Law, Economics and Uniform Contract Law

Jaap Hage


1.  It goes without saying that the Dutch meteorological service is more competent than the author of the present paper to predict the weather in the Netherlands. So, when the question arises whether the temperature on August 8th, 2030 in Maastricht will be 30 degrees Celsius, it seems at first sight obvious that it is better to consult the Dutch meteorological service than the present author. On closer inspection, however, this becomes less obvious. Not because the author happens to know more about the weather than a layman, but because the meteorological service, despite its superior knowledge, probably cannot do better than a layman. There are many factors that are relevant for a precise weather prediction, while the values of many of these factors are unknown. As a consequence, the meteorological service cannot do better than forecast that, given the time of the year and the temperatures of the last few years, there is a fair chance that the temperature will be 30 degrees Celsius. However, another temperature is more probable, if only because there are more other temperatures. In other words, the considerable superiority of the meteorological institute in relevant knowledge is negligible in comparison to the large amount of possibly relevant information.

     It is one of the main tenets of this paper that Law and Economics is in a similar position with regard to the issue whether European contract law should be uniformed. The knowledge that Law and Economics has to offer is in principle relevant, but hardly scratches the surface of all relevant information. The use of Law and Economics for answering the question whether European contract law should be uniformed is therefore limited to an extent that makes it almost irrelevant. Before I can draw this conclusion, I will have to say more about the Law and Economics approach and its role in theorising about the law.


2.  The Law and Economics approach has at least two functions with regard to the law. One is to explain existing law, the other to evaluate actual and possible law. A legal arrangement, such as the presence of a legal institution (e.g. private property), or a particular regulation, can be explained by pointing out that it is efficient. Such a functional[1] explanation can only be plausible on the assumptions that efficient[2] arrangements have a better chance of being introduced, or – even more importantly – of being maintained in the course of time. Whether these assumptions hold good, depends on a number conditions, such as whether the efficiency is recognised as such, and whether there are other factors (ideological[3], for instance) which make that less efficient institutions or regulations survive in the battle for existence.[4]

     A second function of Law and Economics is to contribute to the evaluation of actual or possible legal arrangements. This is the function that is at stake in the discussion whether Law and Economics can help answering the question whether European private law should be harmonised.


3.  A rational evaluation of anything, including legal arrangements, takes two kinds of entities. One kind consists of factual characteristics of the item that is evaluated, including facts that do not belong to the item proper, but which are relevant for its evaluation. For instance, for the evaluation of the harmonisation of European contract law it is a relevant characteristic that the harmonisation makes that contract law will be uniform in the European Union (to the extent that it participates in the unification). For a particular instance of harmonisation additional facts may be relevant. For instance, for the hypothetical harmonisation promoted by the European Commission that should be finished in 2030, facts such as the political climate at the time that major steps towards harmonisation are made, are relevant too.

     The other relevant entities are the standards by means of which the evaluation is performed. An overly simple example of such a standard would be that harmonisation is good (should be brought about)[5], if the total economic production of the participating countries increases as consequence of the harmonisation. A value judgement is the result of applying the standards to the facts about the harmonisation.

     Law and Economics can contribute to the evaluation of the harmonisation of European contract law, both by bringing relevant facts to the light, and by contributing standards for the evaluation of the relevant facts. These two contributions are not independent. The standards by means of which the harmonisation is evaluated determine which facts are relevant. A standard that mentions the total economic production of the participating countries as a relevant factor, makes facts about economic production relevant. A standard that refers to the political climate makes the political climate relevant. Notice that this logical relation between the adopted standards and the relevancy of facts does not imply that the standards are established first and are subsequently used to determine which facts are relevant. It is also possible that some facts are recognised as relevant and that this perceived relevancy is generalised into a standard that attaches a value judgement to the presence of these facts.[6]


4.  How can Law and Economics provide standards by means of which the harmonisation of European contract law can be evaluated and by means of which relevant facts can be recognised as such? It is characteristic for the Law and Economics approach that it operates in terms of costs and benefits.[7] Law and Economics can point to costs and benefits of harmonisation, and can help in balancing them. In fact, these two possible contributions represent two distinct aspects of the Law and Economics enterprise. The identification of costs and benefits of an arrangement are what might be called the ‘material’ aspect of Law and Economics, because it deals with the material aspects of the choice. The theory about how costs and benefits are to be balanced represents the formal aspect of the Law and Economics approach, because it abstracts from the material aspects of a particular choice and deals only with a general theory of rationality.[8]

     The material aspect is not without complications. Are the costs and benefits to be taken broadly, as to include all disadvantages, respectively advantages of an arrangement? In that case the evaluative Law and Economics approach would be nothing else than the theory that those legal arrangements are good (to be introduced or maintained), which have more advantages than disadvantages. If there are several such arrangements, it would be the theory according to which the legal arrangement with the biggest advantage-disadvantage difference is the best, or should be adopted. On this view, it is unclear what the specific role of economics would be, unless economics is taken broadly as the theory of advantage-disadvantage analysis.

