INTRODUCTION

This work seeks to examine the mechanism of limiting damages under the Vienna Convention on Contracts for the International Sale of Goods (CISG). The importance of the issue of limiting damages can be realized when viewed in the light of a role that damages play in a general framework of the CISG. In particular, it has been said that "[n]o aspect of a system of contract law is more revealing of its underlying assumptions than is the law that prescribes the relief available for breach."[1] Further, the right to damages plays the central role in the CISG remedial scheme.[2] Finally, it is to be borne in mind that the CISG regulates international sales contracts, and international trade is one of the main forms of international economic relations. Therefore, certainty and predictability in an exceptionally important field of the Convention's legal regime are necessary. Thus, clarity in the regulation of this issue and the extent, to which its theoretical foundation has been elaborated, are the essential elements in determining the Convention's "underlying assumptions" and in effecting the trends of legal regulation of international sales transactions.

The purpose of using the methods of limiting damages is to restrict the liability in damages. This purpose makes the issue of limiting damages an integral part of general legal regulation of damages. Therefore, it is difficult to achieve certainty and predictability in the rules on damages unless the theoretical basis for regulation of mechanism of limiting damages is developed. At the present moment, it seems that there is room for further development. In this regard, the present work aims to examine the relevant provisions of the CISG and to emphasise some of the issues which require further elaboration.[3]

Before embarking directly on the methods of limiting damages, the present author will examine such issues as the interests protected and, accordingly, the categories of loss covered by the Convention. The reason for these matters being given certain attention is that an examination of the methods of limiting damages will not be full without taking them into consideration.

Further, in some parts of the work, while the relevant provisions and problems related to them will be examined, some attention will be paid to the treatment of analogous issues in some legal systems where similar principles have been extensively developed, as well as in the UNIDROIT Principles. Certainly, bearing in mind the international character of the Convention,[4] the examination of the CISG will not be based on the approaches of domestic legal systems. It is merely thought that comparison is, probably, one of the most efficient ways to underline some of the unique features inherent in some legal regimes (especially in such a document as the CISG because of its self-standing position) and to develop solutions to existing theoretical problems. As has been said in the context of comparative law, the "different systems of the world can offer a greater variety of solutions than could be thought up in a lifetime by even the most imaginative jurist who has corralled in his own system".[5]

I. INTERESTS PROTECTED AND CATEGORIES OF LOSS

1. Expectation and Reliance Interests

The central principle relating to the measurement of damages which is common to many legal systems, is that of the interest which an aggrieved party has in the performance of the contract. It is generally considered that a party has the right to be placed in the same economic position he would have been in had the contract been properly performed.[6] This interest is usually referred to as "expectation interest". It has been said that "expectation measure is … the natural measure of recovery, since it accords directly with the underlying morality of promise keeping".[7] A party's expectation interest will generally represent the actual worth of the contract to that party.[8] In principle, perfect expectation interest will leave an injured party indifferent between performance and breach.[9] In a global context, some believe that realisation of the expectation interest will "stimulate economic activity, facilitate reliance on business agreements and protect the 'credit system'."[10]

The expectation interest is not the only interest that may be protected by an award of damages. Sometimes, the so-called "reliance interest" is protected as well.[11] The core of the reliance interest is to put the aggrieved party into the situation in which he would have been had the contract never been performed.[12] This is usually done by compensating such a party for the losses which he incurred in reliance on the contract.[13] The idea behind this principle is that if the contract has not been duly performed, the aggrieved party may seek to recover those expenses which he incurred having acted in reliance on the contract, as these expenses would otherwise be wasted.[14]

How are these interests reflected in the CISG? It has already been mentioned that "the basic philosophy of the action for damages is to put the injured party in the same economic position he would have been in if the contract had been performed.[15] Following this philosophy, the damages provisions of the CISG are aimed at protecting the injured party's expectation interest.[16] The wording of Article 74 - "loss, including loss of profit" - makes that clear.[17] Consequently, it is logical that the reliance interest is also covered by the CISG, because recovery of the expectation loss should, as a rule, include reliance loss.[18]

With respect to the correlation between these two kinds of interests, the point has been raised as to whether there is an obligation to elect between the recovery of reliance and expectation damages.[19] This concern is not groundless. In some legal systems, only strictly defined situations will give rise to a right to claim damages for both reliance and expectation losses.[20] However, in relation to the CISG, there is no obligation to elect between these interests. The idea, underlying Article 74, is to compensate the injured party fully for the loss suffered as a consequence of the breach.[21] In other words, all kinds of loss suffered by the party should, in principle, be recoverable without the necessity of election.[22] Moreover, the text of the CISG does not prescribe any concrete formula or contain any requirement, in this respect, analogous to those, which can be found in the legal systems, referred to above. The types of loss, which may be recovered, will depend on the circumstances of a particular case. It is suggested that it was neither intended to establish such an obligation, nor is it expedient to elaborate such a scheme.

On the other hand, it is to be borne in mind that overcompensation should not be allowed. In other words, the recovery of damages should not result in a profit to the innocent party.[23] What if the breach of the contract brings certain advantages to the injured party? The UNIDROIT Principles, for instance, directly provide for such a situation. According to Article 7.4.2, in determination of harm account should be taken of "any gain to the aggrieved party resulting from its avoidance of cost or harm."[24] It seems that the solution should be the same under the CISG. It has been said that this rule was implicit in Article 74:

"[T]he party entitled to damages does not suffer a "loss" to the extent that the breach of contract also confers advantages on him which absorb the detriment suffered."[25]

A further issue, which needs to be addressed, is that of the inter-relation between the concepts of expectation interest and full compensation for harm. These two concepts are widely used together in discussion of the principles, underlying Article 74. For example, it has been said that the rule in Article 74 "expresses the principle of full compensation" or, in other words, "the promisee has a right to be fully compensated for all disadvantages he suffers as a result of the promisor's breach of contract".[26] It has been further stated that "[t]hose disadvantages are established by comparing the situation in which the promisee finds himself as a result of the breach of contract with the situation in which he would have found himself if the contract had been correctly performed"(expectation interest).[27] In this regard, it is to be said that the principle of full compensation is the basis for recovery of damages for loss in many Civil law systems. This formula comprises actual loss (damnum emergens) and loss of profit (lucrum cessans).[28] It has been stated that "only owing to this principle will the full protection of the interests of those, who suffer losses … be provided."[29]

Since the two concepts are used in relation to the same subject-matter, it is necessary to clarify the situation in this respect. Three different views can be taken with respect to the inter-relationship of two concepts. First, prima facie, it seems that these concepts mean "the same thing", i.e., they are both used to describe the principle, underlying Article 74. Secondly, it can also be said that the concept of "full compensation for harm", which is reflected in the CISG in the same way as it is established in some Civil law systems, covers all possible kinds of loss. As to the expectation loss, it can be argued that it does not cover all types of loss. In particular, this work will show that the CISG covers, among other types of loss, the losses, which in some common law systems are referred to as "consequential" and "incidental" losses. It has been stated that these two kinds of loss do not form a part of a party's expectation.[30] Thirdly, it can be said that, in essence, these concepts bear the same idea, but the difference is that they "play different roles": expectation interest represents the ultimate goal of the damages claim, and the principle of full compensation for harm is the means or mechanism of achieving that goal or satisfying that interest. It seems that the last view is the most acceptable one.

2. Categories of Loss

Article 74 provides for compensation for "loss, including loss of profit, suffered as a consequence of the breach."[31] Following the logic of this provision, it can be concluded that loss should be divided into two main categories: actual[32] or effective[33] loss and loss of profit.[34] Besides this broad division, Article 74 does not define what concrete types of loss can be compensated. It seems that the principle of full compensation for harm, in the light of the particular contract and circumstances, should be the basis for determining the loss.[35] This principle, in turn, will lead us to the conclusion that all kinds of loss, suffered by the party and caused by the breach, are recoverable.[36]

However, some commentators have gone further. They have worked out the classification in order to identify concrete forms or types of loss besides the main categories of loss mentioned above. In particular, it has been suggested that, in the context of actual loss, "[l]osses caused by breach of contract may take the form of loss by the non-performance as such, incidental loss, or other losses consequent upon the breach [consequential loss]."[37] This commentator has defined each of these "forms" of loss[38] and examined the way these losses should be treated under the CISG.[39]

The present author appreciates this attempt to identify the losses, recoverable under the CISG, and, thereby, to clarify the situation in this respect. However, in this work, this classification will not be relied upon. It is suggested that there is no objective need to try to "embrace", through such a classification, a wide diversity of different "forms" of loss which can arise in practice. Article 74 has been formulated in such a way so as to cover any situation which causes any type or "form" of loss, provided that its requirements are met. Therefore, it seems that, in terms of practical application of this provision, there is no need to "view" a situation through the suggested classification, since the formula in Article 74 will suffice. Nor does the classification seem to be needed for development of the "theory" of the CISG. For the sake of illustration, let us refer to a legal system, where the issue of types of loss is treated in an analogous way.

Article 14 of the Civil Code of the Republic of Uzbekistan provides that damages consist of "expenses, which a person, whose right has been infringed, has incurred or will have to incur in order to redress the infringed right, loss or damage to its property (real loss) [as well as] profits, which have not been received and which would have been received … if his right had not been infringed."[40] Even though this provision has been elaborated in a bit greater detail, it is analogous to the approach taken by Article 74 of the CISG. As in the case with the CISG, this provision is based on the principle of full compensation for harm and serves the purpose of restoring the position, in which the party would have been had the obligation been properly performed.[41] With respect to this provision, it could be equally stated that, in a legal sense, real loss should be constituted by a number of different "forms" of loss. However, the concept of real loss, which have been used since its establishment in Roman Law,[42] has not been treated as being constituted, in a legal sense, of "forms" of loss. So far, this treatment of real loss, in the theory of Civil law, has been rather acceptable. Since the formula used in the CISG, is similar to that provided in the Civil Code of Uzbekistan and some other Civil law systems, the experience of the treatment of this issue in theory of Civil Law should not be disregarded. The classification can serve as a useful example reflecting certain types or "forms" of loss, which can take place in practice. However, in the framework of the CISG, the suggested "forms" of loss should not be treated as legal categories.

Further, it is to be noted that terminology, used to identify the "forms" of loss can lead to some confusion since that terminology is widely used in some legal systems.[43] Moreover, some of the terms are used in a number of different meanings.[44] Taking into account the "international character" of the Convention,[45] this type of terminology should be avoided as much as possible.

Thus, the present author will adhere to the model, which follows directly from Article 74, i.e., actual loss and loss of profit.

(a) Actual Loss

Actual loss can be defined as any reduction in the assets of an injured party as they existed when the contract was concluded,[46] or an increase in his liabilities which, for example, "occurs when an obligee, not having been paid by its obligor, must borrow money to meet its commitments".[47] As mentioned above, there can be a great variety of forms, in which actual loss can manifest. Therefore, it does not seem possible to list all forms that actual loss may take.[48] Nonetheless, so long as the necessary requirements have been met,[49] the compensation for actual loss should be awarded under Article 74.

