Go to Database Directory || Go to Bibliography

Excerpt from Jan H. Dalhuisen, Dalhuisen on International Commercial, Financial and Trade Law, Hart Publishing (2007) 1,358 p. Section 2.3 of this very helpful book is reproduced below with the permission of the publisher. Click here to order the entire text from the publisher's online bookstore; for U.S. orders, click here.

The Uniform International Sales Law

Jan H. Dalhuisen

2.3.1      Usage and Scope
2.3.2 The System of the Vienna Convention: Directory or Mandatory Rules
2.3.3 Applicability of the Vienna Convention
2.3.4 The Sales Law of the Vienna Convention: Remedies
2.3.5 Supplementation and Interpretation of the Vienna Convention
2.3.6 Interpretation of International Sales Contracts under Vienna Convention:
Meaning of Conduct and Custom in Terms of Contract Interpretation
2.3.7 Supplementation of the Vienna Convention: Private International Law and
the Rome Convention on the Law Applicable to Contractual Obligations
2.3.8 The Main Rules of the 1980 EU Rome Convention on the Law Applicable
to Contractual Obligations
2.3.9 The Vienna Convention and the Different Trade Terms in International Sales
2.3.10 Incoterms: Their Status and Relation to the UCC and Vienna Convention
2.3.11 The Vienna Convention and the ICC Model International Sales Contract.
The 2004 Principles of European Law: Sales


International sales have been the subject of uniform treaty law, first in the Hague Conventions of 1964, which separated the formation of the contract from the more substantive parts of the law of sales, and subsequently in the already much mentioned Vienna Convention on Contracts for the International Sale of Goods (CISG) of 1980. It is practically speaking the successor to the Hague Conventions and combines both areas in one document.

It may be helpful at this stage to relate some of the history of the Vienna Convention. It was prepared by UNCITRAL, the UN Commission on International Trade Law established on 17 December 1966 and operating from Vienna since 1969, see also chapter 1, section 1.5.6.

But the idea of a uniform sales law originally emerged in UNIDROIT (Institute International pour l'Unification du Droit Privé or The International Institute for the Unification of Private Law) set up in Rome in 1926 under the authority of the League of Nations. In 1928 Professor Ernst Rabel from Berlin, later at the University of Michigan in Ann Arbor, proposed a project on international sales which led to a first draft in 1935 whilst his own book, Das Recht des Warenkaufs, first appeared in 1936. The project met considerable opposition, especially in the field of offer and acceptance (where common law maintained a much more formal structure).

Since 1939, there were as a consequence two projects, one on formation: Loi Uniforme sur la Formation des Contrats de Vente Internationale des Objets Mobiliers Corporels (LUF) or the Uniform Law on the Formation of Contracts for the International Sale of Goods (ULFIS) and another on the international sale itself: Loi Uniforme sur la Vente Internationale des Objets Mobiliers Corporels (LUVI) or the Uniform Law on the International Sale of Goods (ULIS). They were agreed at the Hague Diplomatic Conferences of 1951 and 1964 which, however, was attended by only a small number of countries, and they were only ratified by a few states, often with reservations (the UK, for example, required that the parties themselves made the Conventions applicable to their contract, which was never done).

Because of their limited support, the Hague Conventions were increasingly seen as not fully representative of the views of the world community at large and especially not of the need of less developed countries. Once UNCITRAL adopted the project, it moved quickly and on a much broader front of participants. It presented a first draft on international sales in 1978 and was able to conclude the matter at a Diplomatic Conference in Vienna in 1980 attended by 62 states, which approved the final text after five weeks of deliberations.

Nevertheless the Vienna Convention as the successor to the Hague Conventions tracks much of their language. In chapter 1 section 1.5.6, the tribulations of UNIDROIT after World War II were already explained. Although created under the League of Nations, it is not now a UN body but rather a private think-tank. When it was reestablished, the sales project was in the hands of an independent Hague Diplomatic Conference which completed the project after conferences in 1951 and 1964. From there it moved to UNCITRAL which, as a UN agency, was thought to be the more proper forum likely to obtain the most ratifications.[214] [page 400]

In the Hague Conventions of 1964, there was no definition of what an international sale was. They limited themselves, however, to the sale of goods as tangible, movable objects. The Vienna Convention does not contain a definition either but notably does not apply to the sale of consumer goods, auctions, execution sales, sales of investment securities and other negotiable instruments or money market transactions, sales of ships or aircraft and sales of electricity (Article 2). The civil or commercial nature of the sale is otherwise not considered (Article 1(3)). For the Convention to apply, parties must either have their place of business in different Contracting States or the applicable private international law should point to the law of a Contracting State for the Convention to apply (Article 1(1)), always assuming that both parties come from different States which in that case need not be Contracting States. If a party has more than one place of business, for example, if it operates through branches in different countries, the place of business with the closest relationship to the contract and its performance counts. If a party has no place of business, the habitual residence will be taken as relevant, Article 10.

The inference is that only cross-border sales between professionals are considered. The Hague Conventions indeed required cross-border movement of goods, but this criterion is now abandoned. It may be more appropriate where proprietary aspects are also considered, which has not so far been the case in the sales conventions. The Vienna Convention is as a consequence applicable even if no assets cross borders. This is in line with its purely contractual scope.

One thus sees a number of elements of international sales implied, cf more in particular section 2.1.2 above, without (for purposes of the uniform law) any single determining factor, although the emphasis is (for contractual purposes) on the different location of the parties, whilst as to substance, the emphasis is on the sale of movable tangible assets between professionals, which assets or parties need not themselves move cross-border. The system of the Vienna Convention is still based on delivery 'ex works', cf Article 31(c), so that again from a proprietary point of view, there may not be any international (cross-border) aspect at all.

The other main differences between the Hague and Vienna Conventions are in a narrower concept of uniform law under the Vienna Convention. It notably allows the application of conflict rules where the Convention itself is silent (but only after applying its general principles, Article 7(2)), see also section 2.3.7 below. Here the internationalist approach of the Hague Conventions is abandoned (see more in particular section 2.3.5 below). The Vienna Convention in its Article 9 is also more restrictive in its recognition of custom as source of law, apparently in order better to protect the unwary in international trade.

The Vienna Convention generally appears to protect the buyer and therefore never accepts automatic termination upon breach, even if fundamental (Articles 61 and 64), gives the defaulter some possibilities to remedy his failure of performance even after delivery (Article 48), especially if the price is paid (Articles 63 and 64(2)), only accepts suspension (and no avoidance) of the contract if there are good reasons for fearing that the buyer will not pay (Article 71), allows avoidance in the case of anticipatory breach only upon notice of the intention to invoke this remedy (Article 72(2)), enhances the buyer's rights unilaterally to reduce the price upon (alleged) non-conform delivery (Article 50, often seen as an excessive right), allows inspection by the buyer within a short period after delivery rather than promptly (Article 38(1)), refers in the absence of a price to the market conditions rather than to prices generally charged by the seller (Article 55), and restricts implied usages to those agreed or established between the parties (Article 9), as they were otherwise considered to favour the strong. The notice [page 401] periods are generally more flexible than in the Hague Conventions and the Vienna Convention never requires prompt answers.

Finally, there are a number of important simplifications, especially in the remedies. The Hague Sales Conventions had distinguished between five categories of performance, each with their own type of remedy in the case of default. It resulted in artificial distinctions and unnecessary complications. Also the concept of delivery was simplified and reduced to its physical aspects and to a number of obligations of buyer and seller in this respect (Articles 31 and 60), whilst the passing of risk was no longer immediately tied to it (Articles 66ff), although the principle that the risk passes upon physical delivery remains unchanged.

As already mentioned above, the Vienna Convention does not apply to consumer sales, the supply of services, auctions, the sale of securities and ships or aircraft (Articles 2 and 3). Even then, the Vienna Convention contains only a partial codification of the law of sales as it relates to the formation of the sale contract and the rights and obligations of parties there under, Article 4 (although this is a very broad formula), but notably not to questions of validity of the contract (therefore in questions of sufficient consensus and the defects therein) and custom, and to the property aspect of sold goods, including the title transfer. In fact, from the subjects covered it is clear that the scope is even more limited as it avoids all general contractual aspects except for offer and acceptance or formation (in Part II).

Areas not covered by the Vienna Convention (and earlier the Hague Conventions) are notably: (a) consent or lack thereof or the usual defenses against binding force in the case of fraud, coercion or duress, abuse of power or undue influence at the time of formation, mistake or misrepresentation, and its consequences, in civil law often referred to as matters of validity, which then also cover questions of avoidance and nullity for other reasons like illegal purposes, (b) capacity and proper authority, including agency, or the consequences of lack thereof, (c) pre-contractual rights and duties, (d) enforcement or abuse of rights obtained under the contract including any re-negotiation duties in the case of hardship or gross imbalances, (e) the transfer of title, its manner and time and the property rights created thereby, and (f) prescription (the UNCITRAL Convention of 1974 on the Limitation Period in the International Sale of Goods suggests a period of four years). By special provision (Article 5), the Vienna Convention, does not apply to seller's liability for death or personal injury caused by his goods either.

The areas not covered by the Convention thus largely relate to the general part of the law of contract or to property aspects, especially the transfer of title pursuant to the contract. What we are mainly concerned with here is the typical contractual infrastructure like questions of capacity, binding force or consensus and the interpretation of the contract, defenses and excuses, pre- and post-contractual duties, and the proprietary consequences of a sale, aspects not likely to be covered in the contract itself either, certainly not if not themselves of a contractual nature, like the transfer and ownership aspects. Because these aspects cannot or are not normally covered in the contract itself, a Convention also covering these areas would have been much more important.

For the contractual infrastructure, the UNIDROIT and EU Contract Principles are now attempting to cover some of this (see section 1.6 above) and may, whatever their strengths or weaknesses, from that point of view become more important, although they are not meant as a binding set of rules. In fact, the Vienna Convention (with the important exception of its provisions on formation relating largely to offer and acceptance) covers particularly those aspects which parties are themselves likely to cover in the contract, like quality, quantity, price, place and time of delivery, default and the in personam remedies including the concept of [page 402] force majeure and the passing of risk. Parties may certainly continue to do so (Article 6). It obviously reduces the importance of the Convention even further. Because of unfamiliarity with its operations or effects and some peculiarities as the probably excessive power of the buyer unilaterally to adjust the purchase price under Article 50, many standard industry contacts exclude the application of the Convention altogether. In any event, it mainly centers on some aspects of offer and acceptance (Part II), on the time and place of delivery (Articles 31 and 33), conform (physical) delivery (Article 35), transfer of risk (Article 69(1)), payment (Article 57), default (Articles 45 and 61), damages (Article 74), force majeure (Article 79) and a duty of care for seller or buyer if something goes wrong in the implementation of the contract in order to protect the goods (Articles 85 and 86). The Vienna Convention is as a consequence of modest help and primarily meant to provide solutions when the contract, standard terms or industry usages (see Article 9) do not provide sufficient guidance themselves, which is usually only the case in oral contracts. They must be considered rare in international dealings.

