1. Mistake in contract law
Erroneous assumptions can affect the making of contracts, last wills, and actes juridiques of all sorts. Such instruments, designed to regulate one’s own affairs, malfunction if influenced by erroneous assumptions. All known legal systems address the issue whether it is possible to treat a contract or deed as invalid because of an underlying mistake. Modern codifications, especially on the civilian side, tend to regulate mistake, fraud, duress (threat, coercion, undue influence), and sometimes misrepresentation as different types of ‘defects of the will’. Resolving a possible mistake involves the interpretation of the contract, so that issues of construction and mistake are intertwined.
In the field of contract law the central problem posed by the issue of mistake is the conflict between respecting the parties’ intentions and protecting the parties’ legitimate expectations as to the legal certainty and reliability of transactions. Accordingly, all legal systems try to strike the right balance between allowing the avoidance of contracts on the basis of mistakes and confirming transactions by rendering certain mistakes inoperative. Various legal techniques are employed to achieve this goal. The minutiae of the requirements for avoiding contracts vary, but the principles involved go to the heart of contract law and of the underlying contract theory.
Erroneous assumptions in the making of last wills (testaments) do not raise the same conflict as mistakes in contract making, since it is a one-sided declaration and there is no contracting party to rely on the apparent validity of the acte juridique. The issue here turns on the question whether the terms of the will can be construed in line with the testator’s hypothetical intentions, had he not been misguided by erroneous assumptions. Some civilian systems contain special provisions as to the avoidance of wills, taking up, in principle, and adapting concepts and provisions as set out for mistakes relating to contracts.
For a long time mistake also played a significant role in the doctrines of unjustified enrichment, on the civil law side as well as in the common law, but many civil law systems have discarded this requirement and the common law may follow suit.
A further variation of mistake exists in the form of mistakes of law (as opposed to mistakes of fact), which can arise in all parts of the legal system, including criminal law. The question whether the normal legal consequences of a person’s conduct attach only if he was aware of them arises in many different contexts. Mistakes of law pose an almost jurisprudential problem for the concept of law itself. Does ignorance of law constitute a valid excuse? Continental systems have long distinguished between mistakes of law (errores iuris) and mistakes of fact, although attempts to come up with unifying doctrines, encompassing mistakes both of fact and of law, have been made now and then. English law does not seem to have stressed the difference between mistakes in law and mistakes in fact as stringently as traditional continental doctrines.
2. Historical developments
a) Civilian legal systems
Civil law contract doctrines are based on Roman law, as adopted in continental Europe since the High Middle Ages in the lengthy process of reception. The Roman lawyers of antiquity never developed an all-encompassing doctrine of mistake, applicable to all kinds of contracts and legal acts alike. For them, different types of contract called for different approaches to cases of mistake. The decisive question was whether an erroneous assumption had obstructed the conclusion of a valid contract. A misapprehension either affected the formation of the contract or it was simply inoperative. In the case of the so-called consensual contracts (sale, mandate, partnership and locatio conductio) the only question was whether a misunderstanding had prevented the parties from reaching consensus. Only correspondence mistakes (mutual mistakes) were operative mistakes. The contract was void, not voidable. The English adage ‘error negatives consent’ very aptly reflects the Roman approach. Since cases of mistake were all handled from the viewpoint of consensus, erroneous assumptions could be operative mistakes only if they related to the subject matter of the contract, ie to the very matter on which there had to be consensus. There was no separate definition of mistake. Consequently, where a mistake had prevented consensus from being reached, there was no special regime as to its effects. Since there was simply no contract, there was no time limit for challenging an assumed contract by raising the issue of mistake. The fact that consensus was not reached was not attributed to either of the parties. The mistaken party did not have to compensate the other party. Deliberations as to protecting reasonable expectations were absent and indeed would have been incompatible with the Roman approach.
From the Middle Ages onwards, civil law has systematized types of mistake. The key systematic criterion was the subject matter of the erroneous assumption: did it relate to (1) the type of contract, eg loan instead of gift (error in negotio); (2) did it relate to the object of the contract (error in corpore); (3) or did it relate to the identity of the contracting party (error in persona)? As to the error in substantia/materia see below. It was now essential that a factual assumption could be classified as one of these recognized types of mistake so as to distinguish operative from non-operative mistakes. While true to the Roman law sources, the underlying concept of mistake had undergone a considerable change. Mistake was now taken to be an erroneous assumption and conceptualized as an occurrence ab extra, apart from the consensus, a concept clearly at odds with the Roman beginnings. A mistake could now be operative though purely unilateral, but in order to avoid having any mistake undermine the binding force of the contract, the scope of operative mistakes was limited by the system of types of mistake. An exhaustive list of operative types of mistake meant that other erroneous expectations could not be relied upon as operative mistakes. The inadmissibility of mistakes relating merely to the environment of the contract (mistake relating to one party’s motivation) became a cornerstone of civilian mistake doctrines. Traditional civilian doctrine saw such mistakes as purely personal complaints, expressions of regret or disappointment over the desirability of the agreement. The classic formula was coined by Friedrich Carl von Savigny (1779–1861) who, returning to the Roman law, juxtaposed the ‘genuine’ mistake, a unilateral erroneous assumption generally without legal effect, with the ‘apparent mistake’ which, in preventing consensus, was not identical with a (unilateral) erroneous assumption.