     It is also possible to take the costs and benefits more strictly. A natural interpretation would be to equate them with monetary costs and benefits.[9] The link with economics would then be obvious, but at the cost of disregarding everything that is not (or cannot?) be evaluated in terms of money. For instance, if the harmonisation of European contract law has monetary benefits because it diminishes transaction costs[10], while there are no important monetary costs involved, the verdict would be that from a Law and Economics perspective, harmonisation is a good thing. If harmonisation leads to loss of cultural identity, this aspect does not count, because it is not expressed (and – arguably – not adequately expressible) in monetary terms.

     Another way to deal with this complication is that the Law and Economics judgement on legal harmonisation is explicitly relativised to the economic perspective, thereby leaving the balancing of economic and non-economic aspects of harmonisation out of consideration. For instance, Law and Economics would pass the judgement that harmonisation of contract law is from the economic point of view desirable, because it diminishes transaction costs, while there are no important monetary costs involved. This last possibility would not be a cause of worry if the formal aspect of the Law and Economics approach would not be useful to balance economic and other aspects of harmonisation. However, balancing economic and non-economic aspects of harmonisation is just another case of balancing advantages and disadvantages. As a consequence, leaving non-monetary aspects out of the discussion means that one does not use an instrument of (Law and) Economics (the theory of balancing) that seems highly relevant.

     It seems possible to take an in between approach by replacing monetary costs and benefits by individual preferences.[11] Preferences are a common phenomenon in economic analysis, and they play a role in issues where purely monetary considerations are irrelevant. It is for instance possible to address the question whether harmonisation of contract law is in accordance with individual preferences if it involves both monetary benefits and costs with regard to cultural identity. Would one prefer harmonisation because of its monetary advantages, even though it involves a loss of cultural identity?

     The replacement of monetary considerations by preferences leads to new complications, however. Let us assume that monetary costs and benefits are an objective matter, and that value judgements based on monetary considerations are in this respect objective. Actual preferences are objective too, but they can be the outcome of false beliefs, or malevolent indoctrination. Suppose, for instance, that most inhabitants of the European Union prefer not to harmonise contract law, because they fear loss of cultural identity. Suppose, moreover, that this preference is the outcome of active campaigning by nationalist groups, supported by national defence industries which hope to gain from nationalist sentiments and the call for armament that results from it. Should such preferences be allowed to play a role in the evaluation of legal arrangements? If not, which demands should be made on preferences which are taken into consideration? Only preferences that are rational? If actual preferences are replaced by rational preferences, subjective values will be introduced in the form of standards for the rationality of preferences. Interpreting costs and benefits as disadvantages, respectively advantages, does not overcome this difficulty. On the contrary, identification of particular facts about a legal arrangement as advantages or disadvantages of this arrangement presupposes evaluative standards which are just as subjective as standards for rationality.

     To summarise this section: the material aspect of the Law and Economics approach has two ‘problems’. One problem has to do with the interpretation of ‘costs’ and ‘benefits’. If they are taken broadly, the typical economic aspect of the Law and Economics approach is lost. If they are restricted to typical economic characteristics of an arrangement, the restriction seems to be arbitrary, or the resulting judgement must be restricted - without good reasons - to a judgement relativised to the economic perspective.

     The second problem is that the identification of costs and benefits threatens to be subjective and possibly error-prone, namely if they are defined in terms of actual preferences, or in terms of advantages and disadvantages that are perceived as such. This subjectivity can be avoided by defining costs and benefits in ‘objective’terms, such as monetary costs and benefits, but then the question is how such costs and benefits can lead to value judgements about legal arrangements. If it is shown that the purely monetary costs of harmonisation of contract law exceed the purely monetary benefits, it does not follow automatically that harmonisation should be rejected. This conclusion only follows on the additional premises that these purely monetary costs are the only relevant factors that should be taken into account and that the net balance of these costs and benefits determines whether harmonisation should be accepted or rejected.