The USA and all EU countries, except the UK, have ratified the Vienna Convention. The UK was represented on the UNCITRAL working group at the time but remains one of the most important commercial nations that have not so far accepted the Convention, although it ratified the earlier Hague Sales Conventions (but with the reservation that parties had to make a reference to it in their contract). The principal reason for the UK's abstention so far may be the one stated by the Law Society of England and Wales concerned about the diminishing role that might result for English law within the international trade arena.[215] Its own sales law for goods or movable tangible assets remains basically the one of 1893 updated in 1979 and may be doubted ever to have set a pattern in international sales, except within the British Commonwealth. Article 2 of the UCC is more influential and emerged in draft very much at the same time as the Hague Conventions. The Vienna Convention is still different in a number of important aspects, notably in the absence of the concept of consideration and in the introduction of good faith notions in the interpretation and of general principles in the supplementation of the Convention itself which may suggest a more teleological or normative interpretation of the legal text. These would certainly be new features in countries like the UK, although less so in commercial law in the USA (see Section 1-103 UCC which in any event allows a liberal interpretation of the statutory text).

In sum, in the more flexible approach to interpretation and in its detail, the USA with its Article 2 of the UCC is closer to the Convention and the USA has therefore not shown a similar resistance and ratified. Unification makes some sense as a system of different domestic sales laws increases cost in terms of legal advice, creates some risks on its own, and may well slow down transactions. It also increases uncertainty. Moreover, there result considerable uncertainties and undesirable consequences from the applicable choice of law rules or even from a contractual choice of a domestic law, see further chapter 1, section 2.2.9. Most importantly, domestic laws tend to ignore the character and dynamics of international sales; see more in particular chapter 1 for the development of a transnational legal order and lex mercatoria to accommodate modern needs. [page 403]

Application of the Vienna Convention alone will not guarantee uniformity as it covers the subject of international sales only partially and notably not the aspect of the transfer of ownership which is the true objective of all sales contracts. In any event, there will remain diversity in its interpretation and certainly in its supplementation where, for the one, reference is made to the widely varying good faith notion and, for the other, to private international law rules, see more in particular sections 2.3.6 and 2.3.7 below.

It is submitted that uniform treaty law can only flourish and its true meaning only become apparent within the context of the hierarchy of norms of the whole lex mercatoria, see chapter 1, section 1.5.4.


The question was raised at the time whether provisions of the Hague Sales Conventions could be of a mandatory instead of a directory nature, especially important as to whether they could in whole or in part be set aside by the parties. Article 6 of the Vienna Convention, following Article 3 of the Hague Sales Convention, is unambiguous on the subject and allows parties to deviate from any of its terms, except where the domestic law of a party with its place of business in a Contracting State requires a written form for the contract of sale, its modification or termination, provided that the relevant Contracting State has insisted on maintaining this principle in international sales upon a special declaration to the effect (Articles 12 and 96) ...

Earlier, an argument developed at the Hague Diplomatic Conference of 1951 to the effect that international trade required some basic mandatory rules on the structure of the international sales agreement, notably to avoid improper practices. This never became the accepted view which remains based on the classical notion of contracts in essence depending on the will of the parties in formation, coverage and interpretation, although corrections would be conceivable on the basis of fundamental legal principles or mandatory custom. They may then be identified as overriding in the context of the lex mercatoria (see chapter 1, section 1.5.2). In international sales, these overriding principles should be of an international nature and domestic considerations should not enter into them--they could at most be a bar to enforcement of such decisions in domestic recognition and execution proceedings. These overriding principles may sometimes be covered by the concept of good faith or abuse of rights notions. Domestically this is known especially in legal systems like the German and Dutch where in pressing cases they may equally amend the contract and its implementation (see more in particular section 1.3.1 above). Again the likely reason is the appeal to fundamental legal notions.

The liberal attitude of the Convention towards party autonomy was made uncontroversial because of its limited scope. The UNIDROIT and European Contract Principles which cover contract law more generally seem to adopt here a fundamentally different attitude, therefore not limited to acceptance of overriding international fundamental principles alone. That was identified in section 1.6.1 as largely the result of a personal dealing and consumer-oriented approach, even where these Principles (especially the UNIDROIT ones) mean to operate only in the international business sphere. It is bound to undermine the credibility of these Principles in international trade (see also the discussion in section 1.6.3 above). [page 404]


The Vienna Convention only applies to the sale of goods or chattels as tangible movable assets between parties whose places of business are in different States. According to Article 1(1), for the Convention to apply, the further requirement is that either the parties must have their places of business in Contracting States or the applicable private international law must point to the law of a Contracting State, although Contracting States may opt out of this latter provision (Article 95), as the USA has done.

If a party has more than one place of business (for example, branches in several countries), the relevant place of business for the purposes of the Convention is the one which has the closest relationship to the contract and its performance, having regard to the circumstances known to or contemplated by the parties at any time before or at the conclusion of the contract (Article 10(a)). If a party does not have a place of business, like most private persons, the habitual residence is the relevant place in terms of the Convention (Article 10(b)). The Convention appears to consider it exceptional that they should engage in international sales of the type covered by it although these sales are not strictly speaking excluded, provided they are not consumer sales.

For the Convention to apply it is thus not necessary for the courts in Contracting States first to determine which law is applicable in order subsequently to determine whether that law has incorporated the Convention for sales of the types covered by it. This is the approach only in the courts of non-Contracting States. Courts in Contracting States must apply the Convention as lex fori if the conditions of Article 1 are fulfilled, that is particularly when the parties come from two Contracting States. The appropriateness of applying the international sales law is then automatically assumed.

Thus only if the parties are not both from Contracting States is there a more traditional conflicts rule inserted and the Convention then only applies if the forum's conflicts rules lead to the applicability of the law of a Contracting State (although under Article 95 Contracting States may opt out of this provision). In finding that the law of a Contracting State is applicable, courts would only subsequently apply the other applicability tests of the Convention, that is that parties came at least from different states and that in essence only non-consumer sales are covered (unless these conditions are waived by the parties), see further also chapter 1, section 2.3.1.

It follows that the Convention may under its applicability rules affect residents from non-Contracting States in respect of whom the law of a Contracting State must be considered applicable under Article 1(1)(b) provided always that the parties come from different States even if not Contracting States (which is logical as there is no reason to apply an international sales law in domestic situations). A contractual choice of law referring to the law of a Contracting State may in this way also achieve applicability of the Convention as a matter of the conflict laws of the forum in or outside Contracting States. It may be safer in such a case, however, to refer to the Convention explicitly, certainly if the law of the country chosen has itself opted to exclude Article 1(1)(b), as the USA has done.

Article 6, on the other hand, allows the parties expressly to exclude application even if they both come from a Contracting State or risk the applicability of the Convention under the prevailing conflict rules. A choice of law in favour of a country that has not ratified the Convention must mean its exclusion. On the other hand, a contractual choice of law making reference to the Convention would mean its applicability even if the parties came from the same State. In that case the diversity requirement of Article 1(1) must be deemed waived. This may make sense if the goods are to be shipped from other countries and the parties, although from the same country, wish the provisions of the Convention concerning quality, quantity, price, allocation of risk and default to apply. [page 405]

As was already mentioned, total exclusion of the Convention is still quite common between commercial parties because of some of the vagaries and eccentricities of the Convention or general unfamiliarity with it.

The result of the application provisions of the Convention is that a German and a Dutch resident, both coming from Contracting States, but the latter being on a visit to Germany, would be selling to and buying from each other under the Vienna Convention, except if the goods were bought for personal, family or household use (unless the seller did not realise that this was the case: see Article 2(2)). According to Article 1(2), this is also not so if the fact that the parties have their places of business in different countries does not appear from the contract or from any dealings between them.

This could be more particularly the case if there was an undisclosed foreign principal represented by an agent resident or with his place of residence in the same country as the other party. In any event, the buying of a loaf of bread by the Dutchman whilst on holiday in Germany would as a consumer transaction not be covered by the Vienna Convention, although according to Article 1(3) neither the nationality of the parties nor the civil or commercial nature of the contract is relevant in this connection. The buying of a combi-car thus probably would.

If, on the other hand, the buyer were an Englishman (therefore someone from a Non-Contracting State) on a visit in Germany, the Convention would still be applied by the German courts if called upon to decide any litigation arising from the sale (assuming they had jurisdiction) as the parties would have their habitual residences in different States (although not Contracting States) whilst under Article 1.1(b) the Convention would apply as a matter of German law, this being the law of the seller who under Article 4 of the EU Rome Convention on the Law Applicable to Contractual Obligations performs the most characteristic obligation.

Should a case arise in England, then the Vienna Convention might still apply if German law were considered applicable to the transaction by the English courts, which would be likely in the circumstances as the UK is a party to the Rome Convention and would therefore be likely to apply German contract law. In that case the English courts would not apply the Convention as part of their lex fori but only as a matter of German law under the applicable conflicts rule.

In this system, the location of the asset or any requirement of delivery in another country is irrelevant. This is a consequence of the property aspects and the transfer of title itself not being covered by the Convention (see Article 4). As we saw, the earlier Hague Conventions still required this cross-border movement of the asset even though it did not cover proprietary aspects either, but this approach is now abandoned.


As already mentioned several times before, the Vienna Convention contains only a partial codification of the international sales law which is always subject to party autonomy. In the following, its major topics, most of which were already highlighted before, will be briefly revisited.

As regards offer and acceptance, the Convention largely adopts common law terminology, concepts and practices, except that the acceptance (if not by conduct or in the manner as indicated by established practice between the parties) is effective upon receipt only, even if by mail, rather than upon dispatch, as is the normal common law rule (Article 18(2)). Like the UCC in the USA, the Convention further dispenses with the traditional common law distinction between unilateral and bilateral contracts, the [page 406] first being under traditional common law for their effect dependent on full performance (and notice thereof to the other party, instead of mere acceptance), as in offers to the public (if not an invitation to treat) and normally in offers for services, not here relevant.

On the other hand, unlike the situation in civil law, offers are not binding per se, even for a reasonable time (Article 16). Here traditional common law continues to prevail, except where the offer states otherwise (which was not even then binding under common law without consideration, but between merchants now also possible under the UCC in the USA (Section 2-205)), or has been reasonably relied upon (like subcontractors' quotations enabling a main contractor to make a final offer). Common law requires detrimental reliance, which may be a stricter requirement. Offers must be sufficiently definite, but they are considered to be so if they fix or make provision for the determination of quantity and price, see Article 14(1), although strictly speaking even that much is not necessary if the offer is otherwise sufficiently definite. Article 55 makes further provisions for price so that in fact quantity is the only term parties must at least generally agree on.

This is also the approach of the UCC in the USA, which requires as the minimum for validity of a sales contract the agreement on quantity. It need not even be accurately stated but recovery is limited by what is said, whilst the price, time and place of payment or delivery, the general quality of the goods or any particular warranties may all be omitted (see Official Comment at Section 2-201). Under the Convention, additional terms, which do not materially change the offer, do not result in a rejection or counter offer, but if bearing on price, payments, quality, quantity, place and time of delivery, party's liability or settlement of disputes, they are considered material per se (Article 19(3)).

