by Gerhard Dannemann1
As a defence, illegality implies that the plaintiff’s own illegal or immoral conduct can defeat a claim in unjust enrichment which would otherwise lie. Whether or not illegality can also serve as a ground for restitution is a different question which is treated elsewhere in this publication.2 The present article will furthermore concentrate on situations where both parties are responsible for the illegality, as one-sided illegality will frequently not render a contract void,3 with the effect that no action will lie in the first place because the enrichment was supported by a causa (in Civilian terminology), or by consideration (in the language of the Common Law).
There is, interestingly, no divide between Common Law and Civil Law as regards the illegality defence. This has largely to do with the fact that this defence can be traced back to Roman law in all legal systems under consideration.4
It could be argued that, next to change of position, illegality is the only other defence which is specific to unjust enrichment claims and which does not depend on how grounds of restitution are structured.5 For in many other cases, one system’s ground of restitution, put in the negative, is the other’s defence against such a claim. Three examples should suffice:
1. If one party intentionally enriches another party, mistake will be a ground of restitution in the Common Law.6 Most civilian systems will grant restitution if the enrichment is not supported by a legal cause. But lack of mistake as concerns such a causa will then operate as a defence, as e.g. in § 814 of the German Civil Code (Bürgerliches Gesetzbuch, BGB.)7
2. A similar point can be made for the defence of bona fide purchase for value.8 This Common Law defence operates where the plaintiff has enriched a first recipient, who then sells this enrichment to the defendant. The defence of bona fide purchase for value serves to protect the reliance which the defendant has placed in his or her contract with the previous recipient of the enrichment. In German law, the concept of performance, which identifies both plaintiff and defendant to an action, serves to restrict the wide general clause and effectively keeps unjust enrichment claims within failed contractual relationships. Thus, a plaintiff must usually sue the first recipient of the enrichment rather than a party who subsequently acquired the enrichment bona fide.9 Again, the main exception which German law allows is tell-tale. In a number of situations, a plaintiff can jump outside the failed contractual relationship and recover from a third party if this third party has acquired the initial enrichment without providing value in return.10 The English counter-defence that the bona fide acquisition was not for value thus becomes, in German law, an exceptional ground of restitution against a third party.
3. If an enrichment was imposed on the owner against his or her will, this can amount to a defence in German law (aufgedrängte Bereicherung).11 However, if the same cases were to be decided under Common Law rules, there would be either no ground of restitution to start with, or no initial enrichment under the rules of subjective valuation.12 The rationale behind this German defence is served by restrictions on the grounds of restitution in the Common Law.
Both Civil and Common Law have some other defences against unjust enrichment claims which, however, are doubtful, of very limited relevance, or not particularly enrichment related; these can be ignored for the purpose of the present article.13
The following example might help to illustrate the rationale which the illegality defence is meant to serve. An instigator pays £1,000 to a thug who in return beats up a victim. The instigator then seeks to recover his payment in unjust enrichment. English, German, and most other laws will agree that this claim must be disallowed. For this, we are normally given one or several of the following reasons: (1) No one will be allowed to found his action on his own illegal conduct.14 (2) Courts would be tainted if they were to assist one villain in his claim against the other.15 (3) The claim must be disallowed in order to deter from, or even punish, illegal or immoral conduct.16
The same victim now sues the thug in tort for injuries suffered. In turn, the thug sues the instigator for contribution to his tortious liability towards the victim. This action can only succeed if the thug can show that he committed the tort in execution of his agreement with the instigator. The thug must therefore plead his own illegal conduct. Furthermore, in allowing the thug’s action, the courts will assist one villain against the other. Additionally, by reducing the thug’s overall liability, contribution equally diminishes the deterrent effect of his tortious liability. So all the above reasons why illegality can be raised against a claim in unjust enrichment should prevent a joint tortfeasor’s claim in contribution. Yet English, German and most other laws agree that the action for contribution must be allowed.17
It looks therefore as if our three explanations of the illegality defence do not stand up to scrutiny. At the very least, they are formulated too widely, so that there must be something which makes these arguments work in an enrichment environment, but not in tort. At the same time, these arguments must be of fundamental importance, as the very purpose of the law of unjust enrichment is to prevent losses from lying where they fall.
It has been suggested before that, when assessing the merit of the illegality defence, it is of paramount importance to keep an eye on the policy rationale behind those rules which make a particular contract, performance or transaction illegal.18 I further suggest that this rationale might help us in reducing the above three general principles to their useful role for justifying illegality as a defence against an unjust enrichment claim.
The reason why a party should not be allowed to rely on its own illegal conduct is that people should not be rewarded for their own illicit behaviour. This is why an unpaid thug cannot recover a quantum meruit from the instigator, but why the same unpaid thug could still sue the instigator for contribution to his tortious liability towards the victim.