(b) Loss of Profit

The second category of loss to which Article 74 specifically refers is the loss of profit.[50] The reason that loss of profit is considered to be in a different category is that its nature is substantially different from that of actual loss. In particular, while actual loss generally means the diminution in the assets of an injured party at the time of the conclusion of the contract, loss of profit means the loss of any increase in the assets caused by the breach.[51] In other words, if the contract had been properly performed and the breach had not been committed, the injured party would have enjoyed an increase in his assets. It has been said that both categories of loss (actual loss and profit loss) are regulated in the same way in the CISG.[52] However, one essential difference between the two, reflected in the Convention's scheme, should be pointed out. This difference relates to the effect of Article 44.[53] Namely, this Article allows the disregard of the provisions of Article 39 and 43,[54] if the buyer has a reasonable excuse for the failure to give the required notice. In the latter case, the buyer may reduce the price or claim damages. However, in his claim for damages, the loss of profit cannot be demanded.[55]

(c) "Lost Volume" Situation

This part of the work will examine the problem, which has been called a "lost volume" situation.[56] This doctrine has received extensive development in American judicial practice and academic writings. The type of loss, implied by this term, is undoubtedly likely to arise in international sales transactions. Therefore, an understanding of the essence of this concept and its place (if any) within the Convention's legal regime is extremely important.

The concept of "lost volume" has been a subject of an enormous debate. Nevertheless, it can be defined. Lost volume is the type of loss, which can only be sustained by a party who acts as a seller. In a lost volume situation, a seller has fewer customers than he can supply.[57] After one of his buyers commits a breach of contract by repudiating the contract, not accepting or rejecting the goods, the seller successfully resells the contract goods to a different buyer. However, the second sale cannot be a replacement for the original contract. The reason is that the second buyer would have purchased these goods from the seller, even if the original contract had been performed. Therefore, the seller has "lost volume". His "total number of sales [has been] reduced by the quantity represented by the original contract".[58] The second transaction cannot make the seller whole. This transaction would have been made in any event. Yet, the breach has prevented the seller from earning one more profit from the original contract.[59]

Let us illustrate this type of situation by an example. Suppose that S has several identical tables he wants to sell. B1 contracts to buy one table. Shortly after the conclusion of the contract, B1 repudiates. S resells the table to B2. S claims that this resale did not put him into the position he would have been in had the contract with B1 been performed. If B1 had not breached, S would have also sold a table to B2, and earned two units of profit, instead of one.

The lost volume concept has been strongly criticised on conceptual and economic grounds. Although the present author will not engage in a detailed discussion of the problem, it is necessary to point out the main grounds for the criticism of the doctrine.

First, it has been said that the lost volume concept is not in line with the seller's duty to mitigate.[60] This issue will be dealt with later on in this work.[61]

Secondly, it has been argued that the seller's expectation interest in the profit from the second sale is unprotected.[62] According to this view, the seller's expectation should be evaluated only at the time of entering into a valid contract. "A valid contract is the basis for legal protection of a party's expectation. Accordingly, at the time of the conclusion of the contract with the breaching buyer, the seller's expectation is legally protected only under this contract. Although the law protects the seller's expected benefit under a valid contract for the sale of goods, it does not protect the seller's expectation as to what his market will be like following the contract[63]… The seller's expectation that he will resell the goods does not arise until after the original buyer has repudiated the deal. The seller does not have this expectation until after he has entered into the contract with the second buyer. In awarding the profit remedy to volume sellers, courts retroactively apply this expectation to the formation of the original contract, the only time at which expectations are relevant with respect to the contract goods. In other words, the seller's expectation of an additional sale is actually a post hoc expectation, which is an oxymoron."[64] It is argued that this point does not withstand close analysis:

"The lost volume seller's expectation in the second sale is unprotected until the seller enters into the second contract. At that time, the seller has a protectable interest in the original contract, and a separate and distinct protectable expectation in the second sale … [T]he lost volume seller is not claiming a protectable interest in two transactions at the time of the original sale. Conversely, the lost volume seller claims an expectation in two contracts entered into at different intervals. Thus, the argument over the lost volume seller is not whether the seller has a protectable expectation in the post-contractual market, but whether, after the second sale is consummated, the expectation on the second sale should be used to reduce the expectation interest on the original sale."[65]

The third critical argument is that the award of damages, flowing from the lost volume, overcompensates the seller. It puts the seller into a better position than the one he would have been in, had the original contract been performed.[66] This view naturally flows from the second argument, that is, that the seller does not have a protectable expectation in being in such a position.[67] Since the second argument has not been supported, the "overcompensation" point cannot be accepted as well.

As regards the Convention's position, it is to be said that it does not explicitly address this issue. However, it is suggested that, in the framework of the CISG, loss of volume should be recognised as a valid legal concept. As has been shown above, the Convention's damages remedy is also based on the concept of expectation interest. In lost volume situations, unless an injured party is properly compensated for this type of loss, he will not be put into the position, in which he would have been, if the contract had been performed. Procedures for measuring damages, stipulated in Articles 75 (so-called "concrete" calculation) [68] and 76 ("abstract" calculation),[69] will not be able to restore a party's expectation interest. For instance, if, in our example above, after avoidance of the contract, S "within a reasonable time" and for the same price resells the table to B2, there will be no difference between the contract price and the price in the second transaction. The seller can recover no damages under Article 75, although, in fact, he suffered the loss of an additional profit he would have received, had there been no breach. If, however, S resells the table at a price lower than the contract price, then he will only be able to recover the difference between the contract price and the price in the second transaction. Again, this difference will not compensate him for the additional profit that he would have received had he sold one more table. Analogous results will follow if abstract calculation under Article 76 is applied.

However, both provisions provide that an injured party may recover "any further damages recoverable under Article 74.[70] It seems that this scheme can adequately cover the lost volume claim.[71]

At least in one case, decided in Austria, the court awarded damages for loss of volume. In that case, the seller agreed to manufacture and sell jewellery to the buyer. The buyer did not pay and delivery was not made. On the facts of the case, it was stated that damage, suffered by the seller, would arise "regardless of a possible resale of the goods ordered to a subsequent buyer, as the later contract would have been formed independently of the [buyer's] order."[72]

In another case, decided in the US, the seller delivered defective compressors, which the buyer intended to use in its manufacture of air conditioners.[73] In the findings of fact, it has been stated that the buyer was unable to obtain substitute compressors from other sources and, therefore, suffered loss in the volume of air conditioners that it was able to manufacture for the selling season. The US District court for the Northern District of New York found that the CISG permitted "recovery of lost profit resulting from a diminished volume of sales". The US Court of Appeals for the Second Circuit seems to have upheld this decision.[74] However, though the loss, in this case, was called "loss in volume", it did not represent the type of loss being discussed in this part of the work. First, the injured party acted as a buyer under the original contract with the breaching party. As has been mentioned above, loss of volume (in the sense it is meant here) can only pertain to a seller. Secondly, in a "classic" lost volume situation, the seller has sufficient supply and insufficient demand. In the present case, the breach diminished the party's capacity to supply, which led to loss of a certain number of customers. Thus, the party did not have sufficient capacity to supply its potential customers. As will be seen later on in this part of the work, the seller cannot be a lost volume seller unless he can prove that he had actual capacity to supply his customers. Moreover, in this case, an injured party seems to have had sufficient demand for its goods. It lost his potential customers as a result of an insufficient capacity. Thus, although the court used the term "loss in the volume", this case does not reflect the situation under consideration.

Although it is argued that loss of volume should be covered by the CISG, an elaboration of strict standards in relation to a lost volume seller is crucial. If there are no such standards, a seller may find himself overcompensated. Fortunately, American legal academics have developed general criteria which a seller must meet in order to qualify as a lost volume seller. Professor Harris has developed three main requirements that a lost volume seller must meet: "(1) the person who bought the resold entity would have been solicited by the plaintiff had there been no breach or resale; (2) the solicitation would have been successful; and (3) the plaintiff could have performed that additional contract".[75] Most American courts and commentators have adopted these requirements.[76] However, another formulation of this test will be relied upon in this work.

In order to identify a lost volume seller two questions must be answered in affirmative: (1) Could the seller have supplied both the original and the resale buyer? (2) Would the seller have sold the goods to the resale buyer even if there had been no repudiation by the original buyer?[77]

The first question focuses on a seller's capacity to supply both buyers. This element is crucial for establishing a lost volume case. Imagine that a small seller produces the goods to the limit of his capacity. If buyer 1 breaches the contract, this seller resells the goods to buyer 2. Had buyer 1 not breached, the seller would not have been able to sell the goods to buyer 2 due to his limited capacity. The seller has not, in fact, suffered any loss of volume. The sale to buyer 2 should be merely considered as a mitigation measure.[78]

Deciding the issue of whether a seller in question was a lost volume seller, the courts will need to rely not on a theoretical ability to supply, but on a practical ability to supply both buyers, basing upon the circumstances of a particular case.[79] It is necessary for a seller to prove that he had, for example, excessive capacity to manufacture the goods or ready access to additional inventory.[80] If he cannot do that, then if he wants to qualify as a lost volume seller, he will need to present the evidence that "he would, in fact, have expanded his manufacturing operations, or sought and found replacement stock outside his regular supply channels."[81]

The second question is whether the seller would have made a second transaction even if there had been no breach by the first buyer. Put in a different way, the question is whether the first sale and resale after the breach are "wholly independent events."[82] If an answer to this question is "no", a seller cannot be regarded as having suffered loss of volume. In order to help us determine whether the second sale would have been made, had there been no breach, several guidelines have been put forward.

First, the fact that the seller has made some special efforts to carry out the second sale is said to serve as an indication that, absent the breach, this sale would not have been made. "Such special efforts might include advertising…or highlighting the breached item on the showroom floor. Any of these actions would lead to indicate that the seller would not have solicited the ultimate resale purchaser except for the breach, therefore, any lost volume is illusory."[83]

Secondly, the needs of a particular resale buyer are to be taken into consideration.[84] Thirdly, the characteristics of the goods under the original contract must be considered. The general idea is that the more specific the goods are, the more likely the resale is not a lost volume sale. Such as resale can be the result of either the seller's special efforts or of the particular needs of the second buyer.[85]

Both requirements -- the "capacity" and "wholly independent events" tests -- must be met. Meeting only one of these requirements is not sufficient to establish a lost volume case. This essential rule in analysing a potential lost volume situation has been repeatedly ignored by American courts. Instead of having applied both requirements (or three requirements in Harris' test), they deemed it sufficient to establish only a "capacity" element.[86] Such a treatment of a potential lost volume case should not be allowed. It is most likely to result in overcompensation. Only provided that both requirements are met, can the seller qualify as a lost volume seller.