The Convention does away both with the requirement of consideration and causa (cf Articles 14 and 29(1)) in the formation. In this respect it also follows the earlier Hague Sales Conventions. Thus under the Vienna Convention, there is no search for a more objective criterion to make promises legally binding besides the mere intention of the parties. In the continental European tradition, any excesses are curbed only by notions of good faith or public policy, but these notions remain underdeveloped in the Vienna Convention. The good faith notion is only mentioned in Article 7(1) in the context of the interpretation of the Convention itself and has here supposedly a different meaning and may more properly stand for teleological or normative interpretation, as we shall see.

The absence of the requirement of consideration or causa in the Convention (and also in the UNIDROIT Principles for International Commercial Contracts) does not appear to mean that it may itself be reintroduced through application of a national law pursuant to applicable conflicts rules under Article 7(2) as a matter of gap-filling in the area of formation or outright as a matter not covered by the Convention at all. It may well be that in international transactions these requirements (of consideration or causa) have lapsed altogether, see also section 1.2.6 above.

Articles 31 and 33 deal with the time and place of the delivery. The normal delivery place remains 'ex works', that is at the seller's place of business, except where the sales contract involves a carriage of goods, therefore a transportation arrangement, when the seller must hand over the goods to the first carrier. If the goods are known to be in a warehouse or particular place of manufacture, they must be put at the buyer's disposal at that place. The time is a reasonable time after conclusion of the contract, which means in fact that the ground rule is immediate delivery at the option of the buyer to be reasonably exercised. [page 407]

The Convention thus concerns itself only with delivery in a physical and not a legal sense. As just mentioned, in essence goods are still put at the disposal of the buyer 'ex works'. The transfer of risk occurs at the same time, although strictly speaking it is at the moment the buyer takes over the goods (Article 69(1)), which could be later. In trade terms, like Fob and Cif, parties often agree, however, to a different regime altogether, see section 2.3.9 below. As proprietary aspects are not dealt with in the Convention, delivery must not be considered here in terms of transferring legal or constructive possession as a precondition for the passing of title in countries which require delivery for title transfer. They may in any event sometimes require mere legal or constructive delivery and therefore use a somewhat different concept of possession in this connection, see more particularly chapter 4.

As regards price and payment, although Article 14 of the Vienna Convention requires the price to be sufficiently definite, according to Article 55, the price if not established in the contract is the price generally charged at the time of the conclusion of the contract for such goods sold under comparable circumstances. Payment must take place at the place of seller (Article 57), unlike notably under French law (insisting on the place of the debtor, Article 1247(3) Cc) and the time is upon delivery (Article 58), all unless agreed otherwise.[216] The European and UNIDROIT Principles also require payment at the place of the creditor.

In Articles 28, 46 and 62, the Vienna Convention deals with specific performance, see also section 1.4.1 above. Subject to the provisions of the lex fori in this manner, the Convention in Article 46 allows the buyer specific performance unless it has resorted to another remedy. In the case of non-conform delivery, substitute goods may only be claimed if there has been a fundamental breach, otherwise the normal remedy is repair. Article 62 gives the seller a more unrestrained right to request payment or force the buyer to take delivery or perform his other duties unless he has chosen another remedy.

As regards other remedies, see also section 1.4.4 above for all types of contracts. To repeat, under the Vienna Convention, Article 47 gives the buyer the facility to set extra time (Nachfrist) for performance by the seller and Article 63 gives a similar facility to the seller to solicit performance by the buyer. The importance of this facility is in its consequences: if the defaulting party does not use the facility, the other party has a right to avoid the contract and thereby terminate his own performance duty, whether or not the breach was itself fundamental. It is then treated as such: see Articles 49(1)(b) and 64(1)(b).

As regards the default remedies of rescission and damages, the key is that, under Articles 49 and 64, a party can ask for a rescission of the contract upon default by the other party, but in the case of fundamental breach only (see for the definition Articles 25 and 26) or if he has set an additional time for performance and the other party has not taken advantage of it. It does not require judicial intervention and discharges the creditor (either seller or buyer as the case may be) from his own obligations and does not rule out any claim for damages under Articles 74 to 77. These may include a loss of profit, incidental (for example, enforcement cost) and often consequential damages, but they cannot exceed the loss foreseeable by the defaulting party at the time of the conclusion [page 408] of the contract. This is the common law approach, which maintains here a more difficult criterion than civil law which ties the foreseeability commonly to the moment the default occurs.

Article 50 contains a special and much criticised remedy in the case of non-conform delivery and allows the buyer unilaterally to reduce the price with the reduction in value of the asset. It does not require judicial intervention to establish whether the delivery was indeed non-conform and what the appropriate reduction is. Article 50 is a combination of the actio quanti minoris and the exceptio non adimpleti contractus except that there is neither an action nor an exception but rather an independent right on the part of the buyer unilaterally to reduce the price upon his allegation of non-conform delivery. It shifts the burden of litigation to the seller. This has given rise to considerable criticism but was asked for by developing countries which felt that their buyers were better protected in this manner. It is an important reason, however, why the application of the Vienna Convention is often excluded by contract. Where the non-conform delivery was due to force majeure whilst the seller was still at risk, he will, however, be excused from paying damages (see Article 79).

As to default, under the Convention, parties may also act in anticipation of a breach to preserve their positions (the anticipatory breach of Articles 71-73) and it is possible to suspend their own performance by giving notice thereof to the other party, provided the deficiency in the ability to perform or in the credit worthiness of the other party is serious or is in preparation for performance or in the performance itself. If it results in adequate reassurances of performance, the party giving the notice must continue his own performance.

The Vienna Convention deals with performance and default only in their in personam consequences, that is to say it deals neither with the transfer of title nor its return upon default or upon rescission or avoidance: see for the various options under domestic laws in this latter respect, section 2.1.10 above. The remedies under the Convention are basically specific performance or avoidance of the contract (if the breach is fundamental) with the possibility of claiming damages depending on the circumstances. As we saw, by way of preventive remedies, the Convention also allows preventive action if a breach is expected (anticipatory breach). It may also mean not performing any further oneself. This is particularly relevant in bilateral contracts (the exceptio non adimpleti contractus) of which the sales contract is a prime example.

Technically speaking, the Vienna Convention, like the UCC, abandons the common law notion of (express or constructive) 'conditions', traditionally distinguished from warranties as a matter of interpretation. The express conditions could be related to the occurrence of an outside event or be a specific term of the deal, for example the specified quality or delivery date. Constructive conditions were those implied, for example a substantial default by the one party, which would condition the performance of the other and release him. As explained in section 1.4.3 above, in common law, technically conditions were considered so material that when remaining unsatisfied, the contractual obligation of the other party did not ripen unless there was a failure to co-operate or a waiver. In this way common law created a method of releasing the non-defaulting party without rescinding the agreement, which it found difficult to do.

Conditions thus automatically discharged a promisor without resort to the notion of breach proper and any right to cure, whilst there could still be a right to damages. Warranties did not have the same status, although they naturally could also be breached.

The Vienna Convention adopts instead the concept of fundamental breach and speaks of the possibility to avoid or rescind the contract in that case (only): see Articles 49 and 64. Fundamental breach is only loosely defined (Article 25) and remains, like negligence, a matter of determination per case in view of all the circumstances. [page 409]

The subject of force majeure is always a difficult one and concerns first the definition of force majeure but also its consequences in terms of a temporary or permanent release of the party required to perform, adaptation of the contract or allocation of the risk: see also section 1.4.5 above.

The Vienna Convention in Article 79 states that a party is not liable for a failure to perform any of his obligations if he proves that the failure was due to an impediment beyond his control and that he could not reasonably be expected to have taken the impediment into account at the time of the conclusion of the contract or to have avoided or overcome it or its consequences.

This is a broad definition, but Article 79(5) further says that nothing in Article 79 prevents either party from exercising any rights under the Convention other than claiming damages. Rights to avoid the contract as a result of fundamental breach or to ask for repairs are thus not impaired whilst avoidance may lead to restitution, but only if that is still possible (if not, avoidance may no longer be possible but other remedies are reatained, Article 82). It means the return of all that has been supplied under the contract, but not as proprietary remedy with which the Convention does not deal.

The Convention does not contain any special provisions allowing for an adaptation of the contractual terms in the case of hardship due to change of circumstances or otherwise.

As regards the passing of risk of deteriorating or lost goods and its relationship to the concept of force majeure, above in sections 2.1.8 and 2.1.9, the basic rules were explained. In summary these were the following:

(a) In a sale, it is the legal owner or, more recently, the physical holder of the goods who must accept whatever happens to the goods in his possession after the conclusion of the sales agreement and before delivery if damage or loss is not caused by anyone in particular. If he is the seller, he is not therefore discharged from his conform delivery duty by force majeure (but he might from other ancillary obligations like servicing the goods).

(b) On the other hand, the seller, if no longer the holder, is no longer liable for any damage subsequently affecting the goods without his fault and is entitled to full payment. He is also excused from any remaining duties in respect of the goods, like a service obligation, in the case of force majeure on his part but might not be able to request payment for the performance of these duties either. This is only different if guarantees of performance have been given.

(c) The seller's own knowledge of dangers or investigation duties in the quality especially of goods which he did not himself manufacture may undermine his claims to force majeure concerning the state of the goods and therefore his reliance on the passing of the risk upon delivery. Defenses of the buyer based on mistake or misrepresentation may then also be valid.

(d) The buyer upon delivery may still be protected even if the seller is not to blame for the deterioration in quality, if the buyer did not have a chance to inspect the goods properly, if there were hidden defects, or if he benefited from an express or implied guarantee (which may also imply a lesser investigation duty upon receipt) so that the risk is still not for him and has therefore not passed.

(e) If goods (as well as the payments made) may be returned upon nonconformity, not discoverable at the time of delivery (even if due to force majeure but also for other reasons like mistake, illegality, etc.), the seller may have to accept the goods in the state in which they then are (except if grossly mishandled by the buyer) and still may have the duty to provide new ones at his own expense, [page 410] repair the old ones to their specified quality, reduce the price or abandon a claim for payment altogether and return all payments already made. So his risk may be increased by the ordinary wear and tear or any (complete) loss of the goods in the meantime (for reasons of force majeure affecting the buyer in possession).

(f) The seller, whilst having the risk but not being able to perform the quality requirements due to force majeure and therefore still being liable for repair or replacement or the loss of the sale price or for a price reduction, does not need to reimburse the buyer for the loss of the bargain or consequential damages except if he did not disclose known dangers whilst not insuring them properly either.

Thus under the modern law of sales, unless the contract states otherwise, the risk in this sense normally passes between seller and buyer when the buyer takes over the goods, or if he does not do so in good time, from the moment the goods were put at his disposal. This is also the approach of the Vienna Convention, cf Article 69. If intermediaries are used, for example in a Fob or Cif sale, the risk usually passes on loading to the first carrier: see also Article 67 Vienna Convention. For goods in transit, the risk passes upon conclusion of the contract (Article 68). Carrier and transit risk is thus now normally for the buyer. The Convention does not, however, deal directly with the other subtleties of the passing of risk, as just summarised, nor with its fluid character, but many of the exceptions and refinements may result under its other provisions.