Since Roman times, error in substantia/ materia has been the odd one out. An endless debate, dating back to Roman times, has raged over this type of mistake. Some Roman lawyers argued that a sales contract was void if the object of the sale was fundamentally different in fact from the object envisaged jointly by seller and buyer (the object made of copper was thought to be of gold). Aliud pro alio was the catchphrase: selling an object for something totally different. While Roman law saw mutual mistakes as operative, error in substantia/materia was the one case where a common mistake was considered to be operative. Since this kind of mistake is somehow related to the quality of the object, it became necessary to distinguish this case from simple ‘lack of conformity’, the more so since in Roman law ‘lack of conformity’ could not be the subject matter of mistake, as the consensus of the Roman sales contract did not encompass terms relating to the qualities of the object. In the absence of general remedies in cases of qualitative defects, error in substantia offered an escape at least in the most severe cases. Since error in substantia made an element operative which technically was not part of the consensus, it came to be seen and be used as a starting point for opening the doctrine of mistake to include erroneous assumptions relating to the environment of the contract.
Whereas the effect of mistake in Roman law was simple—there was no contract—subsequent civil law developed a refined set of rules early on which concerned the avoidance of contracts affected by mistake. The key elements were: (1) Special procedures for avoiding contracts, including, for example, the need to notify the other party. When contracts induced by mistake were no longer seen as void, but merely voidable, the question arose whether the avoidance of a contract was to be seen as a rescission, with effect ex tunc, ie retroactively, or whether it merely operated ex nunc. (2) Perhaps most importantly, while the absence of a contract could be invoked at all times and at will, time-limits were introduced for its avoidance. They were unknown to the Roman law of mistake, but ‘imported’ from the Roman law remedies granted in cases of fraud and coercion. (3) While Roman law saw both parties on an equal footing when no agreement had been reached, the concept of mistake as a unilateral misapprehension created a distinction between the party influenced by erroneous assumptions, solely entitled to avoid the contract, and its opponent, deserving protection and/or some sort of compensation due to his reliance on the contract’s validity. Advanced civilian systems employ all these elements in elaborate combinations in order to balance the interests of both parties in a just manner. In the 19th century, a philosophical approach to the concept of consensus, the so-called ‘will-theory’, made the field of mistake its main battleground, stressing that a legal act which did not originate from an unmitigated (mistake-free) will could not be upheld by law. Some took this idea so far as to argue in favour of the operativeness of the reservatio mentalis. The advocates of the Roman law also stressed that mistakes rendered the contract void, not only voidable. These views were countered by arguments (the so-called reliance theory) stressing the need to safeguard contracting partners from finding the security of their transactions imperilled if mistake-based avoidance were too readily available.
b) Common law
The categories employed by the common law are unilateral mistake, mutual mistake and common mistake. Since not every unilateral mistake is operative, common lawyers, just like their civilian counterparts, had to work at distinguishing operative and non-operative mistakes. Mistaken motives do not make for operative mistakes. To constitute an operative mistake at common law the erroneous assumption must affect the fundamental substance of the contract. A fundamental mistake renders the contract void, whereas fraudulent misrepresentation makes the contract voidable (fraud). As on the civil law side, fault was not originally a factor in determining whether a mistake was operative or not, but considerations of fault have crept in over time.
To further systematize the mistakes which can affect the fundamental substance of the contract English lawyers from the 19th century onwards employed categories taken from the civil law (presumably via the writings of Pothier), and hence also speak of mistakes as to the nature or the identity of the object (error in corpore), the very nature of the contract (error in negotio), to a vital term of the contract, or to the identity of the other contracting party (error in persona). Much attention has been paid to mistakes as to identity, using criteria turning on the distinction between the identity and attributes of the other party and whether the transaction was face to face or not. The development in the courts has been accompanied by doctrinal writings and the issue seems to remain unresolved.
Mutual mistakes, where both parties enter into the agreement on different assumptions, can be seen as a lack of consensus which prevents the formation of a contract (‘mistake negatives consent’), unless by way of interpretation the intention of one of the parties can be given precedence over the intention of the other party. In the latter case, only very exceptional circumstances can be relied upon in order to establish an operative mistake, for example where the contract is wholly ambiguous, or the mistake was known to the other party or negligently caused by him. English law thus strongly intertwines issues of interpretation—especially the so-called objective and subjective tests—with the notion of mistake.
