Abstract
This essay investigates consequentialist reasoning in law. It begins with a brief exposition of the structure of consequentialist reasoning. It then turns to the role of consequentialist reasoning in two aspects of legal decision-making. Legal officials must reason both about ends and about the choice of means to achieve those ends. Legal instrumentalism, however, takes many forms, and different forms identify different officials to engage in the task of reasoning consequentially to choose means. The essay then considers the difficulties posed by the fact that legal decision-makers choose policies or institutions that structure the outcomes they hope to achieve. Finally, the essay considers the role that consequentialist reasoning may play in choosing ends.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Some authors, e.g., Broome (1991) and Woodard (2008) distinguish between teleological and consequentialist views. On their accounts, a consequentialist view adopts an impartial or objective theory of value while a teleological one may have a subjective theory of value. I ignore this distinction here and use the terms “teleological” and “consequentialist” interchangeably.
- 2.
Indeed, the outcomes depend on many other factors as well—the extent to which law enforcement agencies police the speed limit; the level of the fine for violation, etc.
- 3.
One might begin instead with a choice function that identifies, for each possible set of options available to the agent, the option that agent would choose. A large literature investigates the relation between choice functions and rankings. See, e.g., Hansson and Grune-Yanoff (2012) for a brief survey.
- 4.
The text understates the complexity of the comparative process. There are at least two other “answers” to the comparative process. It may be that the ranking is incomplete so that A and B are “not comparable.” So, for example, the Pareto criterion provides a partial ranking for which any two Pareto optimal points are Pareto non-comparable. Ruth Chang suggests that “A may be on a par with B,” a relation that undermines the construction of a ranking. Under this approach, the ranking is complete but it may not be transitive or even acyclic (Chang 2001).
- 5.
Often, the analyst assumes that an agent who must choose an action has rankings over actions, that an agent who must choose a policy has a ranking over policies, or that an agent who must choose an institution has a ranking over institutions. The analyst often assumes, that is, that the agent’s domain of preference corresponds to her domain of choice. But this elision of the two domains in some cases may mischaracterize the agent’s decision problem. The agent requires only a ranking of consequences; she may be unable to rank the options in her domain of choice. Two examples will suffice. First, consider a simple game such as chicken in normal form. Each player has a ranking over each of the four outcomes. But she does not have a well-defined ranking over her actions. She ranks swerve over straight if the other player has chosen straight, but she ranks straight over swerve if the other player has chosen swerve. Second, consider a voter who must choose among several candidates (A, B, and C) to complete a legislature that will then enact a legislative program. The agent must have a ranking over legislative programs, but she need not have a ranking over candidates. She may rank A above B and C if the rest of the legislature consists of X, but she may rank C over A and B if it consists of Y. For further discussion see the text below at footnotes 21–25 and Kornhauser (2003).
- 6.
I have assumed that Liza exhausts her budget; holding cash has no value for her. If holding cash after purchasing fruit has value, then Liza must have preferences over triplets (m, a, o) where m represents the cash remaining after all purchases.
- 7.
This framework corresponds roughly to the framework introduced in Savage (1954). Savage, however, studies decision-making under uncertainty. He thus assumes that the agent has primitive preferences over actions (or choices), and then identifies conditions on these preferences so that the preferences over actions can be represented by a preference over outcomes (itself representable by a utility function) and a set of beliefs about states the world such that one action A is preferred to another action A′ if and only if the expected utility of A exceeds the expected utility of A′.
- 8.
More precisely: “Choose an action A from among those actions for which no better action is available.” When the ranking over actions is complete, the agent chooses from among those actions that are maximal; if there is more than one, she is indifferent among them. If the ranking is incomplete, the agent again chooses among actions that are maximal but now two actions available to her may be non-comparable (on the criterion used by the ranking).
- 9.
Again, more precisely: Choose an action A from among those actions that satisfy standard S.
- 10.
There is a subtle distinction between optimization and maximization that the text ignores. Contrast the phrasing of C1—“Choose A if there is no better option available”—with the phrasing of C1’—“Choose A if A is better than every other option.” (Or “Choose A if A is at least as good as every other option.” When the agent’s ranking is incomplete, no option may satisfy C1’ though many satisfy C1.