The guidelines above are, principally, based on writings of American legal scholars and practice of American courts. But it is suggested that they can be applied to the CISG. As shown above, application of Articles 75 and 76 will not lead to fair results. Article 74 should be the basis for recovery of lost volume. However, it does not explicitly provide for such a situation. Therefore, the proper guidelines are necessary for the courts and arbitrators to be able to orient themselves in such cases. The rules suggested by this work seem to lead to sensible and fair results. The fact, in itself, that these rules are based on the American legal practice and were developed by American lawyers should not impair the "international character" of the Convention. On the contrary, experience of American courts reflects different legal aspects and problems of modern commercial activity. For an "international lawyer" this experience should help deal with analogous problems in the "international context". Usage of such an experience should be careful. The regulation of international transactions governed by the CISG should not be based on the legal concepts and principles used in a particular legal system. Bearing that in mind, the present author believes that, although lost volume is a concept used within the American legal system, it represents a problem which is not "alien" to international transactions. We should not "view" such situations through the prism of American legal principles and rely upon purely domestic sources of law. Rather, we should search for sensible solutions, offered by that legal system without impairing the international character of the Convention and contributing to achieving uniformity in its application. In a lost volume case, the American legal system seems to offer extremely helpful guidelines.

(d) The Problem of Non-Material Loss

How should a situation where non-material loss was caused by the breach be dealt with, if the CISG is applicable? In order to answer this question, it is necessary to define the term "non-material loss", determine the forms it can have and correlate this concept with the nature of legal relationships governed by the CISG.

Non-material loss can be defined as loss, flowing from an injury or damage to non-material values. Non-material values, in turn, are such values that do not have "economic content" and are inseparable from the personality of a bearer of these values.[87] Namely, non-material values can include the following: life, health, dignity, honour, reputation, etc. Accordingly, non-material loss is loss or harm, flowing from injury to health, physical or moral suffering, damage to honour and reputation, etc.[88]

Generally, CISG does not cover non-material loss. First, the Convention is mostly oriented towards the relationships of commercial character. Commercial relations are, generally, aimed at achieving material or pecuniary purposes. These purposes do not seem to involve non-material categories, mentioned above. This aspect can lead us to a general conclusion that, in a commercial setting, non-material loss is not likely to arise and should not be claimed.[89]

Secondly, most commercial players are legal entities (corporate bodies), and the question arises as to whether legal persons can sustain non-material loss. It seems that, generally, a legal person should not be regarded as being capable of suffering this type of loss. For example, the National and International Arbitral Tribunal of Milan in its Arbitral Award No A-1795/51 (01 December 1996), applying the UNIDROIT Principles, excluded compensation for emotional harm and distress, because the injured party was the corporate entity.[90]

Finally, it should be noted that in one case, decided by the Tribunal of International Commercial Arbitration at the Russian Federation Chamber of Commerce and Industry, the plaintiff was denied compensation for "moral harm."[91] One of the grounds for denial was that the Tribunal considered that the CISG did not contain the provisions as to the compensation for "moral harm" in a situation analogous to the case under consideration.[92]

However, it seems that there may be at least two kinds of situations in which this type of loss may be compensated. The first situation is the one where the purpose of the transaction is entirely non-material, and the parties are aware of such a purpose.[93] Accordingly, the loss, caused by the breach, which totally or substantially undermines the whole (non-material) purpose of the transaction, should be recoverable. However, in a context of international commerce, the situations of this kind seem to be quite non-typical.

The second situation is where an injured party's business reputation was negatively affected as a result of the breach. In commerce in general and in international sales, in particular, business reputation plays an important role. It can affect and sometimes pre-determine the state of affairs of a subject of commercial activity.[94] Thus, this part of the work will examine why and how loss of or injury to reputation should be governed by the CISG.

The issue of injury to reputation needs to be approached carefully. In order to understand the legal implications of this form of loss it is helpful to consider the treatment of this matter in English law.

English cases have established a distinction between an injury to reputation as being non-material (non-pecuniary) loss and pecuniary loss, flowing from such an injury. While loss of reputation in itself cannot be recovered,[95] pecuniary loss caused by loss of reputation has been held recoverable in several cases.[96] Should loss of (injury to) reputation in itself be separated from pecuniary loss flowing from it? It is argued that business reputation should be regarded as a separate legal category. It can be defined as an opinion of "business actors" on another subject of commercial activity, which has been formed on the basis of its professional qualities.[97] Although, as we can see from this definition, its nature is wholly non-material, it represents certain value in itself. One can, of course, argue that the ultimate purpose of good reputation in business is to gain profit. Consequently, the loss of (injury to) reputation should have legal significance only when it leads to loss of profit.[98] However, it seems incorrect to consider "legal status" of reputation exclusively in the context of the principal purpose of commercial activity - gaining the profit. Regardless of whether damage to reputation has led to loss of profit or not, reputation in itself will represent a separate non-material category, which has its own value. Consequently, damage, inflicted upon reputation, will, in the first place, entail non-material loss of the value that reputation had.

Thus, it is suggested that, at least in theory, loss of reputation in itself should be recoverable under Article 74. It is the loss, and the principle of full compensation for harm should be the basis for its recoverability. Perhaps, here, it is relevant to cite the critical words of one commentator in relation to the position taken by English courts.

"What about the continued denial of damages for loss of reputation in itself that is a non-pecuniary loss? While perhaps less crucial, there is again no justification for this restriction. Adherence to full compensation dictates recovery, and although proof of this loss may be difficult, as may assessing damages, these are not reasons for blanket refusal …"[99]

In practice, the damages for loss of (injury to) reputation in itself will hardly be recoverable because of the difficulty of proof and meeting the requirements of Article 74.[100] Even if it has been established that loss of reputation had taken place, it will be very difficult to calculate this loss. At best, "all the courts can aim for is a fair and reasonable sum."[101]

As mentioned above, loss or injury to reputation can lead to pecuniary loss in the form of loss of profit. This can be called material manifestation of non-material loss. Loss of profit, in a commercial context, is, probably, the main negative consequence which can be caused by loss of reputation. Being "material manifestation," loss of profit is more likely to be claimed in cases, where reputation has been damaged.[102] As in the case with the loss of reputation in itself, the requirements of Article 74 are of particular importance in establishing the liability of a party in question.

It seems that, in practice, proving this type of loss will not be an easy thing to do.[103] However, once the requirements of Article 74 have been met, this loss should be compensated. For example, the Helsinki Court of Appeals upheld the decision of the Court of First Instance, which had allowed damages, having resulted from loss of goodwill.[104] This decision can, however, be criticised for not having discussed the requirements of Article 74 (foreseeability, for example), which, as has been said, are extremely important in such situations.[105]

Further, English law has identified another type of loss of profit. That is loss of profit, flowing not from an injury to reputation, but from loss of a chance to enhance reputation. Although the cases in question mainly related to actors[106] and authors,[107] it is possible to conceive a hypothetical for international sales. However, hardly can such losses be proved, let alone the establishment of foreseeability, causal link and certainty (if applicable).

Finally, it should be noted that one case, decided by the Tribunal of International Commercial Arbitration at the Russian Federation Chamber of Commerce and Industry, has raised an interesting aspect of the "reputation" problem. That is reputation, not of a businessperson, but of the goods. Case No 054/1999 [108] concerned a contract by instalments. The plaintiff claimed the loss of profit, suffered as a result of a delay in selling and reduction of prices of the goods of the second instalment. This loss, according to the plaintiff, was caused by the fact that the goods of the first instalment had been defective, which, in turn, led to the loss of reputation of these goods on the market. The Tribunal rejected this claim on several grounds. First, there was no causal link between the breach and the loss claimed. Secondly, the plaintiff did not prove that the amount of the claim was commensurate to the breach. Had he proved that damage to reputation had been caused by the breach, he could have been entitled to claim damages only in the proportion, which had, in fact, been caused by the breach. Thirdly, the standard of foreseeability was not established.

However, it seems that had these conditions been met, the Tribunal would have allowed damages for loss of profit, flowing from loss of reputation of the goods. Thus, loss of profit, flowing from loss of reputation of the goods can, in principle, be recoverable. The question as to interrelation of the concepts of reputation of a manufacturer (businessperson) and reputation of the goods may require further elaboration. Here, it will be just stated that, in some cases, reputation of the goods may be considered as a category on its own (separate from reputation of a businessperson) in order to establish the liability in damages.

II. METHODS OF LIMITING DAMAGES

1. General

The principle, which is common to many legal systems, is that of limiting the contractual liability of the party in breach.[109] The purpose of this principle is as follows:

"[T]he full compensation of the expectation and reliance interests would operate either as too strong a disincentive to the assumption of contractual obligations, or to an undue raising of charges to cover such unlimited liability."[110]

The CISG uses a similar approach. It is based on the idea that the recovery of damages cannot be unlimited. This part of the work will examine the methods that the CISG provides in order to achieve this objective, and emphasise the problems, associated with this issue. It is also to be stated that the respective techniques for limitation of damages vary depending on the principles established in a particular legal system.[111] Further, the UNIDROIT Principles, in this respect, represent an interesting example as well: the Principles contain a number of well-known methods of limitation of damages.

2. Foreseeability

One of the methods of limiting damages, which has received an extensive application in various legal systems and international acts, is the principle of foreseeability.[112] This principle has a long history. It was first established in Roman law.[113] Much later, it was established in the Code Napoleon and, consequently, adopted by a number of legal systems.[114]

This rule has been adopted by the Common law as well.[115] It was established in a famous case Hadley v. Baxendale [116] and further restated in Victoria Laundry v. Newman Industries.[117] The UNIDROIT Principles also contain an analogous provision in Article 7.4.4.[118] This rule constitutes the main rule of limiting damages in the CISG as well. Namely, the relevant provision provides as follows:

"Damages may not exceed the loss which the party in breach foresaw or ought to have foreseen at the time of the conclusion of the contract, in the light of the facts and matters of which he then knew or ought to have known, as a possible consequence of the breach of contract."[119]

The purpose, in this part of the work, is to examine the approach taken by the CISG, comparing it with those of other legal systems.[120]

(a) Essential Factors in Evaluation of Foreseeability

(i) Knowledge

According to English law, knowledge is an essential element in evaluating foreseeability. Determination of foreseeability depends on the knowledge that the parties had at the time of the conclusion of the contract, or, "at all events", the party in breach had at that time.[121] Although under the CISG it is only the party in breach whose knowledge matters, the position is analogous. It follows from the text of Article 74 that foreseeability should be established "in the light of the facts and matters of which he then knew or ought to have known."[122] This clearly shows that foreseeability should be examined on the basis of the party's knowledge. Thus, determination of foreseeability is in direct dependence on the party's knowledge.[123]

Under English law, knowledge can be of two kinds: imputed knowledge (which in "the ordinary course of things" is possessed by any reasonable person (regardless of whether the party in breach actually possesses it or not) and actual knowledge (which means knowledge the party in breach actually has of some special circumstances, which lie beyond "the ordinary course of things").[124] Such a division of knowledge into two types flows from the two parts of the rule established in Hadley v Baxendale.[125] The CISG, in turn, does not directly establish the two parts of the Hadley rule, which subsequently gave way to the doctrine of two types of knowledge. However, analogous subjective and objective standards have been established with respect to the party's knowledge: "the facts and matters of which he … knew or ought to have known".[126] Therefore, such wording is likely to cover "the ordinary course of things" case as well as "the special circumstances" case.