If both parties are influenced by the same erroneous assumption and the mistake is fundamental, consensus can be nullified by their common mistake. Here ‘mistake nullifies consent’. This is the approach to the so-called res extincta cases, where the subject matter of the contract does not exist or has ceased to do so. These cases on the borderline between mistake and frustration of contract (involving issues of impossibility/frustration) have also troubled civilian lawyers since Roman times.
It was long thought that a mistake not operative at common law (because not seen as fundamental) could be successfully invoked in equity, if a purely equitable remedy like specific performance was sought. Invoking equity does not lead to a rescission of the contract but to a bar to the enforcement of specific performance. The rule, however, has been called into question; Great Peace Shipping Ltd v Tsavliris Salvage (International) Ltd  EWCA Civ 1407,  QB 679,  4 All ER 689.
If a common mistake arises merely in the way a contract is drafted, equity permits the rectification of an instrument which incorrectly represents the prior consent of the parties, notwithstanding the rule about parol evidence. The divergence of the instrument from what was agreed must be ‘radical’ or ‘substantial’. The rule goes back to the doctrine of non est factum, a defence based on the executed deed having been incorrectly read to the illiterate signatory. Here it is not the contract, but the instrument which is the object of rectification. Rectification is the exception to the rule, universally subscribed to, that an operative mistake does not permit the courts to create a contract free from the mistake but only cancels the contract which has been affected. English law does remedy slips of the tongue (or pen), a core case of civilian mistake doctrine, albeit sometimes on the basis of bad faith in the opposing party, sometimes on the basis of its recognition as a mistake.
If one is ready to generalize very broadly, English law rather stresses the security of transactions, whereas civilian legal systems, perhaps still under the (eroding) impact of so-called will theories, are more generous in allowing the avoidance of contracts on the basis of mistake. To balance this slightly more liberal approach, civilian legal systems rely on mechanisms which, for example, require the party avoiding the contract to compensate the other party (typically for his reliance interest).
3. Recent trends
The national codifications took up civilian contract doctrines as they stood at the time. For centuries, the principle that mistakes as to motive are necessarily inoperative has been challenged again and again in order to allow a more flexible handling of difficult cases. In the absence of significant changes in the doctrines of mistake, ‘neighbouring’ areas of the law have been flourishing. A general tendency seems to favour allowing the ‘cancellation’ of contracts on the basis of mere negligence, whereas this used to be possible only on the basis of (deliberate) fraud. English law has seen an expanding application of misrepresentation and undue influence. Another way of supplementing mistake doctrines perceived as too narrow is to impose liability for the violation of pre-contractual duties (culpa in contrahendo). If damages sought for a violation of pre-contractual duties encompass the undoing of the contract the effect is similar to an actual avoidance.
The civilians also took the doctrine of the clausula rebus sic stantibus (théorie de l'imprévision (French), Wegfall der Geschäftsgrundlage (German)) as a starting point. While the clausula originally dealt with change of circumstances (especially in long-term-contracts), it was stretched so as to apply to fundamental mistakes made already when coming to an agreement. When the Bürgerliches Gesetzbuch (BGB) was partially remodelled in 2002, this doctrine was embraced by the legislature (§ 313(2) German BGB).
Rules set up for the protection of the consumer (consumers and consumer protection law) impose duties to inform. If statutorily required information is withheld, a right of termination may be available independently of the individual consumer’s assumptions (see Arts II.-3.101 ff DCFR).
All in all, the doctrine of mistake, once absolutely central for contract doctrines and theories, has been supplemented by a variety of instruments designed to prevent erroneous assumptions or to remedy them at a later stage. Mistake doctrines have thus been somewhat degraded so as to form but one piece of a larger puzzle.