- 11.
Application of this reformulation does require some care however because it reformulates the agent’s preference in a way that reduces the amount of information available. This reduction becomes significant when we weaken rationality by weakening the requirements imposed on the ranking. See below.
- 12.
Of course these additional criteria would apply to any pair, even when neither option was maximal. Adding a criterion that refines an existing ranking might thus be difficult as its application must not disrupt the ranking of non-maximal elements though it may also serve to reduce the size of equivalence classes.
- 13.
The weights represent her time preferences.
- 14.
Assuming that she cannot trade in any subsequent period.
- 15.
Some analysts think that Liza’s conundrum undermines the standard economic account of rationality. These critics contend that it is rational for Liza to follow through on her plans even if it is not, at some time t, sequentially rational for her to do so. For different accounts, see McClennan (1990) and Gauthier (1990). The arguments offered against standard rationality in the sequential context are often extended to strategic situations such as the prisoner’s dilemma. For further discussion of these issues from a somewhat different perspective see Zaluski, Chap. 6, part II, this volume, on “Interactive Decision Theory and Morality.”
- 16.
One might argue that courts do not first announce a rule and then apply it. But in a common law jurisdiction, we can imagine the announcing rule in a case in which both parents have the best interests of the child at heart and thus will achieve a reasonable allocation of custodial responsibility. In the original story, of course, Solomon announces the rule and then does not enforce it. See also the Chinese play The Chalk Circle and Brecht’s variation on it, The Caucasian Chalk Circle.”
- 17.
For some diseases, this is not true; it is more cost-effective to treat cholera than to vaccinate against it. This trade-off reflects the relative inefficacy of the vaccine and the low cost of treatment (in a well-functioning public healthcare system).
- 18.
The problem facing the government official may look different from the problem of sequential rationality because the legal institution that set the policy differs from the legal institution that enforces or implements it. In thinking about these issues, however, it is helpful to think of the public officials as a team that share an objective function. On this assumption, the problem of sequential rationality re-emerges.
- 19.
See Zaluski (Chap. 6, part II, this volume, on “Interactive Decision Theory and Morality”) for a more extensive development of the theory of games. He does not address, however, the issue discussed in the text.
- 20.
Notice that players with different preferences over the outcomes in matrix 1 will play a different game. It may be helpful to restate the structure of the argument in terms of the prisoner’s dilemma. That game form consists of the matrix that has, in each cell, the sentence that each prisoner receives given the pair of strategies that correspond to that cell. The standard prisoner’s dilemma matrix assumes that each player ranks the outcomes solely on the basis of the sentence he receives and that he prefers shorter sentences to longer ones.
- 21.
And she prefers T to B if Column chooses R but B to M if Column chooses L.
- 22.
Notice that maximin, as it does in the case of decision-making under uncertainty, does yield a ranking of each player’s choice set. Row’s ranking is B >M > T while Column has the ranking C > R > L.
- 23.
Of course, Column under C1 would also not be content with C conditional on Row playing B. Column would prefer to play R. And as before, both Row and Column are better off at (B, R) than (B, C).
- 24.
Note that maximin is not inconsistent with C1; it just reflect an extreme degree of risk aversion. On an alternative account, under Maximin, Row assumes that Column has malevolent preferences and will act to maximize them. But if Column has the preferences he does have and if Row engages in strategic reasoning C1 suggests Nash equilibrium.
- 25.
One might think about the failure of integration of the agent’s multiple criteria into a coherent all-things-considered ranking in another way. An agent with multiple criteria seeks to integrate them into a single ranking. As the prior discussion suggests, many different integrations are possible. A rationality of ends thus requires not only a specification of rational ends or criteria against which to evaluate options but also a specification of rational conditions of integration. Little attention has yet been given to the identification of appropriate conditions of integration in policymaking. Which conditions are appropriate will depend on the nature of the decision problem. As we have seen, economists have intensively investigated the appropriate conditions for the state-by-state and period-by-period integration of a single decision-maker’s criteria. The second area in which intensive investigation of criteria of integration has occurred is social choice theory. Social choice theory investigates the conditions under which individual rankings can be integrated into a social ranking. Arrow (1963) launched the discipline with a result that showed that four conditions of integration were logically incompatible.