5.  The formal aspect of the Law and Economics approach concerns the lines of reasoning by means of which costs and benefits (or their replacement categories) are weighed. If the Law and Economics approach is confined to monetary aspects of legal arrangements, the lines of reasoning are quite simple. There are two kinds of questions that play a role. One is whether a particular legal arrangement is good. This can be established by determining whether the benefits, expressed in some monetary unit, exceed the costs as expressed in the same monetary unit. To put it bluntly and apply the issue to the harmonisation question: Are the profits of the harmonisation of European contract law, as expressed in Euros, bigger than the costs as expressed in Euros? If the costs and benefits are uncertain, this should be taken into account be multiplying the costs or benefits with the probability (on a scale from zero to one) of their occurrence.[12]

     The other kind of question is when there are several alternatives under consideration, and the issue at stake is which of the alternatives should be adopted. Then the line of reasoning that should be followed is to determine for each alternative the net benefits (which may be negative), and then pick the alternative with the highest benefits.[13]

     It is important to notice that these simple lines of reasoning are only available if all the costs and all the benefits of an arrangement can be expressed on a common scale which allows mathematical operations (subtract the costs from the benefits), and negative numbers (if the costs exceed the benefits). If everything that is relevant can be expressed in monetary terms, this condition is satisfied, but the demand that everything relevant is expressed in monetary terms is quite demanding, to state it modestly.

     If not all costs and benefits can be represented on a common scale that allows calculations, the lines of reasoning that are required to make justified judgements about legal arrangements become more complicated. There are several issues that must be dealt with. One is the absence of a numeric scale, or at least a scale that allows calculations. Another one is that different costs and benefits cannot be represented on a common scale, which had led some to the theory that they are incommensurable.[14] There exists, however, theories that deal with reasoning on the basis of factors that are not numeric and that cannot be represented on a common scale.[15] These theories are less powerful than calculation of costs and benefits represented in terms of monetary units, but have the advantage that their scope of application is much broader.[16]

     In combination, the theories about quantitative and qualitative comparative reasoning provide a general framework that can be applied to choices between alternatives. If the costs and benefits of the alternatives can be represented on a common scale that allows calculation, the more powerful quantitative techniques should be applied. If this is not possible, one must be satisfied with qualitative comparative reasoning.


6.  The costs and benefits of some action such as the harmonisation of European contract law consist of the costs and benefits of bringing this action about and the costs and benefits that are consequences of this action. These costs and benefits are always costs and benefits of a concrete action, which actually took place.[17] So if we want to determine the costs and benefits of harmonising European contract law, the relevant costs and benefits will be attached to the harmonisation as it actually took place. These costs and benefits are not available when one merely considers whether to harmonise the law. In advance one can only speculate about the actual costs, and one is forced to confine oneself to the costs and benefits of actions in general that involve the harmonisation of European contract law. All details of the practical implementation must be left out of consideration.

     Let me be more concrete. If one tries to determine the costs and benefits of harmonisation, the only thing that one knows about the action in question is that involves harmonisation. The only consequences that can be taken into account are consequences that are attached to harmonisation in general. Probably these consequences are limited to that European contract law will be uniform, and nothing else. In particular it is not given that the transaction costs involved in studying the law of foreign countries will disappear, for instance because it is not given that the new uniform law will be published. Clearly this is a counterfactual hypothesis. If European contract law will be harmonised, this will probably be accomplished through the introduction of uniform legislation that will be published. However, if this natural assumption is made, the arrangement that is evaluated is not anymore the harmonisation of European contract law in general, but the harmonisation through published legislation. Let us therefore assume that the issue at stake is reformulated as whether it is desirable to harmonise European contract law by means of uniform legislation that will be published. Then it seems safe to assume that at least one of the benefits will be the avoidance of transaction costs.

     Yet, this assumption is not safe at all. Maybe the uniform legislation will contain many open norms, which will be filled in differently in different countries. This possibility can be taken away by making still additional assumptions, for instance that there will be few open norms and that there will be a common supreme court that takes care of legal unity in the interpretation of the remaining open norms. On this assumption the legal arrangement to be evaluated is not merely the harmonisation of European private law, but the harmonisation by means of uniform legislation that contains few open norms and that is kept uniform by means of a common supreme court … etc.