The second argument of the dignity of the courts preventing them from assisting one villain against the other has, in my view, been inflated out of proportion. If this was true, courts would either have a terrible reputation, or else a rather low case load. The law is not against divorcing couples from hell, does not prevent lawsuits between neighbours from hell, and does not prohibit litigation between legacy hunters or between rogues who fiddle company mergers to their own advantage. What could taint courts, though, is if they were forced to allow an action the success of which offends acknowledged legal or public policy. To allow an action for a quantum meruit for having beaten up a victim would have the same effect as declaring that the agreement between the thug and the instigator is a valid contract. This is why courts would indeed endanger their reputation if they allowed such an action. To allow an action by the thug against the instigator for contribution, on the other hand, does not counteract the policy which makes such agreements void.
Finally, the deterrence argument needs some fine tuning. In any lawsuit, whoever wins will feel encouraged, and whoever loses will feel disencouraged. If both parties are to blame, we may have to consider carefully which party, if any, needs to be deterred more. It is easy to state in a lawsuit between two villains that the action must be disallowed because the plaintiff needs to be deterred from illegal conduct – just as easy, in fact, as it is to state the exact opposite, namely that the action must be allowed because the defendant needs to be deterred. Where both parties need to be deterred equally, deterrence as justification of either outcome becomes meaningless. The tortious action for contribution, on the other hand, distributes the deterrence between the culprits, as both are made to pay for the consequences of their wrongdoing.
Let us now take a look at the mechanisms which various unjust enrichment laws employ to operate the illegality defence. I will deal in turn with the nemo auditur approach, the in pari delicto approach, and the discretionary approach.
The first mechanism is procedural and, technically speaking, not a defence but a limitation in making a claim. A plaintiff can only succeed if he or she can present a claim on the basis of facts which do not include the plaintiff’s illegal or immoral conduct – nemo auditur turpitudinem suam allegans.19 This is the position of English law from Holman v Johnson to Tinsley v Milligan,20 and commonly associated with the general maxim of ex turpi causa non oritur actio.21 But the same rule has, in a much more general form, entered the reformed Czech and Slovak Civil Code of 1992.22 It is not, however, the main basis of the illegality defence in German law.23
This mechanism has two main difficulties.24 First and foremost, it can counteract the very policy which makes an agreement between parties illegal. In Tribe v Tribe,25 a father transferred his assets to his son as the father’s creditors were closing in. The father’s claim against his son for restitution of his assets would have been barred by the illegality defence, had it not been for the fact that all creditors were paid and the father considered to have withdrawn from the illegal purpose. I would argue that by allowing the illegality defence in cases involving concealed assets, these assets will be even further removed from the reach of the creditors. This might put the seal on the very situation which the law wants to prevent.26
Second, this rule can produce rather accidental results.27 In a recent German case, the plaintiff was a company which had first prepared and then audited the defendant’s accounts in violation of a statute which requires complete separation of these functions.28 Such a plaintiff is able to plead its case for a quantum meruit for the auditing without the slightest reference to their previous work on the accounts. In this case, the plaintiff, as the professional auditors, seem more blameworthy than their clients. Should this be of no relevance, and should the decision really hang on the fact that the prohibitory norm contains a requirement (previous preparation of accounts) which is not a necessary element in plaintiff’s pleading? The best thing to be said about the nemo auditor rule is probably that it will more often lead to the right than to the wrong result, but that is no praise for a rule of law, which additionally has the somewhat unfortunate inclination towards favouring the more clever amongst two villains.
The second mechanism for the illegality defence has two names, both of which are in Latin. In English Latin, it reads in pari delicto potior es conditio defendentis, and in German Latin in pari turpitudine melior est causa possidentis.29 Two villains sue each other over an illegal or immoral deal: losses will lie where they have fallen. English law recognises this defence next to the nemo auditur rule.30 It is also contained in § 817 sent. 2 of the German Civil Code (Bürgerliches Gesetzbuch, BGB), but limited to situations where the enrichment was given for a purpose which violated a statutory prohibition or offended good morals. Courts will also require that the claimant was aware of the illegality.31 The Austrian and Italian Civil Codes and the Swiss Code of Obligations contain similar rules.32 The German provision is surrounded by many disputes and uncertainties. It has been called “one of the most dreaded perils in the sea of legal doctrine”.33 I will limit my presentation to some aspects which are of particular relevance in the comparative context.