- 26.
Of course the foundations of decision theory are themselves disputed. A number of different formal characterizations of consequentialist reasoning have been advanced. See for example Hammond (1988) and McClennan (1990). These authors study sequential decision-making to probe the underlying premises of rationality so understood.
- 27.
Public choice theory denies that legislatures act in a strongly instrumental way. Rather scholars such as Buchanan adopt an institutionally instrumental stance toward legislatures but expect legislators within the institution to act self-interestedly.
- 28.
- 29.
We shall see, however, that the problem of integration of ends into a single ranking parallels an aspect of the second problem of determining collective ends. For further discussion of the problem of integration of ends, see Kornhauser (1998).
- 30.
In this essay, I put to one side the drafting of constitutions. As constitutions are typically drafted by assemblies—either constitutional conventions or legislatures—many of the same issues arise in the determination of constitutional ends as in the determination of legislative ends. Additional complications, however, may arise from the fact that constitutions are typically ratified by the citizenry.
- 31.
I refer to the collective end embodied in any legal norm as its “statutory purpose.” Not only statutes and ordinances but also administrative regulations and judge-made rules thus have statutory purposes. As these norms are typically announced by collegial bodies, they present similar questions concerning the construction of the statutory purpose.
- 32.
Sartor, however, ignores two issues. First, he ignores the multiplicity of legislators. As the next paragraphs of the chapter indicate, legislators may disagree about the nature of the problem to be solved, the set of constraints the legislature faces, and the causal consequences of each alternative solution. A complete normative theory of legislator behavior would provide guidance to legislators on the resolution of these conflicts. For a parallel argument regarding normative theories of adjudication, see Kornhauser (2013). Second, he ignores the question of representation, i.e., the relation of the legislator’s views to the views of her constituents.
- 33.
I put to one side questions raised by the literature on social choice theory and judgment aggregation. These theories start from the assumption that the statutory purpose should reflect the preferences, beliefs, and judgments of the legislators. Aggregation of these attitudes must confront various logical problems. For a discussion see List (2008).
- 34.
Rather we expect the policeman to decide on the basis of a departmental practice that perhaps was elaborated on the basis of a purposive interpretation of the legal norm. The officials promulgating the regulations governing police searches and seizures are sophisticated interpreters.
- 35.
Some cases are more complex. Consider a motorized wheelchair that might both pose a danger to children and be noisy. On the other hand, the judge presumably faces a set of background constraints about equal concern and respect for individuals with disabilities that might argue for permission.
- 36.
Brown v. Board of Education 347 US 483 (1954).
- 37.
One could understand the Court’s reluctance to require cross-jurisdictional busing as reflecting a different understanding of the relative weights of federalism and equality rather than as a different constraint.
- 38.
Abbott v. Burke 100 NJ 269, 495 A.2d 376 (1985), Abbott v. Burke 119 NJ 359, 575 A.2d 359 (1990), Abbot v. Burke 136 NJ 444, 643 A.2d 575 (1994), Abbott v. Burke 149 NJ 145, 693 A.2d 417 (1997), Abbot ex rel. Abbott v. Burke 199 NJ 140, 971 A.2d 989 (2009).
- 39.
Some legislative acts concern projects rather than programs. In the mid-1950s, for example, the United States Congress began a project to construct an interstate highway system. In subsequent years, Congress appropriated money to fund various stages of this project. The teleological reasoning underlying the appropriation bills was trivial.
- 40.
A number of questions arise concerning how a policymaker should make these predictions. Some authors—most obviously, Brennan and Buchanan (1981, 1985) have argued that the policymaker should use rational choice theory to predict the consequences of different legal rules even if that theory is false. Kornhauser (2002) argues against their claim and suggests that, at least in some instances, prediction does not require an explanatory theory at all.
- 41.
For a discussion of the consequences of installing parking meters in the central business district of Pasadena, CA, see Shoup (2005). His entire seven hundred page book addresses the dramatic (negative) consequences of a range of policies that constitute “free parking.”
- 42.