     As these examples illustrate, the determination in advance of the costs and benefits of a legal arrangement requires that many details of the arrangement are to be specified in advance, because only then the consequences of the arrangement can be taken into account.

It will be clear, however, that it may be possible to make proposed legal arrangements somewhat concrete, but that it is impossible to make them so concrete that all relevant costs and benefits can be determined in advance. The best possible approach is to specify in advance all details that are relevant for the determination of foreseeable costs and benefits and to evaluate the resulting proposal on the assumption that the remaining costs and benefits will be relatively unimportant. In connection with the harmonisation of European contract law this means that one should specify in advance all factors one can think of that might be relevant for the costs and benefits of harmonisation. Then one should formulate a possible line of action, in which all these relevant factors are specified. For instance, one should consider the harmonisation of European contract law by means of central European legislation that contains many open norms referring to local standards, with the establishment of a European court that has the task to look after the progress of the harmonisation, where the members of the European Union are free to determine whether they will participate in the harmonised law. In this way, it is possible to achieve at least some certainty about the consequences of the proposed arrangement, which makes the presupposition more realistic that the costs and benefits that were not taken into account would not tip the balance of costs and benefits.

     The same point can be made in a different way. To evaluate a legal arrangement, one should ideally know all its consequences. Since these consequences are attached to the actual implementation (or maintenance) of the arrangement (the implementation token), while the only thing we know is what kind of arrangement will be implemented (the implementation type), there is a problem. The best way to overcome this problem is to specify the type of the arrangement as much as possible, in order to make it possible to determine as many as possible of its consequences in advance. On the basis of these foreseeable consequences, the implementation of the arrangement should be evaluated in terms of costs and benefits. On the basis of this evaluation, a decision about implementation can be made, under the assumption that the costs and benefits that were not foreseeable and were therefore not taken into account, would not tip the balance of costs and benefits.


7.  This leads me to the issue with which I started this paper. Is the assumption realistic that we can sufficiently specify a way to harmonise European contract law to predict a sufficient large part of its costs and benefits to make a useful evaluation possible? Or is it in practice impossible to foresee sufficiently many of the relevant consequences, and can we only make estimates of the costs and benefits that we know on beforehand to be insufficiently exhaustive. Is the harmonisation of European contract law comparable with predicting the precise temperature in Maastricht on August 8th, 2030? In the end this question is unanswerable, because to answer it we should both know which consequences are foreseeable and which consequences will actually occur. Although we can know which consequences are foreseeable (that is precisely what makes them foreseeable), we cannot know which consequences will actually occur, because actual consequences belong to actual harmonisation and actual harmonisation has not taken place yet when it is discussed whether it should be realised.

     The best approach in this connection seems therefore to look at a hypothetical example. Suppose that the diminution of transaction costs is the main foreseeable benefit of the harmonisation of contract law, and that this gain turns out not to be very important. In fact, it turns out to be less important than the cost of uniform law that the possibility to compare possible legal solutions for issues in contract law disappears.[18] Should we then draw the conclusion that harmonisation is not attractive and should not be brought about? No, because there is a (very) small possibility that the harmonisation of contract law contributes to the public feeling of belonging together and thereby prevents a war between European Union countries. This possibility is far-fetched and therefore I assigned it a (very) small possibility. However, the consequences of such a war would be tremendous to an extent that they can hardly or not at all be estimated. Since the relevance of such a war for the desirability of harmonisation consists of the product of its costs and its probability, the smallness of its probability is counterbalanced by the size of its costs. So maybe this factor should be taken into account.

But is that possible? Can we, if only by approximation, estimate the probability that harmonisation of contract law prevents a war that would otherwise have occurred, or the costs of such a war if it would occur? In my opinion we cannot, especially not where the probability of (the avoidance of) war is concerned. And this leads me to the conclusion that the harmonisation of European contract law is comparable to the long term prediction in detail of the temperature in a particular city: there is relevant information available but there is so much relevant information not available that the value of what is available is very limited.


8.  Should we abandon all hope then, and refrain from deliberating about future actions because we cannot (be certain to) lay hand on all the relevant information? That would be a bridge too far. When deciding what to do, the best (most rational) approach it to work on basis of all the relevant information that is available, while remaining aware of the fact that all the available information may not contain everything that is relevant. To state it somewhat paradoxically:

Rational action takes into account that it is incompletely rational.