As the Convention does not consider the question of the passing of title and whether or when it has passed, the passing of risk (Articles 66ff) acquires for goods a more particular significance. As the ownership question often takes care of itself (resulting under the applicable domestic laws either from the contract or from any subsequent delivery which will in any event be required as a matter of contract performance) and is mostly settled under the applicable law and not by contract, the passing of the risk is often of more immediate concern to the parties (as is its physical delivery) and can always be determined by contract.

As already mentioned in section 2.1.8 above, in international sales under the Vienna Convention, the concept of non-conformity might itself include the concept of mistake (not itself covered by the Convention) as to the qualities of a sold good (Article 35(2)), so that the non-conformity remedies of the Convention apply throughout, including the rule concerning the passing of risk, rather than domestic notions of mistake or misrepresentation, which, in the absence of the coverage of these subjects in the Convention, might otherwise supplement it under applicable rules of private international law (therefore without regard to Article 7(2)).


The Vienna Convention goes into the interpretation and supplementation of its own text. It may be argued that that is less usual or even inconvenient and in any event not absolutely dispositive. Also, the separation of interpretation and supplementation itself may be subject to serious criticism. In a more normative or teleological approach this would not seem to be necessary or even proper. Even in the traditional attitude to code interpretation in civil law, the distinction was uncommon. Applying greatly differing criteria is in any event puzzling. [page 411]

To start with supplementation, questions concerning matters governed by the Convention which are not expressly settled in it must be settled in conformity with the general principles on which the Convention is based and only in the absence thereof pursuant to the national law resulting from the applicable private international law rules. At least so says Article 7(2). These questions are in essence all matters concerning the rights and obligations arising from the sales agreement, cf also Article 4 of the Vienna Convention, including the in personam remedies and excuses, but notably not matters of validity of the contract or of any usage and the proprietary and enforcement aspects of the sales agreement.

This reference to private international law as the last resort was notably absent from the Hague Conventions, cf Article 2 of the Hague Sales Convention, which only accepted conflict references in specific cases, Articles 16, 38(4) and 89. The Hague Conventions did also not distinguish between interpretation and supplementation and Article 17 of the Sales Convention required all matters in principle covered by the Convention to be generally decided in conformity with the general principles on which it was based, thus the first part of the Vienna formula for both supplementation and interpretation of the Hague Conventions. This is in accordance with the general civil law approach to codification which assumes completeness in the areas of the law it covers. Gap filling or supplementation then becomes a matter of interpretation. In the area of uniform law, there is the added argument that this approach allows for a liberal interpretation of that law and of its scope. This approach is therefore in essence uniform law friendly and internationalist.

In areas not covered by the Hague Sales Conventions, conflict of law notions remained naturally applicable, but it was often felt that even in areas that the Conventions did cover this was unavoidable when its general principles were not sufficiently developed. This became the view of the Vienna Convention, at least in the case of supplementation or gap-filling of the Convention. It undermines the reference to general principles itself, which in a codification sense do not strictly need to be identified. There is here an important shift in the nature of the argument and in the Vienna Convention the codification notion and its uniform law friendly attitude seem abandoned. The general principles to which reference is made in Article 7(2) are now probably only the general principles on which the Convention is based in a narrower, more literal sense.

As regards these general principles, in this narrower, non-interpretational or non-judicial sense, it is generally unclear what they are and any party invoking them must then prove them on the basis of its own analysis.[217] They are in that case notably not the fundamental principles of the law in the sense of the ius cogens, customary practices or the general principles common to most commercial legal systems discussed earlier in chapter 1 as part of the hierarchy of the lex mercatoria. [page 412]

Any impact of the fundamental or more general principles of the lex mercatoria is not excluded thereby, however. Their effect depends on the applicability of other sources of the law which the Vienna Convention (and earlier the Hague Conventions) could not regulate, as was more fully explained in chapter 1, section 1.5.2. It is nevertheless of interest that Article 7(2) of the Convention does not even make a reference to (international) custom or usage in this connection as one would have expected. This is reserved for the supplementation of the parties' agreement (and not of the Convention itself) in Article 9, and then only in a limited manner, see the next section. Even so, the validity of custom or usages is not ultimately governed by the Convention itself either, as recognised in Article 4.

It is in any event more normal to talk about supplementation (and interpretation) in the context of the sales agreement itself, with which the Convention does not generally deal. This contractual interpretation and supplementation is only more indirectly covered by the Convention as the reference to the rights and duties of the parties in Article 4 makes clear. Articles 8 and 9, limited to the meaning of statements, conduct and usages, are only relevant in that context (see more in particular section 2.3.6 below).

There is here an element of confusion in the Convention between convention and contract interpretation, as there was in the distinction between interpretation and supplementation more generally. It is clear that a convention of this sort aims at a more general and objective legal regime whilst the sales contract itself is based in principle on party autonomy. This may suggest some differences in interpretation and supplementation techniques but the Convention is not clear on what its aims and policies are in this regard. Again, whatever Article 7(2) of the Convention tries to do in terms of its own supplementation, the norms applicable to international sales and their hierarchy cannot be solely determined by the Vienna Convention itself, even in the areas covered by it.

Thus the failure in Article 7(2) of the Convention to refer (before the rules of private international law) to its own international character (and thus perhaps implicitly to international custom or usage) and to the need for uniformity, as it does in Article 7(1) in the context of interpretation--where therefore (correctly, it is submitted) a distinct legal order is suggested--has no ultimate relevance in this regard as these considerations may intervene anyway, also in Article 7(2). It suggests the relevance of sources of law extraneous to the Convention itself to which the Convention may be subordinated (both in supplementation and interpretation). In fact, it is submitted that they (as well as the Convention) can only be given their proper place within the hierarchy of norms of the lex mercatoria, as explained in chapter 1, in which therefore the uniform sales laws must also find their place. In this hierarchy, even party autonomy figures above the Convention to the extent the latter is only directory (defult) law, see Article 6.

The omission of a reference to these other norms in the Convention is nevertheless unexpected and leaves it incomplete.[218] In may be noted in this respect that Section 1-103 UCC in the US is much clearer. This might only be understood from the point of view of a general wariness of codification and general principle in common law, of custom in civil law, and more generally by an unclear view of the international law merchant and its operation among the draftsmen of the Convention.

As compared to the Hague Conventions, the Vienna Convention (at least in supplementation) lost much of the internationalist approach in the process, but again this cannot prevent it from having to operate in the context of all other norms applicable to the international sale of goods. When it comes to supplementation or gap filling, in the internationalist lex mercatoria approach, transnationalised notions of good [page 413] faith may also play a role in the supplementation, even if in the Convention only mentioned in Article 7(1) in connection with its interpretation (as distinguished from supplementation). This is so certainly where good faith appeals to fundamental principle but also if appealing only to more common legal principles. In a lex mercatoria approach based on a hierarchy of norms, this would be clear. In respect of the Convention more narrowly, the reference to good faith is likely to have a distinct meaning and may refer to teleological or normative interpretation (and supplementation) of the text. Thus the literal approach to statutes, treaties and contractual terms, normally adopted by the UK and towards which there may also be a bias in the Vienna Convention (certainly more so than in the Hague Sales Conventions), may be somewhat relaxed, also it would then appear to follow for supplementation.

In the context of supplementation of the Convention, the reference to private international law after the reference to the general principles on which the Convention is based is, as already suggested above, careless and can easily destroy what little progress is made through the Convention in terms of international uniformity. It narrows in any event the scope of the uniform law. In particular it may even apply to the terminology used in the Convention which on the whole lacks definitional clarity. Where the Convention relies on the rules of private international law, as it does also in determining its scope in Article 1.1(b), for international sales cases pending in EU courts this means the rules of the 1980 Rome Convention on the Law Applicable to Contractual Obligations (see section 2.3.7 below).

Again, the reference to private international law and therefore to domestic law to supplement the Convention can only be properly understood in the context of the hierarchy of norms within the law merchant or lex mercatoria concerning international sales, where it has the lowest rank and its application is discretionary (see also chapter 1, section 1.5.4). In a proper internationalist approach, the appropriate reference for supplementation and interpretation of the Convention, if it needs to be covered, should have been to the text of the Convention itself, its international character, its general principles, and to the need for uniformity in its application. At the same time, it should have been made much clearer that other sources of law remain unimpeded, as the UCC states for itself in its Section 1-103, so that fundamental principle, customs at they develop as well as other general principles remain unaffected (besides party autonomy which is clearly respected in the Convention and supercedes it, see Article 6). The observance of good faith could be added but is in international trade more an issue of interpretation (and supplementation) of the contracts concluded under the Convention therefore a question of party autonomy and its meaning unless, as suggested above, it refers more generally to a normative or teleological interpretation method. Private international law would only come in if no objective legal regime could otherwise be found in the hierarchy of norms here presented. The issue of public policy under state law and its prevalence in international transactions is here another issue which cannot be settled by traditional private international law either (see chapter 1, section 2.2.6).

To repeat, only in respect of its interpretation in Article 7(1) does the Convention refer more specifically to the need to consider its international character, to promote uniformity in its application and even to the observance of good faith in international trade. As already mentioned before, the latter reference did not exist in the Hague Sales Convention either, which in this aspect relied also on its general principles (Article 17).[219] [page 414]

The references to internationality and uniformity, albeit only in the context of interpretation, suggest at least that the Convention operates in a distinct legal order and that its provisions are autonomous, although this does not prevent consultation of domestic legislation, case law and doctrine by way of guidance. Guidance may of course also be obtained from the history of the Hague and Vienna Conventions and their application and now perhaps also from the EU and UNIDROIT Contract Principles. Whatever we make of the use of the notion of good faith in this connection, because of the vagaries of the notion, its true significance under the Convention could, technically speaking, ultimately still remain a question of determination under a national law pursuant to Article 7(2), notwithstanding the reference to the requirement of uniformity in the application of the Convention under Article 7(1) and the reference to 'international trade' which indeed suggest a distinct set of norms. Wildly different good faith notions from domestic law could in this manner be introduced through the conflict rules. This route should not be chosen and good faith should indeed be seen here as a distinct internationalised concept, in the context of the interpretation (and supplementation) of the Convention itself referring probably to a teleological and then perhaps also a normative interpretation method, even if in civil litigation domestic interpretations are still likely to impact because of the general outlook and attitudes of judges groomed in their national systems of law.

It follows that whilst it is likely that the Vienna Convention itself in its approach to interpretation and supplementation maintains much respect for literal meanings and probably even a psychological attitude to contractual intent, borne out in the interpretation provision of Article 8 in respect of statements by the parties, this is not necessarily the end of the story. It might reflect the prevailing attitudes during the formative period of the Convention, which was largely in the first half of the 20th century, and also some common law input. Strictly speaking, the teleological or even normative approaches to interpretation of both the wording of the Convention and the sale contract appeared much later and were probably never substantially considered, not even at the time of the Vienna Convention in 1980.