We will first look at cases where one party performed for the sake of a counter-performance which never came about. Within this category, we can further distinguish (a) cases where the claim is for a quantum meruit, as in the auditor’s case, (b) for restitution of money paid in expectancy of a counter-performance under an illegal or immoral agreement, and (c) for restitution where a performance can be returned in kind.
Two more cases may serve to illustrate quantum meruit claims. The first concerned payment for a series of advertisements for what was described as “Die schönen Stunden zu zweit” (Those lovely hours entre deux), garnished with the additions “just call me”, “habla español” and “parla italiano”.34 The reader may already have guessed the nature of the services which were offered this way. The judgment by the Bundesgerichtshof, anxious to avoid any doubt on this point, illuminates the reader with a graphic reproduction of the offending article and probably makes legal history by placing this particular type of advertisement within an official case reporter. The court explains at some length why this was indeed an advertisement for prostitution and thus amounted to a petty criminal offence, and why the contract was void in consequence. We find nearly three pages on why the defendant’s reliance on illegality against the claim for the agreed prize did not offend good faith, and only one paragraph on the defence in § 817 sent. 2 BGB, which was held to apply both in word and in spirit, in particular in order to deter publishers from offending. (Deterrence of the client who ran the brothel was apparently not an issue.)
Both the auditor and the advertisement cases were preceded by a much more controversial decision which concerned a builder who claimed a quantum meruit for a job which parties had agreed to keep secret from the taxman.35 The court held that this action should, in principle, be defeated by the in pari turpitudine rule, but set this rule aside on the grounds of good faith. It is somewhat puzzling that the Court held honour amongst thieves to be stronger than a provision of the Civil Code.36 More particularly, however, the Court argued that denying the claim was not really necessary to enforce the relevant tax and social legislation, and that it was unjust that the client, who would normally have the stronger bargaining position, should be allowed to keep the windfall. One wonders, of course, whether the same rationale would not apply to the auditor and the publisher in the other two cases, but the Bundesgerichtshof ruled in each of these cases that the builder was to be distinguished, though, unfortunately, without indicating any reasons for this distinction.
We now turn towards cases where one party has paid in advance for an illegal performance. One couple paid DM 15,000 under an illegal adoption agreement whereby the Defendant was to procure, within 2½ weeks, an unnamed child from the Philippines.37 The Oberlandesgericht Oldenburg barely brushed good faith and the purpose of the prohibitory norm to arrive at the conclusion that courts are not meant to sort out the consequences where both parties have acted immorally or unlawfully. And in a mirror case to Parkinson v College of Ambulance Ltd and Harrison,38 the Bundesgerichtshof held that a plaintiff who had paid US$ 50,000 in order to obtain the title of honorary consul of Sierra Leone was barred from recovering when the title failed to come through.39
With the exception of the builder case, it is noteworthy that all these judgments devote comparatively little space to the in pari turpitudine rule itself. They rush to their conclusion as if they were a little embarrassed to deny a claim on such a tedious ground.40 It is also noteworthy that good faith, purpose of the prohibitory norm, and reliance on one’s own illegal conduct are used to explain each other, and rather interchangeably. I do believe, though, that a common rationale can, after all, be found behind those judgments.
I think that the main argument flows indeed from the prohibitory norm, and it is that no claim should lie for the quantum meruit of an act which is in itself illegal. For otherwise restitution would ensure that illegal conduct pays. This clearly explains the advertisement case, but also the auditor case, as the purpose of the statute was to keep the functions of accountants and auditors completely separate. Similarly no claim should lie for the quantum meruit of an act which, although not illegal in itself, should not be linked with a counter-performance in money – because this would enforce via restitution the very link which the prohibitory norm tries to prevent. This could explain why prostitutes or merchants in knighthoods are not allowed to claim a quantum meruit for their services.
On the other hand, the builder’s work was not in itself illegal. Neither was it illegal to build against payment. What was illegal was that no tax or social insurance contributions should be paid from these earnings. Going back to the wording (and perhaps the history)41 of § 817 sent. 2 BGB, one could also wonder whether in this case performance was indeed made for an illegal or immoral purpose.
Neither does the rationale apply to performances which can be returned. On the contrary, allowing a restitutionary claim in kind may be the best or even only way of enforcing the policy rationales which may prohibit certain currency deals, restrict trade with national heritage objects, endangered species, toxic substances, drugs or arms. The same is true for agreements for the illegal concealment of assets. This is where an application of the in pari turpitudine rule is likely to produce the most disastrous results.
