Goal-Setting and the Logic of Transport Policy Decisions
Holger Rosencrantz
STOCKHOLM, 2009
i
This doctoral thesis consists of the following papers preceded by an introduction:
I.
Rosencrantz, H. “Combining policy goals and welfare economics: the case of
Swedish transport,” Submitted manuscript.
II.
Rosencrantz, H., K. Edvardsson, and S.O. Hansson (2007) “Vision Zero – is it
irrational?,” Transportation Research Part A: Policy and Practice, 41(6):559-567.
III.
Grüne-Yanoff, T., and H. Rosencrantz “Beneficial risk increases,” Submitted
manuscript.
IV.
Rosencrantz, H. (2008) “Properties of goal systems: consistency, conflict, and
coherence,” Studia Logica, 89:37-58.
V.
Rosencrantz, H. “Deontic paradoxes: sources and solution,” Submitted manuscript.
Paper II is reprinted with permission from Elsevier. Paper IV is reprinted with permission
from Springer. Rosencrantz was the main contributor to the transport-specific sections of
Paper II, while Edvardsson and Hansson contributed equally to the theoretic sections.
Rosencrantz was the main contributor to the appendices as well as to the graphics of Paper III,
while Grüne-Yanoff was the main contributor to the non-formal discussion and literature
review. Permission to reprint the papers in this volume is granted by all authors of each paper.
The introduction is partly based on the following paper:
Rosencrantz, H. (2006) “Rational policy goals: road safety in Scandinavia,” pp.
151-157 in D.A. Brebbia and V. Dolezel (eds.) Urban Transport XII. Urban
Transport and the Environment in the 21st Century, WIT Press.
Holger Rosencrantz, Division of Philosophy, Department of Philosophy and the History of
Technology, Royal Institute of Technology (KTH), 100 44 Stockholm, Sweden.
© Holger Rosencrantz 2009
ISSN: 1650-8831
ISBN: 978-91-9415-252-4
Printed by US AB, Stockholm, Sweden, 2009
ii
to Hampus and Lucas, my nephews
iii
Abstract
Rosencrantz, H., 2009. Goal-Setting and the Logic of Transport Policy Decisions.
Theses in Philosophy from the Royal Institute of Technology 30. 108 + vi pp. Stockholm.
ISBN 978-91-9415-252-4
The thesis aims at developing approaches to transport policy decisions, based on suggestions
and ideas originating from moral philosophy and philosophical decision theory.
Paper I analyzes the Swedish transport policy goals, and the problem of combining policy
goals with welfare economics. A problem of circularity arises as the Swedish transport policy
goals are conflicting, and hence must be subject to trade-offs, while several of the goals
themselves entail statements on how to prioritize or restrain goals in case of conflict.
Paper II analyzes rationality in road safety policy. Problematic features are identified and
discussed. The paper argues that the Swedish road safety goal is rational, since it is actionguiding and achievement-inducing.
Paper III includes a model of rational choice under risk with biased risk perception. Under
certain plausible conditions, a regulator should raise the population’s risk exposure. By
deteriorating the environment the regulator can motivate drivers to choose behaviour that is
less biased.
Paper IV provides a formal representation of goal systems. The focus is on three properties:
consistency, conflict, and coherence. It is argued that consistency is adequately regarded as a
property relative to the decision situation or, more specifically, the set of alternatives that the
agent faces. Conflict is adequately regarded as a relation over subsets of a given goal system
and should likewise be regarded as relative to the set of alternative that the agent faces.
Coherence is given a probabilistic interpretation, based on a support relation over subsets of
goal systems.
Paper V investigates problems associated with standard deontic logic. A deontic predicate is
derived, which avoids some of the major paradoxes in the area. In particular, paradoxes
occurring when one obligation is derived by logical necessity from another obligation are
dealt with.
Keywords: goals; transport policy; road safety; coherence; action logic; deontic logic
Division of Philosophy, Department of Philosophy and the History of Technology, KTH,
Sweden.
iv
Acknowledgements
First of all, I wish to thank my supervisors Sven Ove Hansson and Till Grüne-Yanoff. Sven
Ove made this project possible in the first place, and rigorously supervised my work while at
the same time giving me much independence. Second, I wish to thank everyone else at the
division of philosophy at the Royal Institute of Technology for making it such an enjoyable
and stimulating environment. My roommate Eva Hedfors deserves special notice for the many
discussions we have shared, as does Karin Edvardsson for co-authoring one of the papers in
this dissertation. Though not formally a participant in this project, Eva has been an important
mentor for me on my way towards completing my PhD.
Among philosophers from other departments, I wish to thank Jan Odelstad for scrutinizing
my licentiate thesis in 2006. Paper IV is a direct outcome of a graduate course in
epistemology at Lund University, led by Erik J. Olsson and Bengt Hansson the same year.