Yet it might now be seen as implicit in the good faith interpretation provision that came in with the Vienna Convention [220] which may otherwise have narrowed the interpretational freedoms inherent in the language of the earlier Hague Conventions. In this respect, the Vienna Convention may yet show some balance, even if it moved in a confusing way with a confusing terminology. This is not to ignore that it took a great step backwards when reducing the impact of general principles and relying on private international law, be it only within its (undefined) concept of supplementation. Again, putting the uniform law in its place in the lex mercatoria hierarchy of norms is here the proper answer.

The formula of Article 7 of the Vienna Convention may (unfortunately) now be found in all UNCITRAL Conventions. It may also be found in the European Contract Principles (Article 1.106). It is even in the UNIDROIT Leasing and Factoring Conventions, where the reference to good faith may be entirely inappropriate as they also cover proprietary matters. It is not, however, the approach of the UNIDROIT Contract Principles (Article 1.6), which revert here to the approach of the Hague Sales Conventions and seem therefore more favourably disposed towards internationalisation. Both sets of Contract Principles also contain elaborate sections on the interpretation of contracts, see section 1.6.3 above. [page 415]


In the previous section it was explained that the Vienna Convention does not deal in any general manner with the interpretation (and supplementation) of the sales contracts governed by it. It only deals with interpretation and supplementation of the Convention itself. The contractual interpretation and supplementation is an issue that in principle seems to be covered by the Convention, however, in view of the reference in Article 4 to the rights and obligations of the parties arising from the sales contracts. The coverage of these rights and obligations by the Convention must imply interpretation and supplementation, if at all distinguishable. This is supported by the fact that in Articles 8 and 9 there follow some more precise interpretation rules, albeit only limited to parties' statements and conduct and to the impact of usages.

Surprisingly, there is in this context no reference to good faith and the contract interpretation rules must therefore largely still be found through the supplementation rule of Article 7(2) with its reference to the general principles on which the Convention is based, which are few in this connection, and otherwise only to the law applicable by virtue of private international law. It is likely to introduce here major variations in national notions and within that context also to good faith.

With respect to the interpretation and supplementation of the Convention itself this might be avoided, as discussed in the previous section, by virtue of its place within the lex mercatoria hierarchy of norms or on the basis even of its own provisions. However, there could be more doubt with respect to the contractual interpretation and supplementation where party autonomy and intent is a key issue, although the lex mercatoria and its hierarchy should equally apply to them in terms of objective law.

Article 8 contains a distinct contract interpretation provision of some sort. Under it, the parties' statements and conduct are to be interpreted according to their true intent, at least to the extent that the other party knew or could not have been unaware of what it was. It ties interpretation principally to psychological intent and the parties' will, but if it cannot be determined whether the other party knew or could have known of the true intent, it allows a reasonable understanding. In determining intent or reasonable understanding in this context, due consideration must be given to all relevant circumstances of the case including the negotiations, practices and usages and any subsequent conduct of the parties, taking into account perhaps also the internationality and commerciality of the transaction and the professionality of the parties.

For a start, it is clear that Article 8 does do away with the common law 'parole evidence rule' which did not allow any contradictory contemporary or earlier evidence against any writing intended by the parties as a final expression of their agreement, cf also Section 2-202 UCC and section 1.2.6 above. A similar rule also exists in France for non-commercial contracts in excess of 50 francs: see Article 1341 Cc. Any express statement, however, that earlier evidence is not to be used would naturally be upheld under Article 6.

Article 9 deals more in particular with usages in terms of contract interpretation. It has been mentioned before that the Convention remains ambivalent on the issue of custom, much more so than Article 9 of the earlier Hague Sales Convention, as implicitly accepted in its Article 4. Sensitivity to accepted practices and custom was earlier identified as being particularly important in international trade, see chapter 1, section 1.5.2. Much international trade law developed on the basis of it. It is a major cornerstone of the notion and operation of the lex mercatoria. Strictly speaking, the Vienna Convention avoids any reference to custom and uses the terms 'usages' and 'practices' instead (Articles 4, 8 and 9). This is in the civil law tradition of suspicion of [page 416] custom as an independent source of law. Thus, according to Article 9(1), the parties are bound by any usage to which they have agreed and by any practices which they have established between themselves. This is simply an extension of the contract and intent principles at the expense of custom as an own source of law. It implies a subjective approach to custom.

Article 9(2) tries to soften some of the impact by accepting as an implied condition all usages of which the parties knew or ought to have known and which in international trade are widely known or regularly observed (but not merely widely operative) between the parties to contracts of the type involved in the particular trade concerned. The UNIDROIT Principles maintain only the latter requirement (Article 1.8), come as such closer to Article 9(2) of the Hague Sales Convention, and do not therefore require that the usages were known or should have been known by the parties concerned, as long as they are widely known and regularly observed. The UNIDROIT Principles are thus somewhat less restrictive (although they introduce a reasonableness test). Article 1.105(2) of the European Principles may even go a little further in stating that parties are bound by any usage which would be considered generally applicable by persons in the same position as the parties (also subject to a reasonableness test).

Article 9 of the Vienna Convention was meant to protect unsuspecting parties, although the need for it appears less obvious in the professional sphere in which international sales as defined in the Convention usually operate. Yet as already submitted several times before, it is not strictly speaking possible for the Convention to be entirely conclusive in this matter as the force of international usages and practices may derive from other sources or from custom itself. This is in fact implicitly recognised in Article 4(a) which excludes the matter of validity of any usage from the scope of the Convention.

The exclusion of usages in Article 4 has been explained as leaving the determination of this issue (as one not covered by the Convention and therefore not subject to its Article 7(2) on supplementation) to applicable domestic law including the validity of these usages itself. It follows that all usages or custom would be seen as essentially a domestic law phenomenon and that considerations of domestic public policy (or good faith like in the Netherlands) may prevail over them. This would be a strange and undesirable result.

Nevertheless to the extent custom is viewed as a contractual term, as it essentially is in the Vienna Convention, rather than objective law, local public order considerations could all the more be able to override it. Where, however, it is agreed that the force of usage cannot be fully determined by the Convention itself, nor indeed the impact of the lex mercatoria more generally, it is likely that there is not only custom outside Article 9 but also that this custom could be international. This could be so especially where within the context of the lex mercatoria, it impacts on international sales. It would as such only be subject to public policy imperatives operating in the international legal order in which there are few and which, if they exist, are more likely to be expressed in fundamental (mandatory) international legal principle or public order requirement. It may be noted in this respect that in the US under Section 1-103 UCC international custom is in pertinent cases generally accepted as binding in commercial matters. This is not to deny that domestic policy considerations (governmental interests) may also not find recognition in the international legal order but then subject to a balancing test, see more in particular chapter 1, section 2.2.6 ff. [page 417]


As mentioned before, in international sales, the 1980 EU Rome Convention on the Law Applicable to Contractual Obligations must, now the Vienna Convention is becoming widely accepted (although still not in the UK), increasingly be seen as a sequel to it within the EU in areas the law of the sale of goods the Vienna Convention does not cover (Article 4) or as a matter of supplementation or gap-filling even in those aspects that it does cover, but in that case only after the general principles on which the Convention is based have been considered (Article 7(2)). It should be realised of course that the scope of the Rome Convention is much wider than sales and embraces all situations where judges in the EU are confronted with a choice between the contractual obligations laws of different countries, whichever (also if non-EU).

The Rome Convention may also play a preliminary function in the context of the Vienna Convention in that under Article 1(1)(b) of the latter, its applicability may result from choice of law rules pointing to the law of a Contracting State. Under the European Principles, but not the UNIDROIT Principles, it also may play a role in the supplementation of these Principles (Article 1.106).

However, as posited all along and especially in chapter 1, section 1.5.4, it should always be borne in mind that the application of both the Vienna Convention and the Rome Convention must be seen in the context of the application of the lex mercatoria as a whole with its own hierarchy of norms in which fundamental principle, international custom, uniform law, and common legal notions are likely to precede the search for a domestic law through conflicts rules (see for a summary section 2.4 below).

The Rome Convention which dates from 1980, coincidentally from the same year as the Vienna Convention, is now ratified by all older EU Member States but operates for a period of ten years only, although it is renewed tacitly for five year periods thereafter if there has been no denunciation (Article 30). Although often viewed as an EU Convention, this is strictly speaking not the case as it is not based on the EEC founding treaty. Even though concluded between Member States, only a limited number needed to ratify for the Convention to become effective. New EU Members need not accede to the Convention, although in 1980 the view was expressed in a Joint Declaration that they should be encouraged to do so.

This situation is very different from the (unrelated) Brussels Convention of 1968 on Jurisdiction and Enforcement of Judgments in Civil and Commercial Matters which was based on Article 220 of the EEC Treaty and is in force between all Members for an unlimited duration and must be accepted by any new Members (and since 2002 replaced by an EU Regulation to the same effect). The consequence of the Rome Convention not being based on the EEC Treaty is that conflicts rules more directly based thereon prevail (Article 20). So far, they particularly exist in the area of insurance under the Insurance Directives, but there is no conflict here as the Rome Convention does not apply in insurance matters when the risk is in the EU. At one stage, EU private international law rules were also contemplated in the area of employment which could have caused a more acute conflict.[221] [page 418]

Another consequence of the Rome Convention not being an EU Convention is that matters of interpretation are not referable to the European Court of Justice in Luxembourg (in fact even the Brussels Convention only made this possible by a special Protocol of 1971). Although two Protocols to the Rome Convention were signed in 1988 to allow an appeal from the highest courts in each Member State to the European Court in order to obtain the necessary prejudicial decisions in interpretation matters (without a duty as there is in the 1971 Protocol), ratification by all parties seems unlikely as some, like the UK, do not see interpretation of the Convention as a European Court matter. In the meantime steps are being taken to redtftthe Rome Convention in an EU Regulation (a project called Rome I). It would give it a proper EU status.

The Rome Convention according to Article 21 does also not prejudice any other conflicts conventions to which Contracting States were already or may become a party. In the latter case consultation with the other Contracting States is necessary, however (Article 24). The most important examples are likely to be in the Hague Private International Law Conventions like those on the Law Applicable to the International Sale of Goods of 1955 and its successor of 1986 (not to be confused with the Hague Uniform Sales laws of 1964). One could also think of the Hague Conventions on the Law Applicable to Agency of 1978 and on the Law Applicable to Trusts and their Recognition of 1985. However, the Rome Convention itself excludes the coverage of agency and trusts.

The applicability of the Rome Convention is dependent on a forum in a Contracting State being faced with a choice between the laws of different countries which need not be EU States (Article 1), although the Convention does not spell out when this may legitimately be the case. It is left to the forum seised to decide. The Convention may apply even within one country with different legal systems like the UK.

It is important to realise from the way the Rome Convention is structured that it does not apply by virtue of the law of a Member State being made applicable to the underlying contract. For example if English law is made applicable to a contract between two Swiss parties, the conflicts rules of the Convention do not automatically apply (as Article 15, which excludes renvoi, confirms). Its application depends on a case being brought in a Member State. It is even doubtful whether the Convention would apply in an international arbitration with its seat in a Member State as an international arbitration is often thought subject to its own conflicts rules which are not necessarily derived from the place of the tribunal. That would not prevent the application of the rules of the Convention as a model or as general principles of conflicts laws which may be thought to apply in international cases but that would not then be a consequence of the applicability of the Convention itself.