Can the same argument be applied to the recovery of money paid in anticipation of counter-performance? In principle, allowing restitution should remove any unwanted link between the illicit act and payment. But allowing recovery would favour the illicit client over the illicit contractor. Those who create demand for illicit acts are, generally speaking, neither less dangerous nor less blameworthy than those who commit them. The person who orders a murder deserves no better treatment than a contract killer. I wonder whether the illegality defence is not used in this context to create a limited amount of equality between those who commit and those who pay for illegal acts. It is the intended link between payment and counter-performance which could ultimately justify the application of the illegality defence in these cases. It is also, arguably, this link which Roman law had in mind when it constructed this particular defence as a bar to a condictio ob rem.42 However, these cases show perhaps more than any others that illegality is a difficult rule, and that there may be cases where a fully satisfactory result is not feasible. It is noteworthy that the new Dutch Civil Code will generally allow the recovery of money paid in advance under an illegal contract, but not recovery for a quantum meruit which should not be valued in money.43
There is one particular problem with the in pari turpitudine approach, which concerns enrichments which were meant to be temporary by both parties, and particular loans or securities. Under a narrow application of nemo auditur, there can be recovery after the stipulated time has elapsed, because the illegal purpose of the loan or security need not to be pleaded. However, the in pari turpitudine approach creates some difficulties, as this defence could preclude any recovery of a loan or security given for an illegal purpose. For loans, German courts have frequently avoided this result by ruling that it is not the money loaned itself, but rather the temporary use of this money which constitutes the enrichment, so that § 817 sent. 2 BGB would not prevent recovery after the stipulated time for the loan has elapsed.44
An equally difficult application of the in pari turpitudine rule relates to void, but fully executed agreements. For denying restitution has essentially the same effects as ruling that the contract is valid.45 But one must also enquire whether undoing the contract will serve the prohibitory norm better than keeping matters as they are. Again, where restitution in kind is possible without further illegality or violation of interests or rights of third parties, this may well be a better way of enforcing the prohibitory rule than allowing the illegality defence.
On the other hand, illicit work or services can normally not just be returned. Putting a price tag on them gives us the same problems which, in my view, can justify the illegality defence in the first place. Not allowing a quantum meruit but allowing restitution for money paid leads to the same imbalance between those who commit illegal acts and those who instigate them with their payment. This may explain why in some situations the prohibitory norm can be served best by leaving both enrichments where they are.
Next to the nemo auditur and in pari delicto rules, one can observe a third approach which has been chosen in particular by legislation in New Zealand and Israel, namely to leave the decision as to whether or not the illegality defence should apply to the courts to be decided on the merits of each case. Both statutes give very broad discretion.46 The New Zealand legislation invites the courts specifically to consider the conduct of the parties, the object of any statute breached and the gravity of the penalty provided, and any other matter which the court thinks proper, but to refrain from granting relief it this would not be in the public interest.47
To a certain extent, both German and Dutch law have taken a discretionary approach. Dutch law has essentially limited the illegality defence to cases where a performance cannot be returned, but leaves the decision whether this performance ought not to be valued in money to the courts, and makes restitution of any counter-performance in a mutually executed transaction subject to reason and equity.48 In addition, the Hoge Raad (the highest Dutch court) has held that unjust enrichment claims are, quite generally, subject to the general good faith provision in Art. 6:2 BW.49 Modern German law should be named in this context because, as the illicit labour case demonstrates, German courts are forever prepared to set aside fairly clearly worded rules on grounds of equity and good faith.50 Indeed, Reuter and Martinek’s influential book urges German courts to develop their own case law by deciding each case on its merits, using the purpose of the prohibitory norm as the guiding principle.51 Finally, the public conscience test developed by Nicholls LJ in the Court of Appeal’s judgment in Tinsley v Milligan is a value judgment based on the adverse consequences of refusing or granting restitutionary relief. This attempt to introduce a discretionary approach to English case law was, however, rejected by the House of Lords in the same case.52
The obvious advantage of a discretionary approach is that it enables courts to avoid results which are unjust to one party or unwise as a matter of legal policy. But its obvious drawback is the legal uncertainty which it creates. On the other hand, this uncertainty needs to be placed in its context. While I would not go as far as stating that legal uncertainty is an appropriate response to parties who place themselves outside of the legal order, one might still feel less concerned about this uncertainty than in many other cases where e.g. English, German or French law have not hesitated to adopt statutory provisions of almost scandalous vagueness.53