Wlodek Rabinowicz supervised my master’s thesis in practical philosophy in 2003 and
encouraged me to proceed to graduate studies.
From outside the realm of philosophy, I have benefited from comments by Patrik Nylander
on Paper I. At the initial stage of the project, I benefited from discussion with an expert panel
including Lars-Göran Mattsson, Staffan Widlert, and Per Norman. Special thanks to Jakob
Nordin at the department of physics, Stockholm University, for helping me draw those nice
graphs in Paper III.
Naturally, I am also indebted to my family and personal friends for encouraging and
supporting me. My sincere appreciation goes to my parents Louise and Göran, to my sisters
Marie and Sophie, and to my brother Niklas, as well as to Magnus Kusoffsky, Maria Sewerin,
Great Cthulhu, Manne Sjöstrand, the members of Osqstämman and Kongl. Teknologkören,
and other friends who have kept me inspired and entertained during these years. While I do
admit that I have had fun and learned a lot, and that I have regarded my job as a privilege, a
PhD in philosophy is a long-term commitment involving a substantial share of unappreciated
and frustrating work as well as loneliness; for me, the outcome that made the project
worthwhile is that I met my girlfriend Maria during the process.
Financial support for this project was provided by VINNOVA. My final words of
acknowledgment are not “Thank you”, but “It was well spent money”.
Stockholm, February 2009
Holger Rosencrantz
v
Contents
Introduction ................................................................................................................................ 1 1. Aim and scope of the thesis................................................................................................ 1 2. Rationality of goals ............................................................................................................ 2 3. How should goal conflicts be rationally managed? ........................................................... 4 4. The logic of goals............................................................................................................... 5 5. Summary of papers I–V ..................................................................................................... 6 6. Afterthought ....................................................................................................................... 7 References ............................................................................................................................ 10 Paper I ……………………………………………………………………………………… 17
Rosencrantz, H. “Combining policy goals and welfare economics: the case of Swedish
transport,” Submitted manuscript.
Paper II …………………………………………………………………………………….. 35
Rosencrantz, H., K. Edvardsson, and S.O. Hansson (2007) “Vision Zero – is it irrational?,”
Transportation Research Part A: Policy and Practice, 41(6):559-567.
Paper III …………………………………………………………………………………… 45
Grüne-Yanoff, T., and H. Rosencrantz “Beneficial risk increases,” Submitted manuscript.
Paper IV …………………………………………………………………………………… 77
Rosencrantz, H. (2008) “Properties of goal systems: consistency, conflict, and coherence,”
Studia Logica, 89:37-58.
Paper V …………………………………………………………………..……………… 99
Rosencrantz, H. “Deontic paradoxes: sources and solution,” Submitted manuscript.
vi
Introduction
Goals are fundamental inputs to many decision problems. Transport policy is an area that
illustrates this. Management of transport infrastructure involves difficult decisions. Major
investments in, for example, new roads or railways compete for public resources; benefits are
to be appropriately distributed; and negative effects, in terms of safety and the environment,
are to be minimised. In a system as complex and important as a national transport system,
with actors involving industrial organisations and individual drivers as well as politicians and
administrators, the concept of rational policy decisions is well worth investigating.
In Sweden transport policy is formulated in terms of goals, which are explicitly formulated in
official documents (Gov. Bill 1996/97:137; Gov. Bill 1997/98:56; Gov. Bill 2001/02:20).
Sweden is not unique in this respect, since national and international transport policy goals
also exist elsewhere. Road safety goals have received much attention internationally (Peden et
al., 2004). In Finland, transport policy goals concerning financing, regional success,
environmental impact, and traffic safety have been formulated (Kommunikationsministeriet,
2008). Similary transport policy goal structures can be found in Norway and the Netherlands
(SIKA, 2008a).1 Policy goals of this kind guide strategies and constitute criteria for
performance evaluation. However, although a rational model of planning presumably entails
an approach aiming at social welfare and justice, national policy goals are set on a political
level and goal conflicts are often resolved by political – rather than economical – tradeoffs.
Given this context, questions of a philosophical nature appear relevant for both specific and
general topics. Is it irrational to set seemingly utopian and unachievable goals, such as the
vision to reduce road traffic fatalities to zero? If not, is there is a price on human lives (Elvik,
1999a; Elvik 2003)? What importance should be given to cost-benefit analyses; should they
be viewed as a moral standard or as a decision procedure (Adler & Posner, 1999; Hubin,
1994)? How should positive and negative effects of congestion pricing be distributed; what
would be fair?