Generally, knowledge, in the light of an objective standard, should be imputed to the party in breach if it can be objectively considered that such knowledge is based on the experience of the party as a "merchant".[127] At that, the circumstances of a concrete case should be taken into account.[128]

As regards the knowledge, based on a subjective standard, it has been said that "[t]he party in breach will be … considered as having known the facts and matters enabling him to foresee the possible consequences of the breach, and therefore, as having foreseen them, whenever the other party to the contract has drawn his attention to such possible consequence in due time."[129] The question then arises: Should the other party to the contract be the only source, which must be taken into consideration in evaluation of the actual knowledge of the breaching party? It is not argued that the other party to the contract is, in the context of a subjective standard, probably, the main source of information. However, it is not the only available source. It has been correctly stated that "[m]odern business practices (and equipment), accounting methods and the extensive communication of information make more knowledge available to both parties … [and] [a] potential breacher today will have available a great deal more information about what can happen concerning the contract and hence "ought to know" a great many more facts than a potential brea cher in the nineteenth century."[130] Therefore, in deciding whether the party in breach can be considered as having known "the facts and matters", a right balance has to be found in relying on available sources. This means that we will need to assess the proportion, in which each of the sources of information can be said to have contributed to the formation of the party's knowledge. However, ultimately, the specific circumstances of a particular case should be decisive.

A further mechanism for determination of actual knowledge of the party has been provided in Article 8. In particular, Article 8(2) refers to the "statements and other conduct" of the party and together with Article 8(3) provide for the rule of interpretation of the statements and conduct of the party.[131] These statements and conduct of the party in breach can sometimes serve as important indicators of the knowledge he had at the time of the conclusion of the contract.

(ii) Terms of the Contract

It has been said that the foreseeability rule "reflects the terms of the contract" and, therefore, "precedence is always given to the express or implied intentions of the parties which define those terms."[132] It is true that the terms of the contract, together with knowledge of the party in breach, are important factors in evaluation of foreseeability. It is also true that in case there are hesitations as to the sequence or priority of application of these elements, precedence should be given to the "express or implied" intentions of the parties with respect to the terms of the contract. The basis for this statement is Article 6 of the CISG.[133] It is also to be mentioned that in this context, Article 8 will be the mechanism of determining the intentions and interpreting the respective terms of the contract.[134]

However, it is submitted that the statement above can bring about the following considerations as well. The statement that the foreseeability rule "reflects the terms of the contract" may seem to confine the entire concept of the foreseeability to the content of the contract. It would be more correct to say that foreseeability is partly reflected by the terms of the contract. Besides the contract terms, there are other elements, which are essential in evaluating foreseeability: knowledge and trade usage.[135] These two elements may or may not be explicitly reflected in the contract. Accordingly, the party's actual foresight and the ability to foresee may not always be explicitly reflected in the contract.

(iii) Trade Usage

It seems that, in some cases, a trade usage can also serve as an additional factor for evaluation of foreseeability. For example, in one case, the German Supreme Court held that subjective and objective tests in relation to foreseeability[136] "can be conclusively met by a showing of trade custom as to foreseeability."[137] It also follows from this decision that a trade usage can be relevant for determining both subjective and objective standards with respect to foreseeability. This statement, in turn, brings about some theoretical considerations, which, however, do not seem to have any practical significance.

Article 9 of the CISG contains both subjective and objective grounds for applicability of a usage to the parties' legal relationships.[138] It seems that where a trade usage is relevant in evaluation of foreseeability, the applicability of an objective or a subjective standard of foreseeability can be linked to the grounds provided for in Article 9.

If a subjective ground is applicable, i.e. if the parties have specifically agreed to a particular trade usage, or established a practice between themselves,[139] or knew of a usage,[140] then such a usage or practice will be likely to determine the actual knowledge of a party in breach. The actual knowledge, in turn, can, on the one hand, establish the actual foresight. On the other hand, the fact that a party actually knew of something does not necessarily mean that he actually foresaw the consequences in question. The actual knowledge can as well lead to the establishment of an objective standard, i.e. that a party, having known of certain conditions, was in a position to foresee the consequences of the breach, but did not in fact foresee them.

If an objective ground for applicability of a usage comes into play, then this ground is likely to impute the knowledge of the party in breach.[141] Provided that a party did not actually possess the knowledge, the imputed knowledge will be more likely to lead to determination of an objective foreseeability ("ought to have foreseen"), rather than of an actual foresight. The reason for this conclusion is that it is highly unlikely that a party will actually foresee the consequences if he does not actually have necessary knowledge.

(b) Whose Foreseeability?

Article 74 makes it clear that it is only the party in breach who is required to foresee or to be in a position to foresee.[142] The position is somewhat different in English law. In particular, in Hadley v. Baxendale, the requirement was that the loss be "in the contemplation of both parties."[143] It seems, however, that this divergence will not produce any substantial differences between the application of the two rules.[144] The reason is that it is the breaching party whose foreseeability matters because it is almost always that "the plaintiff knows his business and circumstances better than the defendant."[145]

(c) Relevant Time for Evaluation of Foreseeability

In English law, the relevant time for evaluation of foreseeability is generally the time of making the contract.[146] This rule "is well settled and has proved remarkably resistant to change."[147]

The position is the same in the CISG. Article 74 directly refers to "the time of the conclusion of the contract".[148] In general, this issue in the CISG seems to be "problem-free". The only point that this work will emphasise is that the time is an important factor in assessing foreseeability. This is so because other important elements of foreseeability such as, for example, knowledge of the party in breach or certain circumstances of the case will be examined only within the limits of this particular period of time, i.e., the time of the conclusion of the contract. Therefore, precision in relation to the time becomes very important. In this regard, it has been correctly stated that the "negotiating leading to the conclusion of the contract may … last a certain period of time."[149] The correct view seems to be to evaluate foreseeability at the time when "the contract came into being"[150] or entered into legal force. This approach is in line with that taken in some legal systems, where the conclusion of the contract is the basis of its entry into legal force.[151]

It is also to be noted that, since the moment of entry into legal force is decisive in the evaluation of foreseeability, careful attention should be paid to the requirements of some legal systems predetermining the entry of the contracts into legal force. This statement is primarily relevant to those countries which have certain requirements as to the form of the contracts and rules on state registration of the contracts, and which have made a reservation under Article 96 of the CISG.[152] The Russian Federation, for example, has made a reservation under this article[153] and provides for certain requirements with respect to the form of external economic transactions.[154]

Thus, the moment of the conclusion of the contract, i.e., the moment of its entry into legal force, is the decisive time in determination of the party's foreseeability. No possible foreseeability, which may take place after this moment, should have any legal consequences.[155]

(d) Objective and Subjective Standards with Respect to Foreseeability

In English law, we have seen the manifestation of objective and subjective standards with respect to the knowledge, which has been established as an essential element for evaluation of foreseeability.[156] What are the standards with respect to the foreseeability test itself?

The first part of the rule in Hadley v Baxendale has been interpreted to mean the objective standard, i.e., that "the defendant is liable for loss which any reasonable person in his position could have foreseen."[157] The second part of the rule was construed as a "mixture" of two elements: "the defendant is liable for loss which could have been foreseen by a reasonable person with the same knowledge of special circumstances as the defendant had."[158] The same commentator, stating the fact that the objective element enters into the second part of the rule, gives an example of two views with respect to this fact.[159] In particular, the first view is that "the rule applies not only where the defendant knew of the special circumstances, but also where he had reason to know."[160] At that, reference is, inter alia, made to Article 74 of the CISG.[161] It is correct that the CISG provides for both standards with respect to foreseeability[162]: "foresaw or ought to have foreseen."[163] But the difference between the provision in the CISG and English law, in this respect, is that Article 74 strictly divides these two standards. Within the rule, the two standards do not "enter into each other." Whereas, in English law, it has been shown that both elements are present in the second part of the rule.

However, even though there is a strict division of standards in Article 74, both of them are equally applicable. In order to determine the foreseeability, it will be sufficient to prove either that the party actually foresaw the loss, or was objectively in a position to foresee it.[164] Therefore, it is not necessary to prove that the party in breach actually foresaw the loss.[165] The proof of an objective element will be sufficient to make the party liable for loss.[166] However, such liability may "be restricted on the basis of a reasonable allocation of risks under the contract."[167] In particular, it may explicitly or implicitly follow from the terms of the contract that certain losses should not be covered by the party's liability, even though they were foreseen or objectively foreseeable.[168]

(e) What Must Be Foreseen?

The foreseeability, in Article 74, directly refers to "the loss … as a possible consequence of the breach of contract."[169] Therefore, it is the (amount of) loss, which must be foreseen.[170] At that, most leading commentators agree that Article 74 does not require the foreseeability of the precise amount of loss.[171] The loss is said to be foreseeable "if the risk that has actually materialized is essentially the same as the risk which was foreseeable at the time of the conclusion of the contract."[172] The crucial question, however, is what concrete factors the party in breach had to foresee or ought to have foreseen to be liable for the loss?

The first such factor is the possibility of the loss.[173] This conclusion flows directly from Article 74, which provides that the loss must be foreseen as "a possible consequence of the breach."[174] There is no doubt that the risk of loss is in direct connection with the type of a potential loss. Therefore, the second factor, which the party had to foresee or ought to have foreseen, is the type of the loss.[175] It is further submitted that foreseeability should also relate to the possible extent of the loss (the third factor).[176] The party in breach should not be held liable for the full extent of the loss, if he could not have reasonably foreseen or was not in the position to foresee that such extent would follow from the type of the loss which he foresaw or ought to have foreseen. The party should be liable only to the extent which he foresaw or ought to have foreseen as the possible extent of the loss. It is also to be noted that in evaluating the possible extent of the loss, the manner in which the loss was caused, or the events which led to the loss having acquired the extent in question, can often be decisive. Therefore, arguably, these aspects can be regarded as necessary factors that the party had to foresee or ought to have foreseen to be liable for the extent of the loss in question.

3. Causation

Next method of limiting damages, which is used by some legal systems, is that of causation.[177] The discussion, in this section, will concentrate on this issue in the framework of the CISG.