I have some sympathies for the Dutch approach. It does not defeat a claim for restitution in kind where it is possible. And it limits the discretion to two fairly precise issues, namely: (1) If a performance under an illegal agreement cannot be returned, ought it to be valued in money? (2) If an illegal agreement has been fully performed by both parties, should restitution to be allowed? There is mainly one point in the Dutch approach where I have my doubts. For it seems to allow, quite generally, those who instigate illegal activity by payment to recover as long as the illegal activity itself has not been performed.54 And while Art. 6:2 BW is available as an emergency brake even in this situation, its use will be hampered by the fact that the Hoge Raad has so far refused to balance the respective wrongdoings of the parties concerned.55
Here, as generally when discretion is to be used within the illegality defence, I suggest the first and foremost consideration should be whether either allowing or refusing restitutionary relief would create, maintain, or prevent the very situation which the prohibitory norm wants to avoid. Only if this does not provide a clear indication for the outcome, notions of deterrence can help to find the right solution. This requires in particular a balancing of which party needs to be deterred more. The relevant factors include the culpability of both parties, any actual harm created by, or danger generally associated with, their behaviour. This will still leave a few situations, in particular as concerns recovery of an advance payment for illicit conduct, where this test fails to produce an unequivocal result. Various ideas have been voiced, all of which present their own difficulties: that the money should be taken from the defendant and given to charity56 or forfeited;57 or that apportionment might be better than an all-or-nothing approach.58
I would like to float one more idea. If the policy reasons for defeating and allowing the action hold the balance, perhaps one should deny recovery of advance payments made for an illegal purpose, but make the defendant pay for the costs of the action for recovery in consideration of the facts that the action does not fail because the defendant has a right to keep the money, and that the defendant has also been culpable of illegal conduct.59 This would at least ensure that both parties lose something. We may find just a hint of such an approach in Parkinson v College of Ambulance Ltd and Harrison.60 In this case, an action for recovery of a donation made with the express purpose of obtaining a knighthood was defeated on grounds of illegality. The Court observed that the second defendant had denied fraud, had administered cross-examination to the plaintiff, and had refused to be examined himself. In the view of the Court, all this amounted to oppressive behaviour. Accordingly, an order was made to deprive him of his costs.
1 Dr. jur.; Fellow, Worcester College, Oxford. I am grateful to Jeroen Kortmann, who has helped me much in my attempts to understand the Dutch law of unjust enrichment.
3 If only one party has violated a statutory provision, a contract will normally not be void: BGH 2.9.1983, NJW 1984, 230; for an English translation, see Basil Markesinis, Werner Lorenz and Gerhard Dannemann The German Law of Obligations, Vol. I: The Law of Contracts and Restitution (1997), case 29; see also ibid pp. 178ss; similar for English law, Phoenix General Insurance co. of Greece S.A. v Administratia Asigurarilor de Stat  2 All ER 152, at 176 (Kerr LJ). Transgressions of legal or moral norms committed by one party only will normally make a contract void only where this is necessary to protect the other party or a third party, and will thus call for restitutionary solutions which reinforce this protection. See generally Markesinis, Lorenz and Dannemann Ch. 3 (1).
4 This is not surprising for Continental Roman law based systems. In English law, Roman law influence appears particularly from Lord Mansfield’s speech in Holman v Johnson  1 Cowp 341 = 98 ER 1120, which contains the nemo auditur turpitudinem suam allegans rule in nearly literal English translation, and where further references are made to ex dolo malo non oritur actio and the rule that potior est conditio defendentis if both parties are equally to blame.
5 Even change of position and illegality can be phrased as elements of a ground for unjust enrichment, namely if surviving enrichment is understood to be the primary object of an unjust enrichment claim, or if illegality is constructed through the nemo auditur rule (see below). On the other hand, it is still the defendant who has the show that the initial enrichment has not survived, and even the nemo auditur rule is applied only to defeat claims which would otherwise lie.
6 See, generally, Peter Birks, An Introduction to the Law of Restitution, 1989 paperback ed. pp. 146ss; Andrew Burrows, The Law of Restitution (1993) ch. 3; Goff and Jones, The Law of Restitution, 5th ed. 1998 by Gareth Jones, ch. 4-9.
7 See, generally, Markesinis, Lorenz and Dannemann (note 3), pp. 736ss.
8 Birks (note 6) pp. 439ss; Burrows (note 6) pp. 472ss; Goff and Jones (note 6) ch. 41.
9 BGH 31.10.1963, BGHZ 40, 272 = Markesinis, Lorenz and Dannemann (note 3), case 134. However, this applies to performance based restitution claims only. A plaintiff claiming under restitution for wrongs (or Eingriffskondiktion) can sue a defendant who has acquired the enrichment from the interferer bona fide and for value, as long as title has not passed to this defendant. See BGH 11.1.1971, BGHZ 55, 176 = Markesinis, Lorenz and Dannemann, case no. 133. See generally, Markesinis Lorenz and Dannemann, pp. 731ss.
10 §§ 816, 822 BGB. See Markesinis, Lorenz and Dannemann (note 3) pp. 748-9 and 766.
