is a unilateral act by which its author expresses his will to be bound by the terms of the offer.

– An acceptance which is not in agreement with the terms of the offer has no effect, except that it constitutes a new offer [emphasis added].

In article 1109-1’s Chapter 2, section 1, on the agreement (consentement), it is provided that there is no agreement when there has not been a meeting of the minds concerning theessential partsof the contract. Can one conclude from these two provisions in different chapters that an acceptance is effective if the disagreement does not affect the essential parts of the contract, or does article 1109 address a situation different from the offer and acceptance rule provided in article 1105-5? If so, which situation? And if not, whose terms should prevail: those of the offeror or those of the offeree? *25  

The PrC has provided articles that leave lacunae and, at least for the uninitiated, some ambiguities *26 . We do not know, for instance, to which extent the PrC intends to uphold the existing case law, It is with some regret that I see the French academics stick to a style of drafting very similar to the one adopted in 1804. They, too, maintain that this is their proud tradition.

4. The PECL and UPICC

The citizens very often have difficulties in finding and understanding foreign laws. For them, and especially for those engaged in cross-border trade, uniform contract rules would be a great relief. This has been confirmed by the success of the UN Convention on Contracts for the International Sale of Goods, which in June 2006 had been adopted by 67 countries.

To give further help to the international trade community, the UNIDROIT Working Group on the Principles of International Commercial Contracts and the Commission on European Contract Law started to work on the UPICC and PECL, in around 1980. Most of the members of these two groups were academics, who arrived with the stylistic traditions of their own country in mind. However, I cannot remember a discussion as to principle on that matter, although the style and structure that were finally adopted in these two instruments differed from those of the statutes and codes of several countries — among them the English statutes, the German Civil Code, and the French Code civil.

The CISG served as model both for the Commission on European Contract Law and for the UNIDROIT group. As described by John Honnold *27 , the authors of the CISG wished to avoid abstract, disembodied concepts, and endeavoured to use plain language that refers to things and events for which there are words with common content in the various languages. Another object was to draft broad rules. Since both groups took many of the CISG’s provisions verbatim, the CISG’s style was adopted without much ado. However, also in those provisions that were ‘new’, plain language as in the CISG was used. Latin words were avoided. The authors were reluctant to use words and expressions with a special legal meaning in English. They went even further than the CISG when they used the expression ‘non-performance’ instead of ‘breach of contract’, since in English law there is breach only if the aggrieved party can claim damages, not when he can only terminate the contract. They needed a term for the failure to perform for which the aggrieved party has at least one remedy, be it damages, termination, reduction of price, or something else.

This style was in accordance with Jean-Étienne-MariePortalis’s famous words on legislative drafting. However, although both groups described their provisions as for principles of contract law, they did not go too far in laying down, through supplying a broad perspective, the general maxims of the law if this might have made the rules difficult to understand. With a few exceptions, they tried to concretise and visualise their intentions.

It seems as if Eugen Huber’srecipe for the Swiss Civil Code *28 guided the majority in the two groups. “The code must speak in popular terms”,he said, continuing thus:“The man of reason who has thought about his times and their needs should have the feeling as he reads it that the statute speaks to him from the heart.” As far as the material permits, he said, the law should “be comprehensible for everyone, at least to those who are involved in the activities regulated by the code. […] Its provisions must mean something to the educated laymen, even if they will mean more to the specialist”. Huberadvocated the use of short articles, short paragraphs, and simple and short sentences.

However, there were people in the groups who did not agree on these devices. There were those who wanted consistency and logic to govern the drafts even at the expense of plain and simple language, and who preferred precision and detail to a language rich in implications. It was not easy always to live up to the visions of Portalis and Huber. Some policies could not be expressed in pithy and pregnant language. The authors had to go into some detail when, for instance, in a rule on change of circumstances they wanted to give a contracting party the opportunity of having his contract renegotiated and of, when re-negotiation failed, having the contract modified or ended by the court. *29 They had to be rather explicit and detailed when they made rules on appropriation of performance in cases where a party has to perform several obligations of the same nature and the performances tendered do not suffice to discharge all of the obligations. *30  

5. Conclusions

Over the last few decades, world trade has expanded considerably. Many of the EU Member States export a large proportion of their domestically produced goods and services. The exporters and importers encounter more and more frequently contracts that are subject to foreign laws, which are hard for them to access fully. In spite of this, commercial law remains national law, and there is still much resistance to the idea of establishing a readily accessible European Civil Code. Given this scenario, it becomes increasingly important for the national statutes and codes — most notably, those relating to contracts — to be accessible to foreigners.

As has been seen, the style of the PECL, UPICC, and CISG did not persuade the Germans who drafted the new rules on obligations for the German Civil Code in 2001; nor has that style influenced those who drafted the French Avant projet of 2005to substitute the rules on obligations of the French Code civil.Admittedly, drafting laws for domestic use is not the same as drafting uniform rules. Those who draft for domestic use do not have to resort to a language for which there are words of common contents in the various languages. This does not, however, mean that the draftsmen should prepare rules that the citizens cannot understand.It is, therefore, submitted that the CISG style as described by Honnold and advocated by Hubershould be of importance in the mind of anyone who drafts laws.