1. Aim and scope of the thesis
This thesis aims at investigating rationality of transport policy decisions, including goalsetting and performance evaluation. More generally, the thesis aims at developing new,
alternative approaches and methods based on suggestions and ideas originating from moral
philosophy and philosophical decision theory. There are good reasons why transport research
and transport policy could benefit from an investigation with such an aim. Setting of goals is a
political issue and methods of evaluations are economical issues. But justifications of both
are, at least partly, philosophical issues. Any reference to welfare or justice, for example,
relates to classical questions of ethics and political philosophy – namely what constitutes a
good life and a good society (Griffin, 1986; Rawls, 1971). The concept of rationality, in turn,
is a main concern of decision theory as well as of moral philosophy – namely what constitutes
norms of choices and reasons for actions (Levi, 1997; Parfit, 1984).2
1
There are even more internationally adopted goals and targets that are not associated, or only partly associated,
with transport policy. Noteworthy examples include objectives of environmental protection and of poverty
elimination (UNDP, 2000; Smith & Todaro, 2006).
2
The term “decision theory” is sometimes narrowly referred to as the theory of rational decisions under risk,
uncertainty, and ignorance (e.g., Mattsson, 2005). In this thesis, however, the scope of decision theory is taken to
1
It is an aim of this thesis to make a constructive contribution to problems of practical concern,
that are, at the same time, philosophically interesting. This is what is commonly referred to as
“applied philosophy”. Applied philosophy requires identification and understanding of
problems of practical concern, and should ideally result in contributions that complement or
even replace contributions made by other disciplines (in this case, primarily economics and
other social sciences). By its very nature, therefore, the aim of making constructive
philosophical contributions to practical problems is ambitious, and demands that the
researcher approaches the task with realistic expectations and with respect for contributions
from other disciplines.
The philosophy of traffic and transportation is not a very big area and there is still much
pioneering work to be done. However, a few topics should be mentioned. One of the first
primarily philosophical works, with a focus on ethical theory, was published in Denmark in
the late 1990s (Zeitler, 1997). An investigation on the redistribution of responsibilities for
automobile safety during 1960-2000 has been published in the United States (Wetmore,
2003). Another notable topic is ethical analysis on moral responsibility in traffic safety
(Nihlén Fahlquist, 2005; Nihlén Fahlquist, 2006a; Nihlén Fahlquist, 2006b; Nihlén Fahlquist,
2007). There is also ongoing work about the ethics of traffic research and about the framing of
decisions in the transport sector. All these different research projects look at, or are at any rate
related to, goals within the transport sector.
There are different kinds of goals. A goal, or a set of several goals, may belong to an
individual or to a collective of individuals. Goals may be implicit or explicit, and they may be
more or less precise. Policy goals are generally explicitly formulated and apply for some
collective of individuals, but other kinds of goals will also be relevant to much of the
discussion in this thesis.
This thesis discusses the concept of rationality in a policy context with particular focus on
transport policy goal-setting and decision-making. This includes topics of philosophical
decision theory, as well as political philosophy. Furthermore, as exhibited primarily in the last
two papers, the scope covers formalization of goal systems and relations between goals. This
includes topics in the fields of belief dynamics and deontic logic.
2. Rationality of goals
Goals have certain functions in decision-making – at any rate, there are ways in which goals
should function. A goal should be meaningful, or non-trivial, if it is to affect any agent’s
decisions and it should not be vague or ambiguous. Furthermore, a goal should make it
possible to plan or coordinate actions over time (and, in the case of goals for organisations or
societies comprising more than one individual, between individuals). To put it more generally,
a goal should say or imply something about how the agent who has the goal should act. The
general requirement that goals should be action-guiding has been proposed in the discussion
on rationality of goals (Edvardsson & Hansson, 2005). More specifically, the requirement of
being action-guiding means that goals must satisfy three rationality criteria: precision,
evaluability, and approachability. In addition, rational goals must also be motivating, since no
agent would want to achieve or even be interested in pursuing goals that are not motivating.
include a wider range of situations, including decisions under certainty, group decisions, and games (Levi, 1986;
Luce & Raiffa, 1957; Resnik, 1987).
2
These four criteria resemble the five criteria behind the mnemonic SMART (i.e., objectives
should be Specific, Measurable, Achievable, Realistic, and Time-bound), often used in project
management in public services (e.g., van Herten & Gunning-Schepers, 2000). Other criteria
may appropriately describe the requirement that goals should guide actions, but it seems clear
that action-guiding goals must provide the agent with information and be adapted to the
agent’s abilities; goals must be meaningful in the sense that they inform the agent on how to
act and how to evaluate acts, and in the sense that the goal is relevant to choices between
alternative actions that are feasible for the agent.