The Rome Convention does not apply in status matters, family matters (relationship and property aspects), negotiable instruments, typical company matters (like creation, capacity, organisation and winding-up), arbitration, agency and trust matters and in matters of evidence and procedure (Article 1(2)). Again, the lex fori seems to determine the true meaning of these exceptions. Proprietary matters are not covered by the Rome Convention either, even if they follow from a contractual disposition as in the case of sales. This would also appear to apply to assignments, although Article 12, especially in its second paragraph, has left some considerable doubt in the matter, see more in particular chapter 4, section 1.9.3. Also Article 10(1)(c) suggests that the consequences of any breach are covered by the law resulting as applicable under the Convention but, in the case of a sale or exchange, that would be unlikely in the proprietary effects of an avoidance of the contract or in any enforcement aspects.

There is a special rule for insurance in that the Rome Convention does apply to reinsurance but not to ordinary insurance policies if the risk covered is located in a [page 419] Member State.[222] If the risk is outside the EU the policy may indicate the applicable law subject always to an appreciation of the foreign mandatory rules (if any) under Article 7 of the Rome Convention, especially in consumer cases when there is the added protection of Article 5.[223]


There are two basic rules in the Rome Convention. First there is the law chosen by the parties, for the whole or part of the contract (Article 3). This choice must be expressed or appear with reasonable certainty from the terms of the contract or the circumstances and can thus not be purely implied as is possible under German conflicts law. The Convention does not limit this contractual choice of law to situations at the disposition of the parties but provides only that if all the elements relevant to the situation at the time of choice are connected with one country, the contractual choice of law may not prejudice the application of the mandatory rules of that country (Article 3(3)). What is to be considered mandatory in this regard remains undefined. There are bound to be considerable differences of view (even within the country concerned) on this point and on the true relevance of such rules in international cases.

The second rule is that in the absence of a contractual choice of law clause, the law of the country with the closest connection to the contract applies. Article 4 allows in this respect different applicable laws for severable parts of the contract. There is a rebuttable presumption that the law with the closest connection is the law of the place of the residence of the party which must perform the most characteristic obligation under the contract (or, if severable, under the part of the contract in question), provided that this characteristic obligation can be determined (Article 4(2) and (5)). The residence in this sense is the habitual one for individuals and the place of the central administration for bodies corporate, except if the contract is entered into in the course of that party's trade or profession, when residence for this purpose is connected to the principal place of business of the party performing the characteristic obligation (Article 4(2)).

Clearly, the reference to the characteristic obligation itself introduces an aspect of considerable uncertainty. Article 4(3) and (4) maintains some special rules in matters of real estate and the carriage of goods, which, like the earlier presumption, is always subject to be disregarded if it appears from the circumstances as a whole that the contract is more closely connected with another country (Article 4(5)). In real estate matters the closest connection is presumed to be with the country where the property is [page 420] located. For the carriage of goods the closest connection is presumed to be with the country of the carrier, at least if this is also the country of the place of loading or discharge or of the principal place of business of the consignor (shipper). No rule is given if none of these combinations obtains and it may well be that in that case Article 4(1) and (2) apply before Article 4(5).

Special rules also apply to consumer contracts (Article 5), which are contracts regarding the supply of goods or products to a person outside his trade or profession, cf. also Article 2(a) of the Vienna Convention and Article 13 of the 1968 EC Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters. The consumer will retain the protection of the mandatory rules of the country of his habitual residence in the case of cold-calling by the (foreign) seller in that country or when the latter induces the consumer to travel to his own for the purpose of making him buy. It means that if a consumer knowingly concludes a contract outside his own country without any special inducement by the seller, he loses the protection of his own mandatory rules. A contractual choice of law is not effective against these principles, nor is the general rule referring to the law of the closest connection (Article 5(2) and (3)).

For individual employment contracts, the law of the country in which the employee habitually carries out his work is believed to be the proper one and a contractual choice of law cannot be effective against the mandatory rules of the country protecting him either (Article 6). It will normally lead to applicability of the law of the place of residence of the employee or alternatively to that of the place where he was engaged if he does not habitually carry out his work in any one country, cf also Articles 5(1) and 17 in fine of the Brussels Convention now superseded by the EU Regulation of 2002, for the assumption of jurisdiction in employment matters.[224]

One may thus see that through contractual choice of law, through the references to the circumstances as a whole in determining the applicable law under Article 4, and through the special protections for consumers and employees, the traditional conflicts approach of hard and fast rules determining the applicability of a legal system regardless of effect and consequence is in a number of aspects amended, although the basic concepts largely remain the same and national law is still assumed to provide an answer to all internationalised contract problems, always subject to the general escape of public policy, provided there is manifest incompatibility with the law resulting under the Convention, cf. Article 16. It introduces a subjective element which is in fact also the case wherever rebuttable presumptions are now used in the Convention. See for the discussion of the broader aspects of the conflicts and transnational approaches more in particular chapter 1, section 2.2.1.

There are a number of specific references to mandatory rules in the Convention: see Articles 3(3), 5(2), 6(3) and 9(6). Article 7(1) contains a special rule on the treatment of a specific type of mandatory rule which could be defined as governmental intervention in the law formation process which cannot be displaced by parties subject to this intervention, even by choosing another law. When application of the law of one country results under the Convention, effect may be given (this is thus discretionary) to this type of mandatory rule of another with which the situation has a close connection in so far as these rules must be applied to the contract under the laws of this latter country. This does not prevent there being varying views in that country itself on the mandatory nature of these laws. The appropriateness of applying them in international situations is here a different matter altogether and may depend also on proportionality and legitimacy of force majeure seen from an international perspective. [page 421]

Article 7(1) only states that regard must be had to the nature and purpose of these rules and to the consequences of their application or non-application. Ratifying states may opt out of this clause (Article 22) as the UK and Germany have done. It nevertheless leaves the problem of what to do with foreign mandatory rules asserted in the case. Note, however, that even under Article 7(1) as now drafted, the interests of other parties and countries are not necessarily considered, nor is the question whether the foreign rule, internationally speaking, might be exorbitant in the light of more general basic principles. The Convention remains here also purely rule-oriented. There is no balancing of different governmental interests when conflicting.

These matters are also not considered under Article 7(2) which allows the forum to apply its own mandatory rules at all times and in all circumstances apparently no matter whether the case in hand has any connection with them. The language of Article 3(3) that all elements of the case must be connected with the country of the forum is not repeated. No special interest of the forum state is apparently required at all for the application of its rules. See for the discussion of Article 7, also chapter 1, section 2.2.6.

As mentioned above, the application of the rules of any other country may always be refused on the basis of public policy (Article 16) if there is manifest incompatibility. In fact, it is rare for courts to give precedence to foreign mandatory rules. This facility was in modern times particularly formulated in the Netherlands.[225] It suggested a bilateralism which even in the Netherlands in practice did not result in much precedence being given to foreign mandatory rules.[226] There is apparently a clear bias for a system not to be unduly disrupted by foreign political imperatives and for mandatory rules of the forum always to prevail. It underlines the importance of the rules of jurisdiction and of any obligatory recognition and enforcement of the ensuing judgment elsewhere, as under the EU Brussels Convention of 1968, now superseded by the Regulation of 2002.

In this connection, the wider problems associated with a contractual choice of law must also be considered. First, in some countries like France there is still a need for some contact between the law chosen and the contract, a concern on the whole more common in the case of a contractual choice of jurisdiction. In any event, the legal aspect concerned must be at the free disposition of the parties, and thus be purely directory.

Indeed, the facility to choose the applicable law is traditionally marred by the operation of mandatory rules, either of the law one attempts to opt out of or of the law one opts into (see chapter 1, sections 2.2.6, 2.2.7 and 2.2.8). Under a contractual choice of law clause, there is bound to remain uncertainty on both accounts, and it is often unclear what the choice of law intends or can achieve in this respect and what its true effect may be, the more so as the definition and intended impact of national mandatory rules may itself be uncertain and vary from situation to situation. Within one country 'ordre public' rules, and modern administrative provisions are normally considered absolutely mandatory, but this is less clear, for example, in property and conveyancing law, status, capacity and other family law matters, company law, bankruptcy and attachments, prescription or procedural laws and concepts. Consequently, it is not always clear which, if any, of these are at the free disposition of the parties and when they may be discarded or opted out of by consent. The opting-in equally presents uncertainties. It is, nevertheless, common to establish the contractual legal framework through a choice of law clause, notably with regard to formation (offer and acceptance) [page 422] and validity (consensus and its defects), all matters that cannot be covered in the contract itself and are therefore in that sense not at the free disposition of the parties either.

This all seems quite normal, yet there remain large areas of doubt. Thus foreigners opting for the application of English law to their contract may not mean to opt for the English rules of consideration or exemption (exoneration) clauses or other instances of investors or consumer protection or generally into any rule deeming the contract itself void or against public policy an health, safety a environmental founds unless there are other contacts with the UK, as application of such rules may never have been in the contemplation of the professional parties as between them. It was already mooted before that in international contracts, the restrictions flowing from the consideration requirement may have lapsed altogether. Under the circumstances, a contractual election of English law may also not include the parole evidence rule. One could argue in this connection that the consideration and parole evidence rules have no mandatory impact as part of a chosen contract law, at least if the case itself has no other major contacts with the law so chosen, see also the discussion in section 1.2.6 above.

In relation to the Vienna Convention, the opting for the law of a Non-Contracting State would imply the exclusion of the Convention even if both parties came from Contracting States. Opting into the law of a Contracting State implies on the other hand the application of the Convention even in respect of parties from non-Contracting States and also of the rules of such States in areas not covered by the Convention, assuming always that such can be considered the intent of the parties and it cannot go further than what makes sense in the circumstances.

A contractual choice of law normally does not have an effect on the proprietary objectives or consequences of a contract. Thus in a sales contract, it is unlikely to determine the law under which the title will transfer. Here the lex situs of the asset acquires mandatory overtones. Yet especially in financial instruments like eurobonds the choice of law by issuers made and reflected on the document is sometimes thought also to govern this aspect, the mode of transfer of ownership, protection of bona fide purchasers, and the types of security interests that can be created in them and the formalities to be observed in this connection, although the final word is here more likely to be with international custom or market practices (see also chapter 1, section 3.2.1).

It is true that the lack of clarity as to the precise whereabouts of these instruments may make the lex situs notion unworkable (as it may be in the case of transient assets like ships and airplanes). The consequence is thus rather the application of transnational concepts instead of those of a chosen domestic legal system. Because of their third party impact, proprietary rights seem to need an objective basis in law, which is unlikely to be provided by the choice of parties to a sales contract or of issuers of securities. Modern registration systems creating security entitlements, may deem the situs for these purposes to be at the place of the entitlements or rather relevant custodian or intermediary (see also chapter 4, 3.2.2).