11 The leading case is BGH 21.12.1956, BGHZ 23, 61 = Markesinis, Lorenz and Dannemann (note 3), case 141.
12 In BGHZ 23, 61, the plaintiff had erected a building on his landlord’s property in full knowledge that he was not entitled to do so; there would be no ground of restitution for his claim for a quantum meruit in the Common Law. See Markesinis, Lorenz and Dannemann (note 3) p. 762; for subjective valuation, see Birks (note 6) pp. 109-114 and Falcke v Scottish Imperial Insurance Co. (1886) 34 Ch.D. 234, 248.
13 These include in particular: (1) § 814 2nd. alt. BGB (plaintiff barred from recovery because of moral (as opposed to legal) obligation towards defendant – this could become a defence in English law as well now that Kleinwort Benson v Birmingham CC and Lincoln CC  3 W.L.R. 1095 has paved the way for restitution based on mistake of law), (2) passing on in the Common Law world (which could be described as being on its way out after Commissioner of State Revenue v Royal Insurance Australia Ltd (1994) 126 ALR 1, at 11-18; Mason CJ; Kleinwort Benson v Birmingham CC  3 W.L.R. 1139, at 1149; see Peter Birks, The Law of Restitution at the End of an Epoch, 28 Univ. of West. Austr. Law Rev (1999), p. 21 note 18; (3) impossibility of counter-restitution (becoming less important and also doubtful whether this a defence (Birks, (note 6), p. 415), or whether the reverse is part of a ground of restitution (Burrows (note 6), p. 420); (4) estoppel, as this is much more closely linked with change of position, which is not the subject of the present article; (5) incapacity, as – from a comparative view – this appears as a particular reason for making a contract void and thus to be related to grounds of restitution.
14 Nemo auditur turpitudinem suam allegans; see below, at III A. A claim must not be founded on illegality.
15 Reinhard Zimmermann, The Law of Obligations. Roman Foundations of the Civilan Tradition (1990), 1996 paperback ed., p. 846; OGH 10.2.1950, OGHZ 4, 57, at 60; Burrows p. 463; Andrew Tettenborn, Law of Restitution in England and Ireland, 2nd ed. 1996 p. 257. See also Konrad Zweigert and Hein Kötz, An Introduction to Comparative Law, translated by T. Weir, 3rd ed. 1998 p. 576.
16 RG 8.11.1922, RGZ 105, 270, 271-2 argued that the defence was intended as a punishment (and justified its operation even against a plaintiff who was of unsound mind and thus without contractual capacity, provided that the defendant could still be liable in tort under § 827 BGB); BGH 31.1.1963, BGHZ 39, 87, 91 (“Strafcharacter”); see Zweigert and Kötz (note 15) p. 576.
17 German law: The instigator and the thug are considered joint tortfeasors under §§ 830, 840 BGB; whoever compensates the victim can sue any other joint tortfeasor for contribution using the victim’s claim, which is assigned to the compensating tortfeasor by operation of the law under § 426 BGB. English law: Sec. (1) and (2) Civil Liability (Contribution) Act 1978.
18 Dieter Fabricius, Einschränkung der Anwendung des § 817 S. 2 BGB durch den Zweck des Verbotsgesetzes?, Juristenzeitung 1963, 85-91; similar Dieter Reuter/Michael Martinek, Ungerechtfertigte Bereicherung (1983) (Handbuch des Schuldrechts in Einzeldarstellungen, hrsg. von J. Gernhuber, Bd. 4), p. 209.; Detlef König, Empfiehlt es sich, das Bereicherungsrecht im Hinblick auf seine Weiterentwicklung in Rechtsprechung und Lehre durch den Gesetzgebern eu zu ordnen?, in: Gutachten und Vorschläge zur Überarbeitung des Schuldrechts, hrsg. vom Bundesministerium der Justiz, Vol. II (1981), pp. 1515-1590, p. 1522 (§ 1.1 (2) (d) of a proposed legal reform act), p. 1542; J. v. Staudinger’s Kommentar zum bürgerlichen Gesetzbuch, §§ 812-822, 13th ed. 1994 by W. Lorenz, § 817 note 2.
19 See Zimmermann (note 15), p. 865 note 196.
20 Holman v Johnson  1 Cowp 341 = 98 ER 1120, at 1121. Tinsley v Milligan  A.C. 340, 354 (Lord Goff), 376 (Lord Browne-Wilkinson).
21 See e.g. Graham Virgo, The Effect of Illegality on Claims for Restitution in English Law, in: W. Swadling (ed.), The Limits of Restitutionary Claims: a Comparative Analysis (1997), pp. 141-185.