The philosophical notion of rationality covers both theoretical (knowledge-related) and
practical (action-related) aspects (Mele & Rawling, 2004; Spohn, 2002). Practical rationality
is often defined in terms of goals, either political goals or individual goals. Although there is
much literature on rational decisions and actions relative to given goals, surprisingly little has
been written about the rationality of goals themselves. Actions are sometimes assessed as
rational relative to goals; an action is rational if it is expected to lead to the achievement of the
agent’s goals. Instrumental rationality, the capacity to choose the most efficient means to
achieve given goals, represents a reasonable but rather narrow concept of rationality (cf.
Simon, 1983). An “efficient” means, in the instrumental sense, need not be optimal – as made
famous by the concept of “satisficing” rationality (Simon, 1997; Rubinstein, 1998). Still, an
underlying assumption is that ends are rational in themselves – or, at any rate, that it is not
meaningful to refer to goals as “rational” or “irrational”.
Some writers have explicitly stressed this assumption. It has been said:
It cannot be too strongly emphasized that there are no criteria for the rationality
of ends as such other than the condition of consistency. Ends are completely
arbitrary. To prefer highly dispersed random outcomes may seem irrational to
the prudent, but for somebody with this penchant, there is nothing irrational
about it. This area is like that of tastes: they are what they are, and differ from
one person to the next. (Allais, 1953[1979], p. 70)
In the cases of what might be called “genuine” goals, goals that truly are ends in themselves
and not means to other ends, this position seems plausible at first sight; if a goal is set without
reference to any other goal, and if the goal is consistent with the other goals that the agent has,
how can we say that it is irrational? Amartya Sen has argued that a person whose goal it is to
cut off her own toes with a blunt knife would be better off by revising the goal to cut off the
toes, rather than by optimizing the goal by using a sharper knife, and that therefore this goal
may be said to be irrational (Sen, 2002). But this goal can be said to be irrational only if it is
understood that the goal is a means to another goal, namely to become somehow “better off”,
and not if it is understood that the goal is strictly an end in itself. It is true that the person in
Sen’s example seems irrational. But if she maintains that her goal is genuine and an end in
itself, then there seems to be no ground for criticism in terms of rationality.
On the instrumental account, as expressed by Allais, goal systems are rational if and only if
they are consistent. And it certainly seems plausible to say that goals are irrational when they
are contradictory, or in some other way stand against each other. But how useful is this
criterion? Two distinct cases of inconsistency can be identified. Logical inconsistency of
goals certainly seems irrational, but would be rare. It would be irrational to set a goal to
reduce road traffic fatalities to zero while at the same time setting another goal to increase
road traffic fatalities to a number greater than zero, but cases of this kind are rare in practice.
Contingent inconsistency of goals may be more common, but it is doubtful whether such
inconsistencies ought to count as irrational. A goal to reduce road traffic fatalities to zero may
3
be in conflict with an accessibility goal to make people able to drive faster on the roads, or
safety may in some other way be in conflict with accessibility, but mere goal conflict is not a
sign of irrationality.
3. How should goal conflicts be rationally managed?
Transport is a policy area in which conflicts between goals are of particular concern.
Examples include goals of road safety and environmental protection. Such goals, primarily
concerned with the negative effects of traffic and transportation, often come into conflict with
accessibility or quality – goals primarily concerned with the positive effects of traffic and
transportation. In these contexts, some seemingly rather uncompromising attitudes have been
expressed. The Swedish road safety goal, for example, explicitly states that it is
“unacceptable” that anyone is killed or seriously injured as a consequence of road traffic
accidents (Gov. Bill 1996/97:137). Similar road safety goals have been adopted in Denmark
and Norway (Færdselssikkerhedskommissionen, 2002; Steinset et al., 2002). The widely
influential statement of the Precautionary Principle in the Rio Declaration of 1992 represents
a stringent attitude in assessments of uncertainties concerning large-scale environmental
impacts (UNCED, 1993). And the definition of sustainability as “development that meets the
needs of the present without compromising the ability of future generations to meet their own
needs” is an explicit statement of an uncompromising attitude (Brundtland Commission,
1987).
However, the word “compromising” in the notion of sustainability may be given several
different interpretations (cf. Chichilnisky, 1996; Heal, 1998). Likewise, the notion that some
consequences of road traffic accidents are “unacceptable” has been interpreted differently by
different authors; some have interpreted it as a lexicographical preference ordering, whereas
others have interpreted it simply as placing a high weight on road safety (cf. Elvik, 1999a;
Hårsman et al., 2002). There is therefore some uncertainty as regards the notion of
compromising and the notion of acceptability, and it is sometimes unclear what the notions
imply for the management of goal conflicts.
Even though a pragmatic interpretation of the road safety goal involves short-term
compromising with other goals, identifying the “optimal” level of accident risk poses a
particular problem: To what extent should a regulator manipulate the risk level, when
individual drivers themselves voluntarily choose risk levels according to their own tastes?