I have merely indicated some of the economic consequences. A policy change may have political consequences as well. The neighborhoods on the border of the metered zone may be upset by the increased demand for free parking in their neighborhoods. Similarly, if the business drops in the metered zone, the policymaker can expect political anger from the businesses there; if business increases, she may be targeted by those businesses outside the metered zone that have lost customers. Of course, if the losing businesses are outside her jurisdiction she may not care.
- 43.
The Framers of the US Constitution did not anticipate the emergence of a party system.
- 44.
So, for example, the United States Congress uses a ten-year time horizon to assess the budgetary impact of legislation.
References
Arrow, K.J. 1963. Social choice and individual values. New Heaven: Yale university press.
Brennan, G., and J. Buchanan. 1981. The normative purpose of economic “science” rediscovery of an eighteenth century method. International Review of Law and Economics 1: 155–166.
Brennan, G., and J. Buchanan. 1985. The reason of rules: constitutional political economy. Cambridge: Cambridge University Press.
Broome, J., 1991. Weighing Goods. Oxford: Basil Blackwell.
Chang, R. 2001. Making comparisons count. Abingdon: Taylor and Francis.
Frankfurt, H. 1982. The importance of what we care about. Synthese 53: 257–272.
Gauthier, D. 1990. Moral dealing: contract, ethics, and reason. Ithaca, NY: Cornell University Press.
Hammond, P. 1988. Consequentialist foundations for expected utility. Theory and Decision 25: 25–78.
Hansson, S.O., and T. Grüne-Yanoff. 2012. Preferences. The Stanford Encyclopedia of Philosophy, ed. E. Zalta. http://plato.stanford.edu/archives/win2012/entries/preferences/.
Hart, H.L.A. 1958. Positivism and the separation of law and morals. Harvard Law Review 71: 593–629.
Kornhauser, L.A. 1998. No best answer? University of Pennsylvania Law Review 146: 1599–1637.
Kornhauser, L.A. 2002. Virtue and self-interest in the design of constitutional institutions. Theoretical Inquiries in Law 3: 21–47.
Kornhauser, L.A. 2003. The domain of preference. University of Pennsylvania Law Review 151: 717–746.
Kornhauser, L.A. 2013. Deciding together. NYU Law and Economics Research Paper 13–37. http://ssrn.com/abstract=2332236.
List, C. 2008. Judgment Aggregation: A Short Introduction. http://personal.lse.ac.uk/list/pdf-files/ja-intro.pdf.
Marmor, A. (ed.). 1995. Law and interpretation. Oxford: Clarendon Press.
Marmor, A. 2005. Interpretation and legal theory. Oxford: Hart Publishing.
McClennan, E. 1990. Rationality and dynamic choice. Cambridge: Cambridge University Press.
Mnookin, R., and L. Kornhauser. 1979. Bargaining in the shadow of the law: the case of divorce. The Yale Law Journal 5: 950–997.
Rawls, J. 1971. A theory of justice. Cambridge, Mass.: Harvard University Press.
Richardson, H. 2003. Practical reasoning about final ends. Cambridge: Cambridge University Press.
Sartor, G. 2010. Doing justice to rights and values: teleological reasoning and proportionality. Artificial Intelligence and Law 18: 175–215.
Savage, L.J. 1954. The foundations of statistics. New York, NY: Dover Publications.
Schmitz, D. 1996. Rational choice and moral agency. Princeton, NJ: Princeton University Press.
Shoup, D. 2005. The high cost of free parking. Chicago, Ill.: APA Planners Press.
Woodard, C. 2008.Reasons, patterns, and cooperation. NY: Routledge.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature B.V.
About this chapter
Cite this chapter
Kornhauser, L.A. (2018). Choosing Ends and Choosing Means: Teleological Reasoning in Law. In: Bongiovanni, G., Postema, G., Rotolo, A., Sartor, G., Valentini, C., Walton, D. (eds) Handbook of Legal Reasoning and Argumentation. Springer, Dordrecht. https://doi.org/10.1007/978-90-481-9452-0_14
Download citation
DOI: https://doi.org/10.1007/978-90-481-9452-0_14
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-9451-3
Online ISBN: 978-90-481-9452-0
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)