This was already known to Karl Popper, who drew from this insight the advice of piecemeal engineering.[19] Changes to social institutions (including the law) should take the form of small adjustments and re-adjustments which can be continually improved upon. This last phrase is in my opinion crucial, and indicates that institutional changes should be made in such a way that the possibility of continuous monitoring and re-adjustment is so to speak built into the changes. In the case of the legal harmonisation this might take the form of small harmonisations (e.g. harmonisations of limited parts of contract law), or locally restricted harmonisation (start with a few countries and extend if possible).


9.  Let me summarise the above. L&E provides relevant information for the decision whether European contract law should be harmonised. This relevant information can be divided into material and formal information. Materially, L&E can help deciding which facts about harmonisation are relevant to determine whether harmonisation is desirable. There are some complications involved, concerning the subjectivity of the standards for what is relevant, but in my opinion these complications are involved in all evaluative enterprises. Formally, L&E provides tools to determine, given the relevant facts about harmonisation, whether harmonisation is desirable. These tools can be quantitative, but often it will be necessary to fall back on tolls for qualitative comparative reasoning. The usefulness of these formal tools is independent of the complications on the material side.

     Although L&E provides relevant information, both on the material and on the formal side, the question remains of this information makes reliable judgments on the desirability of harmonisation possible. In this connection the parallel with weather prediction plays a role. If one has some relevant information, but lacks a lot of other relevant information, a well-founded judgement is not possible. A meteorological institute can for this reason not predict the precise temperature 25 years from now. By means of an example about the prevention of war, I have tried to show that the harmonisation issue is in this respect similar: there are so many possibly relevant things that we do not know, that it is not well possible to make reliable estimations about the desirability of harmonisation.

     However, this does not mean that we should abandon hope and act at random because the issue cannot be solved rationally. Rationality in complex social issues requires that one takes into account that much relevant information is lacking and that we must take decisions under uncertainty. Popper’s advice of piecemeal engineering is based on precisely this insight. With regard to the harmonisation of European private law this suggests that harmonisation, if it is undertaken, should proceed in small steps, and that the continuation and the precise form of the process should be adapted to the findings underway.


Appendix: Qualitative comparative reasoning[20]


Suppose that one must choose between buying a Volvo and buying a Mercedes, and that there is no quantitative scale available by means of which these two alternative scan be compared. A rational theory about decision making should then operate on a qualitative basis. Such a qualitative theory about the comparison can be developed on bais of the notions of reasons pleading for and against alternatives.

A Volvo has two reasons pleading for it, namely that it is a safe car, and that there is a Volvo dealer next door. It has the disadvantage that it is an expansive car. A Mercedes is also expensive, but has (in the example) only one advantage, namely that it is a safe car. There happens to be no Mercedes dealer in the neighbourhood. Under these circumstances, everything that pleads for a Mercedes also pleads for a Volvo, but a Volvo has an additional reason pleading for it, namely the availability of a dealer nearby. Moreover, a Volvo and a Mercedes have the same reason pleading against it, namely that they are expensive. It seems, therefore, that a Volvo is preferable to a Mercedes. This is a reasonable conclusion, even in the absence of any information concerning the (relative) weights of the reasons that the cars are safe, that there is a Volvo dealer nearby, and that the cars are expensive. Analogously it is reasonable to conclude that a Mercedes is preferable to a Porsche, if a Mercedes and a Porsche have the pro-reason in common that they are German cars (for those who like German cars), and they also share the con-reason that they are expensive, while a Porsche has the additional disadvantage that it liable to be stolen. In general alternative A is preferable to alternative B if either:

1.    the set of reasons pleading for A is ‘stronger’ than the set of reasons pleading for B, while the set of reasons pleading against A is not ‘stronger’ than the set of reasons pleading against B; or

2.    the set of reasons pleading against B is ‘stronger’ than the set of reasons pleading against A, while the set of reasons pleading for B is not ‘stronger’ than the set of reasons pleading for A; or

3.    both 1 and 2 hold.


A similar approach can be used to determine whether one set of reasons is ‘stronger’ than another set, because that is also a matter of comparative reasoning that may have to be dealt with in a qualitative fashion.




Chang, R. (ed.) (1997), Incommensurability, Incomparability, and Practical Reason, Cambridge: Harvard University Press.

Cooter, R. and T. Ulen (2004), Law and Economics (4th ed.), Boston: Pearson.

Dawkins, R. (1989), The Selfish Gene (2nd ed.), Oxford: Oxford University Press.

Dennett, D.C. (1995), Darwin’s Dangerous Idea, London: Penguin.