Article 74 provides that damages for only such loss, as has been "suffered … as a consequence of the breach," can be recoverable.[178] Therefore, we can see that there is a requirement as to the presence of causal link between the breach and the loss.[179] The concept of causation in different legal systems gave rise to the development of various "theories" [180] of causation.[181] Does the causal link, established in Article 74, leave us any room for developing the "theoretical background", which would underpin it? Will there be a need to do so?

Some commentators believe that since the foreseeability rule is used, there can be no room for further theoretical development of the issues of causation.[182] Others merely avoid the question, laying everything on the foreseeability rule.[183] The major implication of these views seems to be that the foreseeability rule excludes the possibility of theoretical development of causal problems. This work does not support this position. In order to express this author's view of this problem, the examination of the inter-relationship between the two concepts will be carried out. Is foreseeability capable of fully replacing the potential scope of the causation concept?

There can be no doubt that these two concepts strongly overlap. Their close inter-connection has given rise to confusion in different legal systems. The confusion, primarily, manifests itself in the fact that foreseeability has been used to establish the causal connection.[184] For example, in the American legal literature, it has been stated that the only test of causation was foreseeability.[185] A Swiss author, in defining the theory of adequate causation, essentially, used a foreseeability rule.[186] Such confusion is said to take place even in France, where the two requirements are provided for in separate articles of the Civil Code.[187]

Thus, the present author recognises that foreseeability and causation are closely inter-related and hardly does it seem possible to rigidly separate them from each other. However, such an inter-connection cannot serve as a basis to consider the two concepts as mutually exclusive. Nor is it correct to regard foreseeability as being capable, at least on a theoretical level, of fully replacing the potential "effect" of causation.

The concept of causation requires some examination. It can be defined as "sequence of classes of complex events or conditions"[188] or an objective connection between the events.[189] In legal science and practice, causation is either used for establishing both the existence of liability and the extent of liability, or for determining just one of these elements.[190] In a view to achieve these purposes, this method, at first, artificially limits the "range of events" (otherwise, the events, which we identify as cause and effect, will go away to infinity into two opposite directions).[191] Then, one of the following questions will need to be answered: (a) What caused the event in question? or (b) Is the causal link between the two events in question sufficient to establish either liability or the required extent of liability? In order to answer either of these questions one of the "theories" of causation is used. In answering the first question, we identify one event (breach, for example), out of all preceding events, which contributed to the arising of the event in question (loss, for example), as the cause. In answering the second question, we determine whether there is a required causal connection between one preceding event (breach) and a subsequent event in question (loss). If a causal connection, required by a particular theory of causation, is found, then the event in question (loss) can be considered as having been caused by one of the preceding events (breach).

On the basis of this view of the concept of causation, the present author will emphasise three main reasons why, in theory, foreseeability cannot fully serve as a substitution for causation.

The first reason is that these two methods should generally be used at different stages. As mentioned above, causation artificially establishes the range of events and determines the causal connection between the events. Once this has been done, foreseeability is applied to determine the limits to the consequences, to which liability would have extended had causation alone been applied. In other words, foreseeability limits liability to something less than the loss, which the breach is said to have caused.[192] Therefore, the foreseeability rule should, generally, serve as a final "cut-off" of liability.

However, this may not always be the case. In rare cases, it is causation which should be the "cut-off" of liability (second reason). Contrary to the view that development of the theory of causation is irrelevant within the framework of the CISG,[193] there may be situations where foreseeability alone will not be capable of dealing with the problem of limiting liability. This should be the case in situations where, alongside with the breach, there is another event (or events) that could have equally led to the occurrence of the loss that the breach is said to have caused. Imagine that it is extremely difficult to establish the real cause of the loss:[194] was it the breach or was it that other event(s)? At that, the party in breach foresaw or was in the position to foresee the possibility, the type, and the possible extent of the loss because this loss ought to have been foreseen as a possible consequence of the breach. It seems that, in this type of situations, a certain approach to treatment of causal problems may be necessary.

Thirdly, it can be argued that causation should be established "through" foreseeability. For example, one author has stated:

"The living conviction of … [a] … man … that there is uniformity in the sequence of events, that we can in good measure predict the future from the past, and that we can in some degree ourselves control the future, is all that we are expressing when we assert the relation of cause and effect."[195]

Indeed, foreseeability largely consists of an element of causation. Without an understanding of how events can affect each other and of "a degree of uniformity of sequence of events," it would be impossible to foresee anything whatsoever. However, causation as a phenomenon exists on its own regardless of our knowledge of the world. It is an objective phenomenon.[196] Therefore, it seems incorrect to bring an objective process, which exists independently of our perception of the world, entirely down to the way a person could foresee the potential causal processes. The foreseeability rule under the CISG includes both subjective and objective standards. The way a person had actually foreseen or been in the position to foresee the potential development of events, at the time of the conclusion of the contract, does not necessarily coincide with the way such a development has, in fact, taken place. In determining liability or extent of liability, we need to rely on an objective sequence of events, and not on the way a person foresaw or ought to have foreseen that sequence. As mentioned above, foreseeability should generally be used after an objective sequence of events has been established.

Thus, in theory, foreseeability cannot serve as a substitute for causation. In support of this view, it should also be noted that some legal systems employ both these methods and in spite of the confusion, which sometimes takes place, do not regard the use of foreseeability as excluding the possibility of theoretical development of causal problems.[197] Rather, these concepts should supplement and balance each other. As has been correctly stated by one author, "[the] doctrines on foreseeability and…causation could be applied in a rather consistent manner and Art. 74 is certainly flexible enough to accommodate an application of [these] general principles."[198]

The next question is whether there will be a need for the theories of causation? With respect to contractual liability, the issues of causation are more of theoretical importance, rather than practical.[199] International sales transactions will be, as a rule, based on contractual relationships, and, in essence, the statement above will be applicable to them. Therefore, the problems, connected with the issues of causation, will hardly represent any practical significance. In many cases, it will be, probably, possible to dispense with an examination of causal issues, by using the foreseeability standard only. However, we cannot exclude such a possibility at all. Therefore, it is important to emphasise the types of cases, in which causal problems may be relevant.[200] First, the cases related to damage to property caused by defective goods can be of relevance.[201] These cases seem to be governed by the Convention.[202] Secondly, causal issues may be relevant in the situations where an injured party had to incur expenses in the situation, created by the breach.[203] The third case, where causation may be of importance, is said to be where the party has been deprived of the loss of profit.[204]

Thus, it has been suggested that at least, on a level of theoretical considerations, and for the sake of those rare cases, where causation may be relevant, the development of ways of treating causal problems under the CISG should be carried out.

4. Mitigation of Loss

The next method of limiting damages, which is used in the CISG, is the principle of mitigating loss.[205] Some legal systems [206] and international documents [207] provide for this method as well.

(a) Meaning, Purpose and Status of the "Mitigation" Provision

The central idea, underlying the principle of mitigating loss is that the aggrieved party cannot recover damages with respect to loss which he could have reasonably avoided.[208] The purpose of this principle is to prevent the injured party from passively waiting for the loss to take place and then suing the party in breach for this loss, while such loss could have been avoided by the injured party.[209] From the economic point of view, it has been said that it is "unreasonable … to permit an increase in harm, which could have been reduced by the taking of reasonable steps."[210] In the CISG, this principle is reflected in Article 77 and has been formulated as follows:

"A party who relies on a breach of contract must take such measures as are reasonable in the circumstances to mitigate the loss, including loss of profit, resulting from the breach. If he fails to take such measures, the party in breach may claim a reduction in the damages in the amount by which the loss should have been mitigated."[211]

The requirement of mitigating loss pertains only to the injured party's right to damages.[212] It follows from Article 77 that if the aggrieved party fails to mitigate, the party in breach will have the right to claim reduction in damages by the amount which could have been avoided.[213] Therefore, the failure to mitigate will not affect the injured party's claim for other remedies.[214] The only exception is said to be the case where it was reasonable to expect the injured party to carry out certain actions, for example, in the form of avoidance of the contract or of the conclusion of a cover transaction, [215] in order to mitigate the loss.[216]

As regards the amount, by which the damages should be reduced, the following formula is to be followed:

Another issue, which needs to be considered, is that of the "status" of this provision. It has been formulated as if it imposes an obligation on the party: "[a] party … must take such measures".[218] But does it really represent an obligation as such?

Some sources state that the provision in Article 77 is one of several provisions of the Convention (together with Articles 85-88), which provide for a "duty owed by the injured party to the party in breach."[219] On the other hand, it has been stated that under this provision the injured party "is under an "obligation to herself" to mitigate her loss."[220] It seems that both of these opinions cannot be fully accepted as correct.

First, an obligation can be defined as a legal relationship, by virtue of which one party is entitled to demand from the other party the performance of certain actions.[221] On the basis of Article 77, the breaching party cannot demand from the injured party the performance of this "duty." Therefore, the injured party does not, in fact, owe such a "duty" to the party in breach.

Secondly, we cannot, properly speaking, refer to mitigation as to "an obligation to [him]herself." It would contradict the essence of an obligation as a legal concept [222] and, consequently, the party "cannot owe a duty to him[her]self."[223]

Thus, we can see that even if it is possible to refer to mitigation using such terms as a "duty" or an "obligation,"[224] the nature of this "duty" is substantially different from other obligations under the CISG.[225] In fact, it does not represent a contractual obligation.[226] There are two principal reasons for such a conclusion.

The first reason has already been touched upon, but will be reiterated again: the "duty" under Article 77 does not represent a legal relationship between the parties which gives one party the right to demand a certain action from the other.

Secondly, the breach of an obligation is the basis for liability under the Convention.[227] However, the breach of the "duty" to mitigate will not give rise to any form of liability under the CISG.[228] Non-compliance with Article 77 will entail the loss by the injured party of the right to claim those damages which could have been avoided.[229] Therefore, the view that this provision can conduce to development of a general principle which would establish a duty of "loyalty to the other party to the contract", the breach of which results in damages,[230] is not supported in this work.[231]

In one respect, however, it seems that the "duty" to mitigate may represent the basis for refusal to enforce the party's right to specific performance.[232] Nevertheless, even in that context, this provision should not be construed as an obligation in a legal sense. Its function will be the prevention of the party's exercising his right to the remedy of specific performance.[233]

The opinion, expressed in this work, with respect to the status of the "duty" to mitigate is in line with the approach taken in English law. The relevant position has been formulated as follows: "A plaintiff is under no duty to mitigate his loss, despite the habitual use by the lawyers of the phrase 'duty to mitigate'. He is completely free to act as he judges to be in his best interests."[234] [235]

(b) Reasonable Measures

According to Article 77, measures to mitigate loss must be reasonable in the circumstances concerned.[236] The type of measures that need to be undertaken depends on the criterion of reasonableness.[237] The latter, in turn, depends on and will be construed in the light of the circumstances in question.[238] In general, it has been said that a measure is reasonable "if under the particular circumstances, it could be expected to be taken by a person acting in good faith,[239] or if it is "adequate" and preventive with respect to the loss.[240] In the evaluation of the situation, regard should be also had to the party's skills and position as a businessman, such as, for example, "ingenuity, experience, and financial resources," etc.[241] At that, relevant trade usage, if any, should be taken into account as well.[242] The aggrieved party is not, in any way, obliged to take measures, which, in the circumstances concerned, are "excessive"[243] and entail unreasonably high expenses and risks.[244] If the party refrains from such measures, he will not be considered as not having complied with Article 77.[245]

What type of measures can be possibly meant in Article 77? Article 77 provides that the measures should be aimed at mitigation of "the loss, including loss of profit, resulting from the breach,"[246] Following Article 74, this provision refers to all kinds of loss. It is understandable that, in practice, different types of loss can give rise to a great variety of situations. Consequently, the decision on how and in what way an injured party should have mitigated his loss can be made only on the basis of careful examination of all circumstances of a concrete situation, criterion of reasonableness, and the type of loss in question. Therefore, it does not seem possible to list every single measure which is implied in this provision.