Although in the absence of a contractual choice of law rule, the Rome Convention is unlikely to apply in the proprietary aspects of international transactions as it only seeks out the law applicable to contractual obligations, especially in the case of assignments (Article 11), there is much different opinion and here again contractual freedom to appoint the relevant property law is often advocated (see chapter 5, section 2.3.6).

The contractual choice of law is also not normally meant to include the private international law rules of the law of the country so chosen, cf also Article 15 of the Rome Convention. International arbitrations have no lex fori per se and are not normally thought to be bound by any particular conflicts rule, no matter what law is chosen by the parties to cover the contract or in which country the panel sits. As mentioned in the [page 423] previous section this also applies to the rules of the Rome Convention which may, however, still serve as a model.

The ultimate danger is of course that the selection of a domestic law may be entirely unsuitable or does not lead to any solution at all, not only in proprietary matters if at variance with the lex situs. In terms of the lex mercatoria, the domestic law so chosen would, however, like any other domestic law resulting from conflicts rules, come low in the hierarchy of norms, see chapter 1, section 1.5.4, so that its possible harm would be limited. Where in international cases the lex mercatoria concept may increasingly be relied upon, at least in international commercial arbitrations, a good case may be made for not making any domestic law applicable at all. It could easily be a confusing and disturbing factor.

Another traditional complication of a contractual choice of law is the treatment of any changes in the law so chosen. If such changes signify the normal progression in that law, they may apply, but if there is a clear deviation of the chosen pattern, the changes in the directory rules of the legal system made applicable might not.


As mentioned before, important derogation from the Vienna Convention (or even its total exclusion) and also from national sales law where applicable (leaving aside the problem of the application of national mandatory rules, for example in currency and money transfer matters or in trade restrictions), may result from contract or established trade practices (Articles 6 and 9). They may also supplement the regime of the Vienna Convention and any other applicable sales law.

Wherever such terms derogate or supplement, it is posited that the Convention and its supplementation provision of Article 7(2) no longer apply, for example under trade terms in the area of the passing of risk (Articles 66ff). There are in particular some established trade terms in this respect. Of these, the Fob (free on board port of shipment or loading), F&S (free alongside), Cif (cost insurance freight port of destination or unloading) and C&F (cost and freight) are the most important.

What they all have in common is that they divide the responsibilities for the handling of the goods and allocate certain costs and risks between seller and buyer, differently according to the means of transportation used and according to the stage of the transfer process. Thus, at one end of the spectrum is the term 'ex works' under which the only responsibility of the seller is to hand over the goods to the buyer or his agent at his own place of business, whilst all risks and expenses connected with the goods and their handling are thereafter for the account of the buyer who must also get himself properly organised. At the other end is the term 'free delivered' under which the seller must make and pay for all arrangements necessary to get the goods to the buyer and has all the risks connected with the goods and their transportation in the meantime.

The Fob (free on board) term dates from the early 18th century and the Cif (cost, insurance, freight) term from the late 18th century, each the product of their own situation: the Fob term was logical when the buyer or his agent sailed with the ship or sent his own and concluded sales contracts on the way and supervised the loading of the goods. Even now it means that the buyer nominates the ship and makes all necessary arrangements for transportation and insurance. The Cif condition on the other hand resulted when regular shipping lines were established and the buyer either depended on the seller to make the necessary transportation arrangement or, more likely, the seller availed himself of the opportunity to ship his own goods before any sale and received a bill of lading which he could negotiate later once a proper buyer was found. This gave [page 424] him a great advantage as he could thus sell his goods Cif, whilst they were already sailing or even upon their arrival, directly in the international markets, thus eliminating his dependence on the visiting buyer. It is now by far the most important trade term.

Even where goods are already sold, the Cif term means that the seller is still in charge of the transportation and insurance arrangements. It is often thought that the Fob term is appropriate both in purely domestic and international sales and does not produce a bill of lading, whilst the Cif term always denotes an export transaction with carriage to an overseas destination and always produces a bill of lading. But there may also be a bill of lading under Fob terms. The difference is that under the Fob term the seller will collect it upon loading as receipt on behalf of the buyer, who is in charge of the transportation arrangements. He must as soon as possible send it to the buyer and is here subject to the latter's instruction. Under the Cif term, the seller collects the bill for himself and is under a duty to tender the documents to the buyer only as part of the sale. This buyer may in any event emerge later. In either case, the handing over of the documents may be affected and delayed as part of the payment arrangements (see more in particular section 2.2.4 above) which is likely also to affect the transfer of title in the underlying goods especially relevant in countries like France and England, where in a sale of goods title normally passes upon the conclusion of the sales agreement. In these countries, use of Fob or Cif terms may itself indicate postponement and make the transfer dependent on the handing over of the bill of lading.

There is a modern variant of the Fob term under which the seller undertakes to arrange the shipment on the buyer's behalf and at his cost. In fact, this variant is so widely used that it is sometimes thought to be the more normal Fob arrangement. Under it, it remains in essence the buyer's duty, however, to nominate the vessel, but the seller may be given full powers to do so as an agent acting for the buyer whilst providing additional services to the buyer in terms of collecting the shipping documents and putting these at the buyer's disposal. As he will not pay for the freight, the bill of lading will reflect this and state 'freight collect' (instead of the more usual 'freight paid' under Cif).

The sales contract will specify these services as it is clear that they will not result from the use of the Fob trade term itself. This type of Fob deal is particularly common in established relationships, where it is often standard practice, but less prudent in incidental arrangements as the seller/debtor remains in possession of the shipping documents and remains therefore in a very strong position vis-à-vis his buyer, especially if a bearer bill of lading is issued which is a negotiable document of title.

A Cif sale is sometimes considered a sale of documents rather than a sale of the underlying goods.[227] Tendering documents is then the essential duty of the seller. If so tendered (together with the insurance policy and invoice), the buyer must pay (if it was agreed to take place upon tendering), regardless of the situation concerning the underlying goods. They may even be lost or may never have been properly loaded. For the payment obligation to mature upon the tender of the bill of lading, the bill of lading itself should be an on-board bill of lading (see also chapter 4, section 2.1.1). It means that the goods must have passed the ship's rail and mere delivery of the goods by the seller at a dock warehouse of the carrier or even at its rail is not sufficient for this type of bill of lading to be issued.

However, even with an on-board bill, the goods if shipped in bulk may not yet have been apportioned to the contract, so that title in them cannot strictly speaking pass despite the existence of a negotiable bill of lading. This may give rise to all kind of complications and actions but it does not excuse the buyer from payment upon the [page 425] tendering of the documents.

In this arrangement, the documents may be rejected upon tendering if inaccurate on their face (especially relevant under letters of credit). This right must be clearly distinguished from the right to reject the goods, which is unconnected and can only be exercised after landing, claiming and examining them when they are found not to be in conformity with the contract. If the bill is used to obtain early payment, for example upon tendering of the documents, any rejection of the goods later may of course still give rise to an adjustment of the sales price and reimbursements but does not affect the original payment and its validity itself.

In this connection, the situation concerning the transfer of title, although not covered by the trade terms themselves, is of interest especially in systems which require delivery for title transfer. The reason for the trade terms not traditionally covering the title and its transfer is the considerable differences in the various laws on when title may pass and the mandatory rules in this respect in many legal systems. In systems passing title upon the mere conclusion of the sales agreement (unless postponed), like the English and French, title passes immediately as we saw. In systems that require delivery for title transfer, like the German and Dutch, the seller must as a minimum put the goods at the disposal of the carrier, see for this system chapter 4, section 2.1.3.

Under a Fob clause, it is common to view the carrier as the agent for the buyer who nominated the vessel and the title passes in systems requiring delivery therefore normally at the ship's rail. If a bill of lading is also issued, title may only pass upon the handing over off the bill, relevant especially if used in the context of a letter of credit. Under a Cif clause, it is also conceivable to view the carrier as an agent for the buyer, even though the buyer strictly speaking does not nominate the vessel and may not yet exist. In this approach, the delivery to the buyer may also be considered to have taken place at the ship's rail, which may complete the title transfer at the same time, provided always that the goods are identifiable (the more likely civil law approach) or have been properly appropriated to the contract (the more likely common law approach). It is, however, more likely that under a Cif sale, title is considered passed only if the documents have been tendered to the buyer and, upon any negotiation of the bill, title will only be considered to have passed further when the bill is handed over (plus endorsement if to order). Intent acquires here a special role.

Indeed in systems in which title transfers upon the mere conclusion of the sales agreement, the bill of lading only has significance if the title transfer is deemed to be postponed until tendering or delivering the bill of lading and this then depends entirely on the intent of the original parties. Under Fob and Cif terms, this intent may, however, be deemed implied, at least if the bill of lading is intended also to play a role in payment protection schemes. Otherwise, especially under Fob terms, the title is usually considered passed when the goods pass the ship's rail, as we saw. This may even be the case in France under the Cif term, although in England the tendering of the documents seems to be the moment (see also chapter 4, section 2.1.4).

Further delay will result, however, if the underlying goods have not yet been identified or set aside to the contract or if title in them has been reserved pending payment, no matter the tendering, delivering or negotiating of the bill of lading. If the goods have not yet been sold, but are simply shipped by the owner with a view to a later sale, the seller/shipper naturally remains the owner and also has the bill of lading. That sale may still be Cif (and this is normal for shipped goods) even though the loading has already taken place (on a ship nominated and paid for by the seller as in an ordinary Cif sale). It is also common to on-sell goods Cif even though in that case shipped by a previous seller who paid for the transportation. [page 426]

As regards the passing of risk, the trade terms tend to be specific on the subject and especially in Fob and Cif terms the passing of the risk always takes place at the moment the goods pass the ship's rail and is thus entirely independent from the transfer of title or of legal possession. Thus the physical act terminating the seller's control of the goods constitutes at the same time the moment the risk in the goods passes: see for the concept of the passing of risk further section 2.1.9 above. If under a Cif contract the sale happens after the goods are already afloat, the passing of risk will be retroactive to the moment of loading.


In Europe, the most important trade terms like the Fob and Cif terms have been compiled and restated by the International Chamber of Commerce since 1936, in the so-called Incoterms, the last edition being of 1 January 2000. They also cover a considerable number of other trade terms like 'ex works', 'free on rail', 'free on truck', 'arrival or ex ship', 'ex quay', 'delivered at frontier' or 'delivered duty paid' (DDP), etc. The Incoterms are sometimes still thought not by themselves to have the force of law. They may, however, have acquired the status of industry custom but not necessarily in all circumstances and detail, as may be seen shortly. At an early stage it was proposed to add them to the Hague Sales Conventions but it was not pursued. In the USA the Fob, Fas, Cif, C&F and 'free delivery, ex ship' terms have been codified under domestic law in Article 2 UCC, but not the others. The trade terms are seldom codified in other domestic laws.

One has to be aware of some differences between the USA and European practices in this field, especially relevant for the Fob term which in the USA allows reference to a destination rather then to a port of loading, a practice now also seen in Europe, for example where pipelines are used, and then implies free delivery to the designated destination point, although the risk may pass sooner.