22 Section 40a sent. 2 Obcanský zákoník: “A person who has caused a juridical act to be invalid may not raise the issue of invalidity.” Translation taken from: The Civil Code (“Obcanský zákoník”), translated by Trade Links, Trade Links, Prague 1993.
23 But see Reuter/Martinek (note 18) pp. 205 and 214.
25 Tribe v Tribe  3 WLR 913.
26 See also below III B 1 for further examples.
27 Similar Asser-Hartkamp 4-III “De Verbintenis uit de Wet” (1998), para. 346.
28 BGH 30.4.1992, NJW 1992, 2021 (violation of § 319 subs. (2) no. 5, subs. (3) no. 2 of the Commercial Code (Handelsgesetzbuch, HGB).
29 This should imply that English law is concerned with defendants in actions concerning illegality, whereas German law caters for those who are in possession of an enrichment obtained through an immoral transaction. In fact, however, no such distinctions are being attributed to the different choice of words in either law.
30 Holman v Johnson  1 Cowp 341 = 98 ER 1120. Tinsley v Milligan  A.C. 340, 354-5; Virgo (note 21); Birks (note 6) pp. 424ss; Burrows (note 6) pp. 461ss.
31 RG 30.6.1939, RGZ 161, 57.
32 Austria: § 1174 Allgemeines bürgerliches Gesetzbuch; Italy: Art. 2035 Codice civile; Switzerland: Art. 66 Obligationenrecht. Arguably, the Swiss Code of Obligations contains the best worded version of this rule.
33 Zimmermann (note 15) p. 864; similar Reuter/Martinek (note 18) p. 199.
34 BGH 5.5.1992, BGHZ 118, 182.
35 BGH 31.1.1990, BGHZ 111, 308, 312; English translation in Markesinis, Lorenz and Dannemann (note 3) case no. 136.
36 The judgment has been criticized as being contra legem by several authors, e.g. J. von Staudinger’s Kommentar zum Bürgerlichen Gesetzbuch, §§ 812-822, 13th ed. 1994 by Werner Lorenz, § 817 note 10; Larenz/Canaris, Lehrbuch des Schuldrechts, 2. Band, Besonderer Teil, 2. Halbband, 13th ed. 1994, § 68 III 3 g.
37 OLG Oldenburg 13.12.1990, NJW 1991, 2216.
38 Parkinson v College of Ambulance Ltd and Harrison  2 KB 1.
39 BGH 5.10.1993, NJW 1994, 187. Unlike in Parkinson, though, the recipient was not a charitable institution so that the payment could hardly be classified as a gift.
40 The same is not true for OGH 10.2.1950, OGHZ 4, 57, at 60ff, which contains a long discussion of § 817 sent. 2 BGB.
41 For the history of the in pari turpitudine rule, see Zimmermann (note 15) pp. 846-7.
42 Zimmermann (note 15) pp. 846-7 and 863-5.
43 Art. 6:211 subs. (1). Burgerlijk Wetboek (Dutch Civil Code, BW); Eltjo Schrage, Restitution in the new Dutch Civil Code,  Rest. Law Rev. 208-221, 214. See also below at note 49.
44 However, this concerns mainly usury cases, where only one party (the loan shark) has acted illegally or immorally (see RG 30.6.1939, RGZ 161, 52, 57; Reuter/Martinek (note 18) p. 216; Staudinger-Lorenz (note 18) § 817 note 12). A different solution was found in BGH 15.3.1990, ZIP 1990, 915, concerning a bank loan given in the full knowledge that it would be used to finance a brothel ship in the Mediterranean. It helped that the money went through a probablybona fide middlewoman and that § 817 sent. 2 BGB contains a loophole. It does not provide that the illegality defence will also operate if the plaintiff, but not the defendant acted immorally. The bank was thus awarded the unjustified enrichment of the middlewoman, which consisted in her claim against the operator-to-be of the brothel ship. See Staudinger-Lorenz (note 18) § 817 note 11. Lorenz argues that not allowing restitution in this case would have legalised a situation which is frowned upon by the legal order.
45 Similar Zimmermann (note 15) p. 864.
46 Sec. 31 Israeli Contracts (General Part) Law 1973, whereby courts may relieve a party from restitution claims under sec. 21 in cases involving contracts which are void on the grounds of illegality or immoraligy (sec. 30) “if it deems just to do so and on such conditions as it sees fit”. Quoted from (and see): Nili Cohen, Illegality: the Case for Discretion, in: Swadling (ed.), The Limits of Restitutionary Claims: a Comparative Analysis (1997), pp. 186-211, 188.