However, in order to illustrate how wide a range of possible mitigating measures can be, some examples of such measures will be given. Mitigation can, for instance, have the form of making a substitute transaction (resale or repurchase),[247] avoidance of the contract,[248] finding a sub-contractor,[249] expediting shipment of the goods, which have not been purchased in a cover transaction,[250] sub-charter of a vessel,[251] or contacting a party in breach and submitting the documents, proving the claim, in order to receive necessary information, which could help in mitigating the loss.[252]

Further, it is also noteworthy that a peculiarity can be discovered in some cases of mitigation. The problem is that mitigation itself can bring about certain forms of loss.[253] In other words, mitigation can often be the source of loss. In taking certain mitigating measures, an injured party may have to incur a number of different expenses such as, for example, the costs of storage, repair costs or brokerage costs. Is it required and is it possible to mitigate this type of loss, or, in other words, to mitigate a measure aimed at mitigation? It is argued that the wording of Article 77 is broad enough to cover this situation and, therefore, requires mitigating this type of loss as well. It is also submitted that it is not impossible to mitigate this kind of loss. For example, suppose that the Buyer informs the Seller that he will not be able to accept delivery and pay for the goods. The contract has been avoided, and the Seller mitigates his loss by reselling the goods. At that, in such a cover sale, he had to incur a certain amount of brokerage costs. If it can be proved that it was reasonable to avoid these brokerage costs, then these costs should not be included in the claim. Likewise, if it were reasonable to incur lesser amount of the brokerage costs, the claim should be, accordingly, reduced. Finally, it should be said that since these damages can be caused by a diverse number of situations, the measures preventing this loss will vary accordingly.

(c) Mitigation of Loss in Case of an Anticipatory Breach

It has been said that "one challenging area for the prospective operation of the duty to mitigate" is its applicability with respect to an anticipatory breach.[254] Indeed, is the application of the measures to mitigate required in connection with an anticipatory breach and can the measures be applied in such a case? This question requires some examination.

It has been stated that the "duty" to mitigate should apply in case of an anticipatory breach of contract.[255] Generally, such a conclusion is based on the following reasoning: "The aim of Article 77 is to encourage mitigation of loss. To this end, measures directed at mitigating the loss are to be taken as soon as the party to the contract could foresee the danger of breach of the contract by the other party and of his potential loss."[256]

With respect to a fundamental breach, in particular, it has been said that if it is clear that such a breach will take place,[257] the party concerned "cannot await the contract date of performance before he declares the contract avoided and takes measures to reduce the loss arising out of the breach by making a cover purchase, reselling the goods or otherwise."[258] Further, it should be remembered that Articles 71 and 72 are the provisions in the CISG which govern the conduct of the parties in the "anticipatory breach" situations. It is important, for the purpose of this discussion, to bear in mind that the procedure in these provisions is of non-mandatory character: it does not oblige the aggrieved party to suspend his obligations where it is apparent that the other party will not perform a "substantial" part of his obligations;[259] in a similar vein, it is not required that the aggrieved party avoid that contract when it is clear that a fundamental breach will take place.[260]

Some commentaries recommend that in case of an anticipatory breach, in particular, in case of a fundamental breach, the procedures prescribed in Articles 71 and 72 be used.[261] It is further stated that if the party does not follow this procedure, i.e. does not suspend his performance and avoid the contract, and insists on the performance by the other party, there will be a risk for the party to be found not in compliance with Article 77.[262] Therefore, if the aggrieved party, in a situation of an anticipatory breach wants to comply with Article 77, he should either follow the procedure in Articles 71 and 72 and, if a positive result does not follow, subsequently mitigate; or he should mitigate as soon as he could foresee the breach.[263]

However, as mentioned above, while the duty to mitigate may, in certain cases, be applied to an anticipatory breach situation, there is no such general requirement.[264] If, for example, prior to the contract date, one party refuses to perform his obligations and wants to repudiate the contract, nothing in the Convention obliges the other party to follow Article 72, i.e., to avoid the contract or accept the repudiation. He is fully entitled to continue his performance and to demand the performance from the other party. Accordingly, there will be no need for him to mitigate if he continues to perform and expects the same from the other party.[265] Therefore, generally, if the performance from the party in breach does not take place at the contract date, the injured party can subsequently sue for damages, without apprehending that his claim can be reduced.

However, it is further suggested that this solution should not always be the case. There are categories of cases where the duty to mitigate, in a situation of an anticipatory breach, should be regarded as necessary in order not to suffer a sanction of reduction in damages under Article 77.

Why should there be such a necessity? In order to find an answer to this question let us, first, consider whether analogous considerations arose in some legal systems. It seems that English law can be particularly helpful in this respect.

In particular, a general rule is that when one party repudiates the contract prior to the performance date, the other party has an option. He can refuse to accept the repudiation and treat the contract as subsisting. In this case, the contract continues to exist and no need to mitigate arises. Alternatively, he can accept the repudiation and treat the contract as at an end. In this case, he has the right to sue the breaching party at once. This right will then be subject to the mitigation rule.[266]

However, despite this rule being a well-established one, it is subject to two exceptions: (1) where it could simply not work; and (2) where it would lead to "wholly unreasonable" results.

The first category includes cases where performance of the obligations is based on the cooperation between the parties. Where performance of the obligations by the innocent party depends on that of the party in breach, it may turn out that the former will merely not be able to carry out his part of the obligations without the latter's cooperation.[267] "In most cases by refusing co-operation the party in breach can compel the innocent party to restrict his claim to damages..[268] Therefore, where it is established that performance was not possible without the cooperation of the parties, the innocent party can be found as having been bound to accept repudiation.[269] In this case, his right to damages will be subject to the "duty" to mitigate.

The second category includes the situations where the absence of the so-called "legitimate interest" on the part of the innocent party has been proved. "[I]f it can be shown that a person has no legitimate interest, financial or otherwise, in performing the contract rather than claiming damages, he ought not to be allowed to saddle the other party with an additional burden with no benefit to himself. If a party has no interest to enforce a stipulation, he cannot in general enforce it: so it might be said that, if a party has no interest to insist on a particular remedy, he ought not to be allowed to insist on it."[270]

An example, given in White and Carter (Councils) Ltd. v. McGregor, will help illustrate the concept. A company had concluded the contract with an expert. Under the contract, the expert undertook to go abroad in order to compile a report. Shortly after the conclusion of the contract and before anything was done, the company repudiates the contract. The expert still intends to carry out his part of the obligations. It has been said that "[t]o allow such an expert then to waste thousands of pounds in preparing the report cannot be right if a much smaller sum of damages would give him full compensation for his loss."[271]

What are the criteria of determining the legitimate interest? At present, no distinct criteria have been developed in English law. Nevertheless, certain rules, in this respect, have been established.

First, from the example above, it may seem that reasonableness is the criterion. Certainly, reasonableness is a very important factor. However, the proof of the innocent party's having acted unreasonably will not suffice to prove the absence of the legitimate interest.[272] In order to determine the presence or absence of the legitimate interest, it will be necessary to distinguish between "merely unreasonable" and "wholly unreasonable" actions.[273]

Secondly, in order to establish the legitimate interest it will not be sufficient to determine only the interests of the innocent party. The innocent party must take into consideration the interests of the breaching party as well.[274] Presumably, this rule represents a further development of the primary statement of the basic rule in White and Carter (Councils) Ltd. v. McGregor, which provided that the breaching party should not be "saddled … with an additional burden."[275]

Thirdly, the legitimate interest should be established on the condition that damages are an adequate remedy.[276]

Since the concept of legitimate interest is rather abstract these rules may seem to be quite "vague". It has been correctly pointed out that an "absolute certainty can never be attained."[277] Nonetheless, these rules will represent the basis for determining the presence or absence of the legitimate interest in the context of the circumstances of a concrete case.

Now, are the situations involving the elements of cooperation and legitimate interest likely to arise in international sales? Should such situations be treated as "exceptions" to the general position under Article 72? The answer to these questions should be "yes".

Let us start with the cooperation element, at first. Suppose that it is "clear" that a fundamental breach is going to occur.[278] The innocent party, however, thinks that he is perfectly entitled to the performance and does not avoid the contract. The contract, of course, remains in existence. The innocent party needs to continue to perform because he does not want to be in breach himself. The problem, however, is that he cannot perform unless the breaching party cooperates. In other words, the performance of the innocent party's obligations is pre-determined by that of the breaching party. For example, in an Ex-works contract, the buyer has an obligation to take delivery of the goods when they are placed at his disposal.[279] However, he will not be able to take delivery if the seller does not carry out his delivery obligation, i.e. does not make the goods available to the buyer at the seller's premises or some other place where the goods are stored. In an FOB contract, the seller will not be able to make a delivery if the buyer does not nominate a ship. The seller will not be able to manufacture the goods for the buyer if the latter does not perform his obligation to supply the seller with necessary materials. In these types of situations, the innocent party may merely be compelled to treat the contract as avoided. If he is compelled to avoid the contract, he will be subject to Article 77.

The question arises: Can the innocent party require specific performance in such cases?[280] Here, it is necessary to draw a line between the obligations, which are already due, and which are not. For example, in the FOB contract the refusal to pay and accept a delivery before the due date can be regarded as an indication of an anticipatory breach. But in the context of cooperation, the innocent party will need the enforcement only of those obligations, which are necessary for him to perform his part of the contract. Therefore, in order for the FOB seller to make a delivery, he will need the enforcement of the obligation to nominate a ship. If the seller wants to go on with the contract, the buyer's refusal to perform future obligations (to pay and to take a delivery) will not affect his ability to deliver the goods. Accordingly, at this stage the seller does not need to worry about the enforcement of the future obligations. He will need to worry about it, when he succeeds in enforcing the buyer's obligation to nominate a ship, completes his part of the contract, and expects the buyer's final performance. When the seller seeks to enforce the obligation to nominate a ship, he will be dealing with an actual, but not an anticipatory breach. This is so because this obligation will be due, sooner or later, since it precedes the seller's delivery date. Accordingly, the innocent party will be entitled to specific performance with respect to this obligation.