Drivers do not unconditionally minimize risks, but make trade-offs between safety and other
goods, such as the convenience of driving fast and arriving on time. For example: Mandatory
wearing of seat belts reduces the likelihood of death or injury in case an accident happens, but
makes people drive less cautiously (Wilde, 1994). It has also been shown that cars outfitted
with antilock brakes are driven faster, more carelessly, and closer to the car in front, braked
more abruptly, and have no lower accident rate per hour of exposure than cars without these
devices (Fosser et al., 1996; Sagberg et al., 1997). Cars fitted with studded tires, similarly, are
driven faster on slippery roads. Thus, studded tires only marginally effect accident rates
(Elvik, 1999b). Wider road design has been shown to increase people’s speed and lane
position, offsetting potential safety effects from changing road width (van Driel et al., 2004;
Lewis-Evans & Charlton, 2006). Last, investigations showed that driver training or a
mandatory course of driving on slippery roads does not reduce accident risk (Christensen &
Glad, 1996).
4
In addition to the evidence that individuals voluntarily choose to increase risks in order to
attain benefits of speed, there are empirical studies suggesting that individuals are often
unable to perceive risks adequately; that is, individuals tend to overcompensate risks when
they try to optimize them. For example, when lane markings were introduced to improve
nighttime road visibility, drivers increased their speed to the extent that they habitually
overdrove their headlamps, thus effectively worsening their safety (Cottrell, 1988; Kallberg,
1993; Rumar & Marsh, 1998). In the case of nighttime lane markings, the psychological
mechanism underlying such overcompensation has been studied. Owens and Tyrrell (1999)
suggested that lane guidance is mediated by visual mechanisms that are less adversely
affected by lower light levels, providing drivers with a false sense of their own visual
competence. As a consequence, drivers routinely overdrive their headlights because all
available feedback tells them they are driving at a safe speed.
The hypothesis that individuals generally optimize, rather than minimize, risks is stressed in
risk homeostasis theory (Starmer, 2000; Wilde, 1994). The hypothesis that risks are often
underestimated lends support from general observations suggesting mental heuristics leading
to large and persistant biases (Kahneman et al., 1982). Paper III combines these two
approaches. The paper argues that, under certain conditions, a “deterioration” of the
environment may affect people’s behaviour in such a way that the final outcome is beneficial.
Regulators may, for example, opt to make a roundabout smaller in order to make people drive
more slowly and cautiously.
4. The logic of goals
Goals play a central role in artificial intelligence and multiagent systems. Goal systems are
commonly used in existing agent programming languages, and discussed and analyzed in
formal agent theories (e.g. Broerson et al., 2002; da Costa Pereira & Tettamanzi, 2007;
Dastani & van der Torre, 2002; Rahwan & Amgoud, 2006; Shapiro et al., 2005). Along with
philosophical writings in the tradition of theory change (Alchourrón & Bulygin, 1971;
Alchourrón et al., 1985; BonJour, 1985; Gärdenfors, 1988; Hansson, 1999; Hansson &
Makinson, 1999; Levi, 1980; Levi, 1984), this provides a starting point for investigating and
formalizing concepts such as consistency, conflict, and coherence. Goals are similar to beliefs
in the sense that they can be expressed as sentences – in the former case as factual statements
that are believed to be true, and in the latter case as factual statements that some agent wants
to realize. Consequently, there are reasons to investigate epistemological topics in a context of
goals.
Allais (1953[1979]) stresses the concept of consistent goals. Millgram (2000) and Edvardsson
(2006) have investigated the concept of goal system coherence, though in a non-formal or
quasi-formal discussion. In the epistemological literature, it is generally agreed that
inconsistent sets of beliefs must be consolidated – i.e., made consistent (Fuhrmann, 1997;
Hansson, 1991; Olsson, 1997). There is also an ongoing discussion on how to characterize
and measure coherence of belief sets (Hansson, 2006; Hansson & Olsson, 1999; Lewis, 1946;
Olsson, 2002; Olsson, 2005; Shogenji, 1999).
One of the points stressed in Paper IV is that the above mentioned epistemological concepts
are only conceivable as relative to specific situations; more precisely, it does not make sense
to refer to a goal system as inconsistent or incoherent unless additional information about the
decision-maker’s situation is (implicitly or explicitly) assumed. Consequently, a complete
formal representation of goals must include a logic of action. Paper V involves constructing
5
such a logic. Writings on the logic of action is already prominent, originating with classical
writings of Aristotle, St. Anselm, and David Hume and continuing with contemporary
literature in philosophy and artificial intelligence (e.g. Bratman, 1984; Bratman, 2006;
Davidson, 1980; Giordano et al., 2000; Hughes et al., 2005; Hughes et al., 2007; Segerberg,
1992; Wooldridge, 2006; Zhang & Foo, 2001).