As already mentioned, in essence these terms all aim at a certain division of labour and costs in terms of physical delivery, transportation and insurance. They also bear on the place of the transfer of risk (normally at the port of loading) and insist on notice (including procurement of the invoice), require the parties to keep each other properly advised on what is happening and, under the Cif terms, demand tender of a transportation document (bill of lading) and insurance policy to the buyer.

It is not uncommon in this connection to refer to 'C' terms, 'F' terms and 'D' terms: in 'C' or 'F' terms, risk passes at loading, in 'D' terms (like DDP) upon unloading, whilst the difference between 'C' and 'F' terms is in the liability for the cost of transport and insurance. Also under the Incoterms, the proprietary consequences remain an area for the applicable national law, if not of some advanced notion of the lex mercatoria, under which one must normally assume that after delivery to the ship's rail, the buyer is owner, the carrier his agent and the possession of the bill of lading the proof. It acquires then the status of transnational paper operating under its own rules, just as negotiable instruments of title largely do.

Whenever specifically referred to in a contract, the Incoterms are automatically incorporated and may then derogate from the Vienna Convention or from national law where otherwise applicable. But it is still necessary in this respect to determine their precise meaning where they lack detail. This may be decided on the basis of general principles or may itself entail references to a national law under the private international law reference of Article 7(2) of the Vienna Convention. Under the Hague Sales Convention, [page 427] Article 9(3), the interpretation of terms was to conform to their usual meaning in the trade concerned. This is now less clear as this particular sub-section was not retained in the Vienna Convention.

This is all the more problematic where the Incoterms have not been made explicitly applicable to the contract. The question has arisen whether in this context the Incoterms, if not expressly made applicable to Fob and Cif references, at least in Europe, may be deemed to apply customarily. The Incoterms themselves (cf their 1990 Foreword) suggest that they must be included in the contract to apply, but this cannot be decisive as to their status as custom or usage. As already discussed in section 2.3.6 above, the Vienna Convention in Article 9 requires that a usage is, or ought to have been, known to the parties and must also be widely known in the trade concerned. Even though this can also not be ultimately decisive as to the true impact of international custom, it may well apply to the Incoterms in the commodity trade in Europe. In any event, one may expect the courts and commercial arbitrators to look to the Incoterms for lack of any better guide when the relevant sales terms are used in a contract without definition or reference to the Incoterms, even if they are not considered custom.

As a minimum, the Incoterms may thus have some explanatory, supporting or pursuasive effect and serve as guides, especially when there is no substantial statutory or case law explaining these trade terms in the jurisdiction identified by the conflicts rules as providing the applicable national law. Especially in the UK such a body of case law exists but that remains exceptional.[228] Even then, it is likely that the (English) courts will have regard to the need for uniformity in the interpretation of these terms and domestic precedent should not determine all issues, which in the UK is also the approach taken to the interpretation of the Hague (Visby) Rules.[229]

Where the law of a state of the USA would be applicable under the pertinent conflicts rules, the UCC would be applied instead to the extent that it covers the relevant terms. Again the interpretation elsewhere cannot even then be fully ignored and international practice remains relevant, all the more so for the terms not defined in the UCC, in which connection the Incoterms may be of prime importance even in the USA.

It finally leaves the question whether conflicts may arise between the meaning of the established trade terms, whether or not supported by the Incoterms, and the Vienna Convention, in which case the trade terms prevail, assuming their meaning is clear. In this connection the Incoterms figure either if explicitly made applicable or as custom. Yet conflict will be rare. As a start, the need to deliver the goods to the carrier under both Fob and Cif terms is fully compatible with the provisions of Articles 30 and 31 of the Vienna Convention. Article 34 requires documents to be tendered at the time and place and in the form as required by the contract.

The trade terms will be specific in these aspects and then supplement this provision. As regards the passing of risk, the Vienna Convention (Article 67(1)) again accepts the established trade term practice of the risk passing when the goods are handed over to the carrier (except in 'D' terms). Where the trade terms (indirectly) bear on or refer to other aspects of substantive law, they should prevail as contractual terms but will be amplified by the Vienna Convention, unless clearly meant to operate otherwise. An example may exist in Article 67(2) requiring for the passing of risk the clear identification of the goods, which may not strictly be necessary under the trade terms which would then prevail. There may also be some doubts: for example the Fob terms do not involve a carriage of goods proper so that the references to the handing-over point and the passing of risk in Articles 31 and 67 of the Convention may not, strictly [page 428] speaking, apply. Even so, the Incoterms themselves clarify these aspects as if there was a carriage of goods.

Earlier, the packaging duty under Fob terms to which Article 35(1) of the Vienna Convention makes a reference, existed probably only until the place of loading was reached, in derogation from Article 35(2)(d), but a further duty now appears implied under the Fob terms (as restated at the time in the 1990 version, Comment 9).


As was pointed out many times before, the Vienna Convention is only a partial codification of the international sales law. As we saw, it has its own rules of supplementation in the areas it covers in principle. They are all best explained in the context of the lex mercatoria with its hierarchy of norms, see sections 2.3.5 and 2.3.6 above. Also the UNIDROIT and European Principles may supplement them within that context.

Within the ICC, an effort has further been made to produce a Model International Sale Contract which is divided into two parts: (a) specific conditions setting out terms that are special to particular contracts of sale and (b) general conditions setting out standard terms common to all contracts incorporating the ICC General Conditions of sale. Both the special and general conditions concern manufactured goods intended for resale when the purchaser is not a consumer and the contract is an independent transaction rather than part of a long-term supply arrangement. The model contract is subject to the Vienna Convention (if not excluded) but its general conditions serve in practice as an amplification of rules of the Vienna Convention. The specific conditions on the other hand contain an easy checklist of the basic deal terms. The trade terms used are as defined in the Incoterms (except as otherwise agreed).

There are also Principles of European Law: Sales, assembled at the University of Utrecht as part of the larger European Civil Law project aiming at Restatements in various areas (see also chapter 1, section 1.5.6).


214. The International Sales of Goods Convention had been Unidroit's major project. After it was re-established it diversified into other areas: see more particularly chapter 1, s. 1.5.6.

215. See the Law Society of England and Wales, Law Reform Committee of the Council, 1980, Convention on Contracts for the International Sale of Goods (1981). The Department of Trade and Industry, United Nations Convention on Contracts for the International Sale of Goods: A Consultative Document (June 1989) was more neutral and did not express a clear preference.

216. It is clear that the Vienna Convention does not adhere to the principle of a price certain for the validity of the contract, as Roman law had required (the pretium certum, also of importance for the transfer of risk) and as French law in its wake still does, see Article 1591Cc, but the Cour de Cass. in a decision of 1 Dec. 1995 [1996] JCP 22.565, allowed the price to be unilaterally established if subject to objective criteria, like for telephone companies a generally applicable telephone charging scale. Article 1591Cc is now merely seen as a protection against abuse by the seller. German and new Dutch law do not retain the idea of a fixed price and, like the Vienna Convention, allow the determination of the price later with reference to objective, often market-related standards.

217. These principles must be distilled from the underlying approach of the Convention in its material provisions, which is not easy. Honnold, above n. 183, 129ff. sees a general principle of reliance (on conduct), e.g. on oral representations made after the contract was concluded (Articles 16(2)(b), 29(2) and 47), disclosure (Articles 19(2), 21(2), 26, 39(1), 48(2), 65, 68, 71(3), 72(2), 79(4) and 88), and of mitigation of damages (Articles 77, 85 and 86). He suggest nevertheless caution and restraint in invoking these general principles now that the reference to private international law has been added and that matters of pure interpretation have been separated out in Article 7(1). Case law so far collected by UNCITRAL and regularly published in their Case Law on UNCITRAL Texts has not shed a great deal of light on these general principles either. Another issue is what the matters governed by the Convention are. Although in general the excluded areas are clear from Article 4, and the coverage as a partial codification is quite limited, it must be assumed that in cases of amplification of the Convention, e.g. in the area of the passing of risk, underlying principles of liability must first be distilled and expanded to cover any gap pursuant to Article 7(2) before a domestic law can be invoked.

218. See for the law merchant impact more fully section 2.4 below.

219. The reference to good faith interpretation of international conventions is somewhat of a strange compromise but may well have its origin in the Vienna Convention on the Law of Treaties of 1969, Article 31, but it concerns here the regime of agreements between states, in which connection the reference to good faith seems much more traditional.

220. See for the English literal and purposive approaches to statutory interpretation, chapter 1, s 1.2.9.

221. Others have appeared in the 1998 Settlement Finality Directive (Article 9(2)), 98/26/EC [1998] OJ L166 and the 2002 Financial Collateral Directive (2002/47/EC, OJ L168) with respect to security interests in securities that are recorded in a book entry system, and in the 2000 Directive combating late payment in commercial transactions (Article 4, 2000/35/EC, OJ L200) with respect to reservations of title. All are, however, in the proprietary area and therefore out of the scope of the Rome Convention proper.

222. Each judge must use his own internal law to determine where the risk is located: Article 1(3). The Second EU Non-Life Directive of 1988 (Article 7) and the Second Life Directive of 1990 (Article 4), maintain distinct systems when the risk is within the EU. They are inspired by respect for the mandatory contract rules of the Member States where the risk is situated for (mass) non-life policies and of the Member States of the commitment for life policies. These rules are explicitly not harmonised so as to protect the beneficiaries of mass non-life and life policies under their own law. The importance of the conflicts rules of the Insurance Directives is that the insurance policy may set aside these rules only if the applicable law allows it, which for this purpose is itself tied to the habitual residence of the beneficiary. If there is no chosen law in the policy, it is this law which is in any event presumed to apply. It means that only insurance products allowed in the country of the beneficiary may be sold into that country and are to be structured in accordance with that law. From this point of view there is no free circulation of insurance products even though the rendering of insurance services has been liberated in the EU.

223. If there is no such choice of law, the Rome Convention relies, under its general rule of the characteristic obligations, on the law of the place of the insurer rather than of the place of the risk which is the more traditional approach. There is no rule for a situation in which the policy covers risks in and outside the EU at the same time, e.g. in the case of a fire insurance in respect of premises of a single party insured in the USA and the UK.

224. See for a comment also CGJ Morse, 'Consumer Contracts, Employment Contracts and the Rome Convention' (1992) 41 ICLQ 1.

225. See the Alnati case, HR, 13 May 1966 [1967] NJ 3 in which it was said that 'in respect of a contract…it is possible that a foreign state has such an interest in the application of its own mandatory rules…that Dutch courts must consider this interest and may have to give precedence to such rules and ignore the law chosen by the parties in their contract'.

226.See also Arnhold Karberg & Co. v Blythe, Green, Jourdain & Co [1915] 2 KB 379, 388.

227. Arnhold Karberg & Co v Blythe, Green, Jourdain & Co [1915] 2 KB 379, 388.

228. See C Schmitthoff, The Law and Practice of International Trade 10th edn (2000), 7ff.

229. See Stag Line v Foscolo Mango & Co Ltd [1932] AC 328.

Pace Law School Institute of International Commercial Law - Last updated November 7, 2008
Go to Database Directory || Go to Bibliography