However, this remedy may not always be available. Article 28 provides that "a court is not bound to enter a judgement for specific performance unless the court would do so under its own law in respect of similar contracts of sale not governed by the Convention."[281] If the innocent party is not successful in enforcing this remedy, then we will have no choice but to treat the contract as avoided and will, certainly, be under the duty to mitigate his loss.[282]

Will the innocent party's right to specific performance be affected because of the breaching party's inability to perform? Generally, the answer should be "no". In a legal sense, this right will remain unaffected.[283] He will lose that right only if the inability is based on the conditions, set out in Article 79.[284] However, it seems that in practice, it will be difficult to force a party to perform when he objectively cannot do so.

Thus, in international sales, the cooperation element may sometimes compel the innocent party to avoid the contract and to be subject to the "duty" to mitigate. However, the innocent party will, normally, be entitled to specific performance. This remedy will, presumably, be targeted at the actual breach. But we should still regard the co-operation element in the light of an anticipatory breach because the actual breach is the result of a general anticipatory breach. The performance of the obligation, which was actually breached, was to take place earlier than the performance of the obligations, an anticipatory breach of which occurred, is to take place. Further, Article 28 may limit the innocent party's right to specific performance. And even in some situations, where the party still has that right, it may be impossible to enforce it due to the breaching party's objective inability to perform.

The second exception to the right, provided in Article 72, should be the proof of the absence of the legitimate interest. It has already been shown that it is virtually impossible to give a clear definition to this concept. The best way to illustrate it is to consider it in the light of a hypothetical example.

Example A:[285] On June 1, Buyer A and Seller B made a contract for B to produce and deliver to A 10,000 sheets of steel on August 1 at $50 per sheet. A needed the steel for use in manufacturing. On July 1, B notified A that production difficulties in B's steel mill would prevent delivery of the steel by August 1. B also stated that the production difficulties might persist for an unknown period after August 1 and urged A to obtain the steel elsewhere. Comparable steel was available in A's area. The price at all times remained at $50. For unexplained reasons, A did not seek or obtain the steel elsewhere. As a consequence, A's production facilities were shut down for the month of August. A sued B for damages based on shut-down losses of $10,000 per day, or $300,000. Seller B argued that, under Article 77, A failed to "take such measures as are reasonable in the circumstances to mitigate the loss" so that there should be a corresponding reduction in the damages.

It seems that, in this example, A's damages claim should be reduced by the amount, which could have been avoided. This conclusion is based on the absolute absence of any interest A could further have in this particular transaction. What did A's treatment of the contract as existing lead to? Can we trace through realisation or manifestation of any form of interest, and economic interest, in the first place? If A had bought the steel, and he had every opportunity to do so, he would have most likely prevented part of his loss (loss of profit, for example). B, in turn, would have given him an adequate compensation for the loss,[286] which had been caused by the breach and could not have been avoided. The closest that we can get in formulating a general criterion for determining the legitimate interest is to say that A's conduct, in these circumstances, was "wholly unreasonable."

It has been said that, in this situation, A's right to specific performance is irrelevant.[287] The reason is that a court would not be able "to overcome its production facilities."[288] In this case, we, once again, see the party's inability to perform as well as the court's inability to enforce specific performance in spite of the existence of the right to this remedy. However, it is argued that the proof of the absence of the legitimate interest should prevent the innocent party from exercising his right to specific performance.

Example B: [289] B undertook to manufacture certain goods for A. The goods are to be produced according to A's particular specifications. It is not possible to find any market for these goods, since the goods are suitable only for a very specific use. Before B starts the production, A makes it clear that he no longer has an interest in the goods and urges B not to start that manufacture. Should B, nevertheless, start the production, exercise his right to specific performance, and force A to accept the goods and pay for them?[290]

It seems that B should not be allowed to do so. Knowing that damages can provide with adequate compensation, B does not have any legitimate interest in manufacturing the goods, which will be of no use whatsoever. In this case, it is particularly important that the innocent party bear in mind the interests of the breaching party.[291] The court, in turn, should not grant specific performance. The exercise of this remedy will impose substantial burden on A without any benefit to B.[292] The correct solution should be B's avoidance of the contract and receiving an adequate compensation in the form of damages.

How can we reconcile this approach with a legal right to specific performance established by the CISG? It seems that the only "counterbalance" to this right is a "duty" to mitigate. It has been correctly pointed out that "the different theoretical approaches may fade away when transformed into practical realities."[293]

Thus, we have seen that in some situations, keeping the contract alive may be "wholly unreasonable and untenable."[294] In such situations, the innocent party will not, as a rule, have any legitimate interest in further performance of the contract. Although the concept was developed in English law, it potentially embodies the practical situations, which can equally arise in international sales. Moreover, rules for determining the presence or absence of legitimate interest, established in English law, can serve as extremely useful guidelines in practice. Finally, the right to specific performance, which the innocent party can try to exercise, should, in such cases, be restricted.

(d) Mitigation in a "Lost Volume" Situation

This work has suggested that a "lost volume" situation should be governed by the CISG. A "lost volume" seller should be allowed to recover damages flowing from this type of loss, provided that a number of requirements, suggested by this work,[295] and standards for limiting damages have been met. However, the peculiarity of a "lost volume" situation is that it makes it, in principle, impossible for one of such standards to be carried out. Namely, the operation of the mitigation rule is not possible in such circumstances.

It has been stated that, depending on the circumstances, mitigation measures can have different forms.[296] However, when a seller suffers loss of volume, his mitigation measure (if it were possible to mitigate) should, generally, have the form of finding a substitute buyer and reselling the goods under the original contract. Where a seller makes such a resale, thinking that he, thereby, performs his "duty" to mitigate, he does not, in fact, avoid his loss. He will not minimise the profit, lost as a result of the breach of the first transaction, by making a second transaction, because the second transaction would have been made even if there had been no breach. Let us illustrate the point.

A agreed to buy a bicycle from B for $100. This deal would give B a profit of $10. A breaches the contract by refusing to take a delivery of the bicycle and pay for it. B resells the bicycle to C, to whom he would have sold an identical bicycle in any event, even if A had not breached. Therefore, the sale to C will not affect B's actual damages. B should be entitled to claim $10 as his lost profit.

An analogous situation has arisen in a case, decided by Oberster Gerichtshof 28 April 2000.[297] The court found that, after the buyer had breached the contract by having refused to pay for the jewelry, the seller did not conclude a substitute transaction. The buyer contended that the seller had failed to mitigate his loss, as required under Article 77 CISG. The court held that this argument was ineffective "as far as the promisee, in performing the substitute transaction, would have lost another similar transaction bringing the same profit as the first transaction … [Buyer's] (completely unsubstantiated) objection that [seller] failed to mitigate damages is therefore irrelevant, because the Court of First Instance found that a miscellaneous resale of the goods intended for the [buyer] would have materialized independently of the [buyer's] order."[298]

Thus, it can be, generally, concluded that a "lost-volume" seller cannot mitigate his damages. Can this conclusion be reconciled with the provision, stipulated in Article 77? It seems that Article 77 does not preclude us from making such a conclusion. It provides that mitigation measures should be such "as are reasonable in circumstances." Measures which are known to lead to no mitigation whatsoever can hardly be considered to be "reasonable."[299]

However, it can be argued that circumstances of a particular case can reveal some ways of mitigation. Imagine that in a case, referred to above,[300] although there was no possibility to mitigate by reselling that piece of jewelry itself, because there was no demand whatsoever for that particular design, it is possible to avoid whole or part of the loss in a different way. For example, the seller has broken up that jewelry to pieces, and sold the precious stones, of which the jewelry was made. For our purposes, let's assume that these measures were "reasonable" in the meaning of Article 77 (although it is likely that in many cases such measures can be considered to be "unreasonable" or "excessive"). It seems possible to prove that these measures were, in fact, mitigation. How does this situation correlate with our conclusion that it is impossible to mitigate in a lost volume case? The answer is that at the moment the seller manages to find a "reasonable" way to mitigate, he cannot be regarded a lost volume seller, even though, at the first glance, the situation seemed to be a lost volume situation. First, the central point of the lost volume doctrine is that the seller cannot realise the expected volume. In our example, it may be said that, strictly speaking, the seller lost volume of sales of the jewelry itself. However, in essence, he did not lose volume, because, in one form or another, he managed to resell "the unit". Secondly, one of the requirements that the seller must meet in order to qualify as a lost volume seller is that the first and second transactions must be "wholly independent events,"[301] i.e., there should not be any causal connection at all between these two transactions.[302] In the example in question, the seller would not have broken up the jewelry and sold it in pieces had there been no breach. Therefore, this cannot be a lost volume situation.

5. Certainty

Most legal systems have a requirement as to certainty of damages claimed.[303] An analogous requirement can be found in the UNIDROIT Principles.[304] This part of the work will examine the position of the CISG in relation to this issue.

The CISG does not contain any express reference to certainty.[305] However, this work suggests that this limitation can still be applied to the cases regulated by the CISG.[306] Several ways of how this rule can be applied will be pointed out.

First, it can be applied through the procedural law of the forum.[307] It has been said that the "[p]roblems of proof and certainty of loss are procedural matters which remain within the province of national law, and procedural conceptions may still serve as covert limitations on CISG consequential awards."[308] It follows from this statement that the procedural issues are beyond the scope of the CISG.[309] Therefore, if, for example, the procedural law of country A contains the requirement as to the proof of certainty, the court may consider it mandatory to apply this requirement to a dispute governed by the CISG. In this case, such a decision will reflect the court's opinion that procedural rules are not regulated by the Convention. For example, in a case No 304/1993 (decision dated 3 March 1995) considered by Tribunal of International Commercial Arbitration at the Russian Federation Chamber of Commerce and Industry, the plaintiff was denied compensation of "moral harm." One of the reasons for such a decision was the fact that the amount of the claim was not substantiated.[310] Presumably, the certainty standard within the framework of this Arbitration Institution is based on the requirement that every party must prove the circumstances to which it refers as the basis of its claims and defense.[311] Therefore, the requirement of proving the amount of "loss" has been, to some extent, imposed. The basis for it was the procedural requirement in the respective rules. It seems that the decision implied that the procedural rule was beyond the scope of the CISG.