4. European unification projects
Whereas mistakes affecting last wills may well be dealt with on a national basis for some time yet, unification projects concentrate on the law of contract. The most recent proposals take as a starting point the concept of fundamental mistake, Art II.-7.201 DCFR; Art 4:103 PECL; Art 3.5 UNIDROIT PICC. There are no restrictions as to the subject matter of a mistake: it need not be related at all to the subject matter of the agreement. Avoidance can thus be based on mistakes of any sort: ‘Mistake is an erroneous assumption relating to facts or to law when the contract was concluded’ (Art 3.4 UNIDROIT PICC). To make up for the absence of restrictions regarding the kind or subject matter of the ‘erroneous assumption’ other criteria are employed to limit avoidance on the ground of error. The mistake must concern an element which the opposite party could have recognized as being fundamental; the mistake must have been induced or at least shared (ought to have known) by the other party. (The English lawyer will recognize the criteria used at common law in deciding whether a mutual mistake is operative.) A unilateral mistake undetectable by the other party thus remains irrelevant. There must not be any fault on the side of the party seeking to rely on his mistake. The requirement that but for the mistake the contract would have been concluded on ‘fundamentally different terms’ introduces an elastic clause. Last but not least, the contract cannot be avoided if the risk of the mistake was assumed, or in the circumstances should be borne, by the mistaken party (Art 3.5(2)(b) UNIDROIT PICC). There are no provisions for compensation due to the other party. This is understandable, because the other party must have been involved, one way or another, in the emergence of the mistake. What is striking is the lack of a time limit for avoiding the contract. All in all, many restrictions try to prevent avoidance from becoming dangerously easy, given the open concept of mistake. DCFR and PECL do not take up the idea of allowing avoidance only when the contract is still unperformed (res integra concept), a speciality of Austrian law (§ 871(1) Austrian ABGB), which contains a number of other peculiarities characteristic of natural law. Article 3.5 UNIDROIT PICC, however, does employ the res integra criterion among others.
The DCFR concept of mistake covers both mistake of law and mistake of fact. Mistake of law is not confined to mistakes relating to the legal effects of the agreement. Rather, all sorts of mistakes of law can become relevant, such as mistakes as to the tax effects resulting from the contract and/or its performance. So open a concept tends to render a separate concept of fraud superfluous. Just as a mistake resulting from fraud need not meet any specific requirements in order to be relied upon (see Art 28(1) Swiss OR: a contract can be avoided on the basis of fraud, ‘even if the induced mistake as such was not substantial’), mistake as such is also freed from any requirement as to its subject matter. One could thus dispense with the concept of fraud altogether. Fraud becomes just a ‘special case of “induced” mistake’ (Hein Kötz).
The DCFR does not require the erroneous assumption to be related to the agreement and its terms. The distinction between agreement-related and external mistakes, long cherished in the civilian tradition, is discarded. None of the restrictive criteria employed by the DCFR, the fundamental nature of the mistake, its excusability, or the causation of the mistake, is related to the constitutive elements of the contract. By contrast, the DCFR does not regard the slip of the tongue, a core case of civilian mistake doctrines, as constituting a mistake at all: inaccuracy in the expression or transmission of a statement is acknowledged as a basis for avoiding the contract, but only because it is expressly provided that it is to be treated as if it were a mistake (Art II.-7:202 DCFR; likewise Art 4:104 PECL and Art 3.6 UNIDROIT PICC). We seem to have come full circle: the mistake in motivation (Savigny’s ‘genuine mistake’), long held to be inoperative by civilians and common lawyers alike, now seems to be the central instance of mistake, whose relevance, however, is sharply reduced by making avoidance dependent on a bundle of quite strict requirements. To apply the same strict requirements to slips of the tongue and inaccuracy in transmission seems odd.
While DCRF, PECL and UNIDROIT PICC are rather restrictive in allowing avoidance of contracts due to mistake, there will always remain some ‘misunderstandings’ which prevent consensus. Consensus as the core requirement for the conclusion of contracts (Art II.-1:101 DCFR; Arts 2:101–2:103 PECL) could thus turn out to be the source from which a new doctrine of mistake might spring, since one can hardly ignore the relevance of mistakes which prevent consensus. Mistakes which are relevant due to the requirement of consensus would fall outside Art II.-7:201 f DCFR and thus would not need to meet the elaborate restrictions in this provision which narrow down the avoidability of contracts.
Werner Flume, ‘Irrtum und Rechtsgeschäft im römischen Recht’ in Festschrift Fritz Schulz, vol 1 (1951) 209; Samuel Stoljar, Mistake and Misrepresentation (1968); Jacques Ghestin, La notion d’erreur dans le droit positif actuel (2nd edn, 1971); Vittorino Pietrobon, Errore, volontà ed affidamento nel negozio giuridio (2nd edn, 1990); Ernst A Kramer, Der Irrtum beim Vertragsschluss. Eine weltweit rechtsvergleichende Bestandsaufnahme (1998); David J Ibbetson, A Historical Introduction to the Law of Obligations (1999) 210, 225 ff, 289 ff; Martin J Schermaier, Die Bestimmung des wesentlichen Irrtums von den Glossatoren bis zum BGB (2002); Ruth Sefton-Green (ed), Mistake, Fraud and Pre-Contractual Duties to Inform in European Contract Law (2005); Wolfgang Ernst, ‘Irrtum—Ein Streifzug durch die Rechtsgeschichte’ in Reinhard Zimmermann (ed), Störungen der Willensbildung bei Vertragsschluss (2007) 1; Catharine MacMillan, Mistakes in Contract Law (2010).