Since obligations and intentions are closely related to goals, there is a natural connection
between action logic and deontic logic. Standard deontic logic, as it is commonly referred to,
is based on a semantical construction of possible worlds; an obligation to realize some
sentence p holds, it is said, if and only if p is true in every possible “ideal” world (Føllesdal &
Hilpinen, 1970; Hansson, 1969; Hansson, 2001; von Wright, 1951; von Wright, 1956; von
Wright, 1968; von Wright, 1998). This gives rise to a number of paradoxes (Åqvist, 1994;
Åqvist, 1967; Prior, 1954; Prior, 1958; Ross, 1941) that can be avoided by deriving a deontic
logic from an action logic.
5. Summary of papers I–V
Paper I discusses the Swedish transport policy goals, and problem of combining policy goals
with welfare economics. On an instrumental account, goals are taken for granted as inputs to
decision problems, and decisions are assessed as rational to the extent that they result in that
goals are achieved. Social welfare, while frequently assessed as a criterion for rational tradeoffs between goals, is itself is an ambiguous goal. In Sweden, the overall goal for transport
policy is to provide a system that is both economically efficient and sustainable in the long
term. This overall goal is specified by six sub-goals: (1) an accessible transport system, (2) a
high standard of transport quality, (3) safe traffic, (4) a good environment, (5) positive
regional development, and (6) equal opportunities. The first two are concerned with
promoting the positive effects of the transport system – the purpose why there is a transport
system in the first place. The next two sub-goals, in contrast, are concerned with preventing
the negative effects of the transport system. Finally, the last two sub-goals express distributive
concerns – that positive and negative effects should be fairly distributed between regions and
between the sexes. The six goals often come into conflict and trade-offs must be rationally
and consistently managed.
Paper II (Rosencrantz et al., 2007), written together with Sven Ove Hansson and Karin
Edvardsson, discusses rationality in road safety policy. Sweden’s ambitious road safety goal,
“Vision Zero”, has met with much criticism. Among other things, the goal has been accused
of being illusory and irrational. The paper states that such accusations of irrationality have
been unnecessarily imprecise, since no reference is made to independently developed criteria
of rational goal-setting. The paper argues that the Swedish road safety goal is rational, since it
is action-guiding and achievement-inducing. The goal satisfies the criteria of precision,
evaluability, approachability, and motivity. Nevertheless, some problematic features – such as
the lack of temporal precision in the long-term goal and the difficulty of comparing fatalities
and serious injuries when evaluating performance – are identified and discussed.
Paper III includes a model of rational choice under risk with biased risk perception. On its
basis, it is argued that sometimes, a regulator aiming at maximizing social welfare should
raise the population’s risk exposure. It is shown that by deteriorating the environment – i.e. by
making it riskier – the regulator can motivate drivers to choose behavior that is less biased.
The conditions under which such a beneficial effect exists are investigated. The benefits of
such an intervention in a heterogeneous population are investigated.
6
Paper IV (Rosencrantz, 2008) provides a formal representation of goal systems. The focus is
on three properties: consistency, conflict, and coherence. An aim is to attain conceptual clarity
of these properties. It is argued that consistency is adequately regarded as a property relative
to the decision situation or, more specifically, the set of alternatives that the agent faces.
Moreover, as a condition of rationality, consistency is stronger than some writers have
claimed. Conflict is adequately regarded as a relation over subsets of a given goal system and
should likewise be regarded as relative to the set of alternative that the agent faces. Coherence
is given a probabilistic interpretation, based on a support relation over subsets of goal
systems.
Paper V provides an analysis of paradoxes in standard deontic logic. Two features of these
paradoxes are pointed out; (1) they involve inference of obligations with arbitrary disjuncts,
and (2) they involve failure to properly distinguish between actions and contexts. After an
initial presentation of these paradoxes and their sources, the paper constructs a logic of action.
A deontic predicate is derived, and it is observed that this predicate does not satisfy all of the
axioms of standard deontic logic. In violating these axioms, the predicate avoids the
paradoxes.