Secondly, the view can be taken that certainty is a matter governed, but not expressly settled in the Convention. Certainty can be either treated as a procedural issue, "indirectly" governed by the CISG,[312] or merely as a substantive rule governed but not expressly settled in the Convention. Recourse in this case, must be, first, had to one of the general principles on which the Convention is based. The issue of whether the Convention contains a general principle in relation to certainty of damages can be arguable and may require further elaboration. If no relevant general principle is found, the matter must be settled in accordance with the applicable rules of Private International Law (PIL). However, it is important to note that, if certainty of damages is treated as a procedural matter, recourse to rules of PIL may turn out to be irrelevant because, as a rule, rules of PIL point at the substantive, rather than the procedural part of the legal system in question.[313] In such a situation, the only reasonable way to follow is to apply the procedural rules of the forum containing the provisions on certainty.[314]

Thirdly, it is to be stated that the issue of certainty of damages is directly related to the problem of proof.[315] In practice, the proof of the precise amount of damages may not always be possible. Therefore, the extent of compensation can be determined on the basis of a mere discretion of a judge or an arbitrator. Such a solution of the problem of certainty can, first of all, derive from a relevant provision of an applicable law.[316] This result may follow from either of the two approaches, discussed above, i.e., where the issue of certainty is regarded as being either outside the scope of the CISG or "governed, but not expressly settled" in it, as well as from an application of the UNIDROIT Principles.[317] However, an analogous result may also follow where there were no grounds for such discretion. It seems that this approach has been taken in the ICC Arbitration Award 8611/HV/JK of 1997. Namely, it has been stated that because of "the arbitrator's lack of reliable documents concerning the number of the machines for which [buyer's] customers did not pay because of non-delivery of replacement parts, the arbitrator must judge the damages according to his own conviction having taken into consideration the circumstances."[318]

Fourthly, one author argues that the regulation of the issue of certainty can be carried out on the basis of the UNIDROIT Principles.[319] Since one of the purposes of the Principles is to interpret or supplement international uniform law instruments[320] such as the CISG, they can be used to supplement those provisions, which are within the Convention's scope, but not expressly settled in it.[321] That author suggests that the UNIDROIT Principles be regarded as "a component part of the 'general principles' underlying the CISG."[322] Thus, considering the issue of certainty as falling within the Convention's scope, the commentator states that the "UNIDROIT Principles article 7.4.3 complements CISG article 74 by emphasizing that the existence and extent of the harm to be compensated must be established with a reasonable degree of certainty."[323] If this degree of certainty cannot be achieved, the court will have the discretion to assess damages.[324] It is submitted that this treatment of certainty represents a workable solution, which is conducive to maintaining the Convention's international character and contributing to uniformity in its application. However, one can still argue about whether or not this issue is governed by the Convention and whether the UNIDROIT Principles can be regarded as a part of the general principles, underlying the Convention.

In order to develop this discussion further let us analyse other cases, decided in different jurisdictions.

In the Delchi case, the court found that damages for loss of profit must be proved with reasonable certainty.[325] This finding was based on certainty as it has been established in common law.[326] However, in the present context, it can be said that this decision was flawed in two respects. First, it did not contain the basis for the decision to apply a common law standard.[327] Secondly, since certainty had been considered to be a procedural rule, its application should not have exceeded the procedural limits.[328] Namely, the following has been said:

"There is a distinction between a court determining that evidence is unreliable or uncertain and a court not allowing any evidence of a type of loss because the law of the jurisdiction refuses to allow damages for that type of loss as a matter of law".[329]

Thus, the certainty rule, even if it is applied to the CISG case, should not prevent the injured party from claiming the loss, which, otherwise, can be legally claimed under the Convention. Even though that is the way it would be applied under that particular legal system, the court does not have the right to restrict the Convention's legal regime and the legitimate rights, established by the CISG. Moreover, the "international character" of the Convention as well as the need to promote uniformity in its application [330] should prevent the courts from applying this standard in such a way.

A case, decided by the German Supreme Court on the basis of the ULIS,[331] does not make it clear whether, in determining the amount of damages, the court was guided by national law or by the principles of ULIS.[332] Similarly, it is unclear what standards of certainty some other German courts applied. In one case, one of the grounds for rejection the buyer's claim for damages was that the buyer "failed to substantiate her purported damages in detail."[333] In another case, the court stated that under Article 74 CISG, the buyer had to "exactly calculate her damage."[334]

Thus, the standard of certainty can be and is sometimes imposed on the parties in the CISG cases, even though the Convention does not directly provide for it. This work has suggested several ways, through which the requirement of certainty, in different forms, could be applied. The analysis of several cases has revealed different approaches to and the lack of clarity in the treatment of this issue. The question what is a correct approach remains open.

6. Fault

Under the CISG, fault is neither the basis for liability, nor the requirement for availability of any remedy or determination of the extent of the liability. Accordingly, the right to recover damages under the CISG is not connected to "proof or even presumption" of the party's "culpable breach."[335] This conclusion derives from the fact that the basis of liability is any kind of objective non-performance of the obligations under the contract and the CISG.[336] Therefore, this concept cannot produce any legal effects within the framework of the Convention. It has been said, however, that the liability under the CISG cannot be regarded as "absolutely strict"[337] because the party can be exempt from liability under Article 79.[338]

7. Burden of Proof

The importance of the issue of burden of proof should not be underestimated. Although burden of proof is a procedural matter in nature, the way it is allocated between the parties can often pre-determine the outcome of a case. Certainly, this issue is of particular importance when it comes to proving the standards of limiting damages. In order to determine who will bear the burden of proving these standards, it is necessary to identify a general principle of allocation of burden of proof. The problem, however, is that the CISG does not explicitly provide for such a rule. The Convention's silence on this problem has produced divergent opinions of legal scholars and, most importantly, divergent interpretations and applications of the CISG. Namely, some commentators believe that the issue of burden of proof is not governed by the Convention and should be regulated by applicable domestic law.[339] Several cases have reflected this view. In one case, a Swiss court held that the CISG did not contain rules on burden of proof and decided to rely on the rules of Private International Law of the forum.[340] In the ICC Arbitral Award No. 6653, the Tribunal was also of opinion that the issue of burden of proof was not governed by the CISG.[341]

One case has revealed another view. Namely, the Arbitral tribunal regulated the issue of burden of proof on the basis of general principles of law.[342] The discussion of this approach can, ultimately, lead to continuation of a long lasting debate on the status of the concept of lex mercatoria in regulation of international commercial transactions.[343] Here, it will be just stated that this approach to regulation of burden of proof is not supported.[344]

It is argued that the CISG governs the issue of burden of proof.[345] As has been pointed out by many commentators, Article 79 contains a rule, which specifically allocates burden of proof.[346] Accordingly, it cannot be asserted that the CISG does not govern this issue.[347] A number of cases, decided in different jurisdictions, can be referred to in support of this view. For example, several courts have clearly stated that, although the CISG does not expressly deal with burden of proof, it governs this issue[348] and should be interpreted for the relevant principle could be found.[349]

Thus, since the Convention governs this matter, the allocation of burden of proof should be determined on the basis of a general principle, underlying the Convention.[350] The Tribunale di Vigevano has identified such a principle.[351] That is, the party, which invokes its right to assert a claim, must demonstrate the facts, supporting this claim.[352] Basing upon this principle, this court has, in essence, formulated another principle: if a party relies on an exception, it must prove the factual prerequisites of that exception.[353] It is to be noted that these principles have already been formulated in scholarly writings.[354]

Thus, applying these principles to the issue of damages, it can be stated that if the injured party asserts non-performance by the other party and seeks damages, it bears the burden of proving the non-performance and existence of the damage.[355] Further, it is up to the injured party to prove the foreseeability of loss by the other party,[356] the causal link between the breach and the loss[357] and, depending on the requirements of certainty, the actual amount of loss, suffered.[358] As to mitigation, the rule should be as follows: the party, who argues that the injured party has not taken appropriate mitigation measures, bears the burden of proving this allegation.[359]

CONCLUSION

The purpose of this work was to examine the methods of limiting damages under the CISG as well as to highlight issues which need to be developed further. The first part of the work focused on the issue of interested protected and the categories of loss, covered by the Convention. Special attention has been given to the examination of the problems of "lost volume" and "non-material" loss. Suggestions as to regulation of these types of loss under the Convention have been made. The second part of the work concentrated on the methods of limiting damages. It has been shown that potential problems may, primarily, arise in practice with respect to certain aspects of those methods, which are provided in the CISG, as well as with those, which are not directly mentioned in the Convention.

The examination of the foreseeability rule did not reveal any particular difficulties. However, essential factors for evaluation of foreseeability have been emphasised, and some guidelines have been given with respect to different aspects related to this rule.

As to causation, this work suggests that, in the framework of the CISG, it represents a field in which further theoretical development may be necessary. It has been shown that there is a room for such an elaboration. Moreover, diverse practical situations may call for the solving of problems related to causation, although such problems are not likely to arise in international sales transactions very often.

A number of important factors, connected with the mitigation principle, have been considered. In general, it can be concluded that the problems with the mitigation principle are particularly acute in a situation of an anticipatory breach. In this regard, some hypothetical examples have been given, and possible solutions to the problems have been suggested. Further, this work has demonstrated that the operation of the mitigation rule was impossible in the "lost volume" situation.

The certainty concept, in turn, is a principle, which has not been directly provided in the Convention as a method of limiting damages. Nevertheless, this work has shown the ways, through which its application could be possible in practice. However, although its application can be justified in certain cases, it is important to bear in mind the international character of the Convention and the need to promote uniformity in its application.

Finally, the fault principle does not exist in the CISG and cannot produce any legal consequences.

In the end, the present author would like to emphasise the importance of further development of these problems. Only provided that there is a "firm" theoretical basis underpinning these issues will the uniformity in application of the Convention become more realistic.

* Djakhongir Saidov (Republic of Uzbekistan), Bachelor (University of World Economy and Diplomacy, Uzbekistan), LL.M. (University of East Anglia, UK). The author is grateful to Mr. Alastair Mullis for his valuable comments and suggestions on the draft of this work.

1. E. Allan Farnsworth, "Damages and Specific Relief", 27 Am. J. Comp. L., 1979, p. 247.

2. See Herbert Bernstein & Joseph Lookofsky, "Understanding the CISG in Europe", 1997, p. 96.

3. The issue of contractual limitation of damages will not be covered by this work.

4. See Article 7(1) of the CISG.

5. K. Zweigert & H. Kötz, "An Introduction to Comparative Law", 2nd Edition, 1987, Oxford University Press, p. 15.

6. See Treitel, "Remedies for Breach of Contract: A Comparative Account", 1988, p. 76.

7. Andrew Burrows, "Remedies for Torts and Breach of Contract", Second Edition, 1994, p. 20.

8. See Restatement 2nd of Contracts, Pamphlet No 3, 1981, paragraph 344, comment (b), p. 104.

9. Robert Cooter & Thomas Ulen, "Law and Economics", 2nd Edition, 1997, p. 204.

10. Ogus, "T