Hage, J.C. (1997). Reasoning with rules, Kluwer Academic Publishers, Dordrecht.

Hage, Jaap C. (2001a). Legal logic. It’s existence, nature and use. In A. Soeteman (ed.), Pluralism and law. Dordrecht: Kluwer Academic Publishers, p. 347-374..

Hage, J.C. (2001b). Formalizing legal coherence. Proceedings of the 8th International Conference on Artificial Intelligence and Law, ACM, New York, p. 22-3Hare, R.M. (1963). Freedom and Reason, Oxford University Press, Oxford.

Keeney, R.L. and H. Raiffa (1993), Decision with Multiple Objectives. Preferences and Value Tradeoffs, 2nd edition, 1st edition 1976, Cambridge University Press.

Kerkmeester, H. (1999). Methodology: General, in B. Bouckaert and G. de Geest (eds.), ***

Popper, K.R. (1957), The Povery of Historicism, London: Routledge and Kegan Paul

Lando, O. (1992), Principles of European Contract Law, *** 261-273.

Rosenberg, A. (2000), Philosophy of Social Science, in W.H. Newton-Smith (ed.), A Companion to the Philosophy of Science, 451-460.

Smits, Jan M. (2002), The Harmonisation of Private law in Europe: Some Insights from Revolutionary Theory, Georgia Law Review 31, no. 1, 79-99.

Smits, J.M. (2003), De missing link in het debat over unificatie van privaatrecht: het evolutionair perspectief, NTBR 2003/5, 241-246.

Verheij, Bart (1996). Rules, Reasons, Arguments. Formal studies of argumentation and defeat. Thesis Maastricht University.

Wagner, G., The Economics of Harmonization: The Case of Contract Law, Common Market Law Review 39 (2002), 995-1023.

Wright, G.H. von (1963). Norm and Action, Routledge and Kegan Paul, London.



[1]     An explanation is functional if a phenomenon is explained by pointing out that it performs a useful function. Functional explanations play in particular a role in psychological, sociological and evolutionary settings. Cf. Rosenberg 2000, 455f.

[2]     Presently, I leave the issue what efficiency is deliberately open. A brief discussion of different economic notions of efficiency can be found in Cooter and Ulen 2004, 16f.

[3]     I will not discuss the issue whether ideological considerations play a role in determining efficiency.

[4]     Discussions of the (continued) existence of social phenomena in evolutionary terms can be found in, amongst others, Dawkins 1989, 189f. and Dennett 1995, 335f. A different evolutionary perspective is sketched in Smits 2002 and 2003.

[5]     In this paper I ignore the difference between evaluation and prescription and simply assume that the best legal arrangement is the one that should be brought about. This close relation between evaluation and prescription is quite often assumed, but should be scrutinised is a more extensive study.

[6]     This kind of generalisation is discussed Hare 1963, 7f.

[7]     This is another way of stating that the method of economics is a rational choice approach. Cf. Kerkmeester 1999, 384. The connection between rational choice and cost/benefit analysis is the assumption that the rationality of a choice depends (solely) of the costs and benefits connected to this choice (in relation to its alternatives).

[8]     The relativity of this distinction between formal and material aspects of a discipline, and more in particular between form and content in logic, is discussed in Hage 2001a.

[9]     According to Kerkmeester 1999, 387, this approach is chosen by Coase and Posner.

[10]    Cf. for instance Lando 1992.

[11]    An efficient arrangement would then be an arrangement that has the highest utility, meaning that it leads to the highest preference satisfaction. Cf. Kerkmeester 1999, 386.

[12]    Cf. Keeney and Raiffa 1993, 5f. This book also discusses more complex techniques of decision making.

[13]    The first kind of question can be treated as a question of the second kind, if one treats adopting and rejecting an arrangement as the only two alternatives under consideration.

[14]    See Chang 1997 for a general discussion of the problems.

[15]    Verheij 1996, Hage 1997 and 2001b.

[16]    The appendix gives an impression of what such a theory would look like.

[17]    In the terminology of Von Wright (1963, 36f.) consequences are tied to individual acts (act tokens) and not to generic acts (action types).

[18]    Cf. Smits 2002 and 2003 and Wagner 2002, 1001f.

[19]    Popper 1957, 66f.

[20]    This appendix is drawn from the paper Qualitative comparative reasoning, which is intended as part of a book Studies in Legal Logic that is still under review.