6. Afterthought
In Paper I and in Paper II, the Swedish transport policy goals are analyzed. Among other
topics, problems in policy goal-setting are identified and simplistic views of rational tradeoffs are challenged. Two questions may be posed regarding these two papers: Is this
philosophy? And in what way does philosophy make a constructive contribution? The easy
way to answer these questions is to say that, by virtue of being part of a thesis in philosophy,
the two papers are to be classified as philosophical papers and that the constructive
contribution may still be difficult to identify. But more can be said about the relation between
philosophical activity and practical problems. While the two papers may provide an
interesting summary of “philosophical” problems recognized by non-philosophers (or
philosophically oriented social scientists) who have analyzed the Swedish transport policy
goals, it might be argued that these goals are not suitable as objects for rigorous philosophical
scrutiny. The goals have been formulated by people with little philosophical background and
are not intended to be evaluated according to philosophical standards. They are intended to be
communicated primarily among non-philosophers, and to coordinate actions across agents
who likewise lack philosophical training. As such, the goals manifest a notion of rationality
that is arguably different from the philosophical notion. They are not to be taken literally, but
are mere means of communication between politicians on the one side and administrators or
the public on the other side, and it should be admitted that the goals are rhetorically
formulated. The question, then, is whether it is in any way more constructive to approach
problems in transport policy decision-making or goal-setting from a philosophical point of
departure than from the perspective of a social scientist with methods of empirical research
and faculties of critical thinking. While the answer to this question is pending, it may be said
that these two perspectives are not mutually exclusive; a philosopher may also be a scientist,
or a practitioner, and vice versa. Whether Paper I and Paper II are writings in the
philosophical tradition, and whether they address problems to which philosophy exclusively
has the potential to make a specific contribution, is not as important as whether the arguments
they present are of interest to philosophically oriented scholars in general. Neither is the
question whether a particular text is “philosophy” as important as the question whether
acquaintance with philosophical literature legitimizes claims of expertise, or whether
7
familiarity with philosophical concepts provides an advantage; while the former question
precedes the latter questions, it is of interest only in light of what could be said about the
benefits of doing philosophy.
In Paper III, a decision theoretical problem of relevance to transport policy is investigated.
The main argument is illustrated by some rather specific functions representing the possibility
frontier and the perception bias. It should be emphasized that the argument is not dependent
on assumptions that either of these function uniquely represents the decision problem. The
general problem consists in combining a set of plausible assumptions with the condition that
the net effect is beneficent. Let the probability triangle represent the set of pairs 〈p, q〉∈[0,
1]×[0, 1] such that p+q ≤ 1. With f being a real-valued function, and with q = f(p), the domain
of the probability triangle is defined as the set {p∈[0, 1] : p+f(p) ≤ 1}. Then the following
general conditions can be formulated:
(1) f is continuous over the domain of the probability triangle. (p. 3).
(2) f is continuously differentiable over the domain of the probability triangle (p. 3).
(3) f is a monotonically increasing function of p (pp. 3-4).
Since the possibility set demarcated by f is (strictly) convex (p. 4), it is also assumed that
(4) f is a (strictly) concave function over the domain of the probability triangle.
The assumption that the indifference curves are upward sloping, linear, and parallel follows
from the expected utility hypothesis (which in turn follows from the standard assumptions of
ordering, continuity, and independence):
• Let L be a lottery of the form {p, H; q, A; 1–p–q, ¬(A∨H)}.
• Then u(L) = p×u(H)+q×u(A)+(1–p–q)×u(¬(A∨H)).
• Then q = (u(¬(H∨A))+p×(u(H)–u(¬(H∨A)))–u(L))/(u(¬(H∨A))–u(A)).
• With u(A), u(H), and u(¬(H∨A)) fixed for each individual, and with u(L) fixed for
each indifference curve, it is then clear that the indifference curves are linear with the
slope k = dq/dp = (u(H)–u(¬(H∨A)))/(u(¬(H∨A))–u(A)); note that k is positive, since
u(A) > u(¬(A∨H)) > u(H).
As for the real-valued function b, each of the three specific functions examined in the paper
satisfies the following condition:
(5) p ≤ b(p) ≤ 1 over the domain [0, 1].
It is also noted in the paper (p. 14) that it is implausible that, for some interval, the agent
decreases her risk judgment with increasing risk. This corresponds to the following condition:
(6) 0 ≤ db/dp over the domain [0, 1].
Two definitions provide the central concepts of the paper:
• A pair of real-valued functions 〈fα, fβ〉 is a risk increase if and only if fα(p) > fβ(p) for
all p such that p+fα(p) ≤ 1.
• A pair of real-valued functions 〈fα, fβ〉 is kb-beneficial if and only if fα(b(pα*)) <
fβ(b(pβ*)), where pα* and pβ* are such that fα′(pα*) = fβ′(pβ*) = k.
The general argument is that there exists a triple of real-valued functions 〈fα, fβ, b〉 and a
positive number k such that:
(a) each of fα and fβ satisfies the conditions (1)-(4),
(b) b satisfies the conditions (5)-(6), and
(c) 〈fα, fβ〉 is a kb-beneficial risk increase.
Since this theorem concerns the existence of a beneficial risk increase under a set of plausible
conditions, the argument does not fall with the fact that the observations reported in Table 1
(p. 15) concern specific functions.