47 The Illegal Contracts Act 1970, sec. 7 subs. (3):
(3) In considering whether to grant relief under subsection (1) of this section the Court shall have regard to-
(a) The conduct of the parties; and
(b) In the case of a breach of an enactment, the object of the enactment and the gravity of the penalty expressly provided for any breach thereof; and
(c) Such other matters as it thinks proper;
but shall not grant relief if it considers that to do so would not be in the public interest.
48 Schrage (note 43) p. 214; Art. 6:211 BW provides that the claim for the counter-performance “is also barred so far as it would offend reason and equity”. Translation taken from:  RLR 202-207, at 204. (Name of translator not indicated.)
49 HR 28.6.1991, NJ 1992, no. 787, in a case involving an illegal form of temporary employment business: claim for quantum meruit for work performed denied. The Hoge Raad held (at 3.3) that unjust enrichment claims could generally be denied on grounds of reason and equity (redelijkheid en billijkheid) under Art. 6:2 (2) BW. See also Asser-Hartkamp 4-III “De Verbintenis uit de Wet” (1998) no. 348, and Hugo J van Kooten, Artikel 6:211 en de Engelse law of restitution, Ars Aequi 43 (1994), pp. 311-320.
50 It is noteworthy, though, that any element of discretion or balancing of culpability was rejected by OGH 10.2.1950, OGHZ 4, 57, at 60ff. This was an action for return of the purchase price of a radio which was bought on the fraudulent representation of being brand new; the buyer’s action failed because both parties had violated a maximum price regulation in force at the time.
51 Reuter/Martinek (note 18) pp. 210-211.
52 Tinsley v Milligan  Chan. 310, 319 (C.A.), per Nicholls L.J.;  A.C. 340 Lord Goff p. 358-363; Lord Browne-Wilkinson p. 369.
53 English law has, inter alia, a statutory provision which, quite simply, makes it an offence to publish any material which tends to “deprave and corrupt ” readers (Obscene Publications Act 1959, sec. 1,(1), sec. 2). British law requires for the purpose of naturalisation as a British citizen, that the applicant “is of good character” (British Nationality Act, Schedule 1 Sec. 1 (1) (b). French law provides general tortious liability for all damage caused to another by fault (Art. 1382 Code civil). Within the German law of obligations, about any desired result can be achieved via the good faith provision in § 242 BGB as applied by German courts, see Markesinis, Lorenz and Dannemann (note 3) ch. 7.
54 For Asser-Hartkamp (note 49) para. 348, this is a welcome consequence of Art. 6:211 BW.
55 HR 28.6.1991, NJ 1992, no. 787. But see also Kooten (note 49) p. 315, who believes that Art. 6:211 BW permits courts to allow or deny a claim on a balance of the wrongdoing of each party.
56 Zweigert and Kötz (note 15) p. 576s trace this idea back to Thomas Aquinas.
57 I 16 §§ 173 General Land Law of Prussia; Zweigert and Kötz (note 15) p. 577.
58 Some of the wording of the judgment of the Bundesgerichtshof in the builder case (BGH 31.1.1990, BGHZ 111, 308, 312; English translation in Markesinis, Lorenz and Dannemann (note 3) case no. 136; suggests that the deductions which the Court imposed on the builder’s claim for a quantum meruit amount to partial recovery. The Court stated (p. 313) that the enrichment claim “must at any rate be subject to certain restrictions”. On the other hand, on p. 314 the “very considerable deductions” which need to be made are dressed up as reflecting the lower economic value of an illicit labour job. Perhaps the Court was equally anxious to avoid either full recovery in the given case, or the official introduction of partial recovery into the law of restitution.
59 Additionally, the defendant could be allowed to avoid being burdened with the costs of the proceedings e.g. by paying the sum into court on first demand. Depending on the merits of the case, the judgment could then award this sum to the plaintiff, have it returned to the defendant, or give the money to a charitable purpose. To which extent such mixed substantive and procedural law solutions are feasible does, of course, depend much on the applicable procedural law.
60 Parkinson v College of Ambulance Ltd and Harrison  2 K.B. 1, at p. 17.
© 2000 G. Dannemann. This HTML edition © 2000 University of Oxford.
This research has been supported by the network ‘Common Principles of European Private Law’, within the Training and Mobility of Researchers project (TMR) funded by the European Commission (Contract No ERB FMRX-CT97-0118), administered from Münster by Professor Dr Reinhard Schulze. The partner universities are: The University of Oxford (UK), Universidad de Barcelona (Spain), Humboldt-Universität zu Berlin (Germany), Université Jean Moulin Lyon III (France), Westfälische Wilhelms-Universität Münster (Germany), Katholieke Universiteit Nijmegen (Holland) and Università di Torino (Italy).
The contents of this page may be downloaded and printed out in single copies for individual use only. Making multiple copies without permission is prohibited.