8
In Paper IV and Paper V, formal theories based primarily on belief dynamics and deontic
logic are developed. While the focus is on goals and obligations, respectively, the two topics
are related; a goal may be interpreted as an obligation to realize a particular state-of-affairs, in
the sense that an agent who has a goal ought to achieve it. Hence, the framework in Paper V
may be suitable for analyzing and developing the concepts formalized in Paper IV. In
particular, the support relation defined in Section 5 of Paper IV may benefit from a more
refined logic of action. In this way, the notion of normative coherence could be further
developed. A suggestion for further work in this area will briefly be sketched out here. In
order to clarify the concepts formalized in Paper IV, the framework provided in Paper V can
be expanded such that the decision problem is modeled as a simple belief network. In
particular, this can provide a clearer account of the notion in Definition 5.1 that an agent
“attempts to realize” a set of sentences. Let ℒ0, ℒ1, and ℒ be defined as in Section 3.1 of
Paper V, and let S be a subset of ℒ. Let p be a function from S′ to [0, 1], where S′ is a set such
that
(1) If α∈ℒ0∩Cn(S), then α∈S′.
(2) If m(φ, α, β)∈Cn(S), then m(φ, α, β)∈S′.
(3) S′ is closed under the truth-functional operations ¬ and ∧.
(4) Nothing is an element of S′, except by virtue of (1)–(3).
Thus, given the additional assumption that p is a probability measure satisfying the usual
axioms, the decision problem of Paper IV can be represented by the pair 〈S, p〉. In network
terminology, each counterfactual outcome β is a child of at least three parents: an action φ, a
fact α, and a subjunctive conditional m(φ, α, β) stating that the outcome would result from
performing the action when the fact holds. The conditional probability that β would be the
outcome of performing the action φ can therefore be calculated as PS(β|φ) = ∑{p(α∧m(φ, α,
β)) : α∈T}, where T is a set such that
(1) {Cn(α) : α∈T} = ℒ1∩S′↓⊥. (The operator Cn is a map from T to the set of maximally
consistent subset of ℒ1∩S′.)
(2) If α,α′∈T, then Cn({α}) ≠ Cn({α′}). (There is no pair of logically equivalent
sentences in T.)
The notion that an agent “attempts to realize” a set X of sentences, denoted att(X) in
Definition 5.1, can now be defined as the disjunction of those actions that maximize the
probability of realizing X; a plausible interpretation of this notion is that the agent performs
one action among those actions that maximize the probability of realizing X. This can be
formally expressed as attS(X) = ∨{φ∈AS : (∀φ′∈AS)(PS(∧X|φ′) ≤ PS(∧X|φ))}, where AS =
{φ∈ℒ1 : (∃α,β∈ℒ1)(m(φ, α, β)∈S′)}. In the special case where there is a unique action that
maximizes the probability of realizing X, attS(X) will of course be equal to this action. Section
5 of Paper IV can now be reformulated with P(∧Y|att(X)) = PS(∧Y|attS(X)). This suggested
course for further work is yet to be elaborated. The main benefits of Paper V may therefore be
that it provides a framework preceding further work relating to the concepts formalized at the
end of Paper IV, and the results regarding paradoxes in Standard Deontic Logic may turn out
to be but a spinoff in the work towards a complete account of goal system coherence.
In short, much more can be done. While the problems addressed in this thesis are indeed
intriguing and leave many new questions to be answered, however, it is time to conclude;
further work might have to wait, but hopefully not for too long.
9
References
Adler, M.D. and E.A. Posner (1999) “Rethinking cost-benefit analysis”, Yale Law Journal,
109(2), pp. 165-247.
Alchourrón, C.E. and E. Bulygin (1971) Normative Systems. Springer, Vienna.
Alchourrón, C.E., P. Gärdenfors, and D. Makinson (1985) “On the logic of theory change:
Partial meet contraction and revision functions”, Journal of Symbolic Logic, 50(2):510-530.
Allais, M. (1953[1979]) “The foundations of a positive theory of choice involving risk and a
criticism of the postulates and axioms of the American school”, in Allais, M. and O. Hagen
(eds.), Expected Utility Hypotheses and the Allais Paradox. D. Reidel, Dordrecht.
Åqvist, L. (1967) “Good Samaritans, contrary-to-duty imperatives, and epistemic
obligations,” Noûs, 1:361-379.
Åqvist, L. (1994) “Deontic Logic,” in D. Gabbay and F. Guenthner, eds., Handbook of
Philosophical Logic: Volume II Extensions of Classical Logic, Kluwer.
BonJour, L. (1985) The Structure of Empirical Knowledge. Harvard University Press,
Cambridge (Mass.).
Bratman, M. (1984) “Two Faces of Intention”, Philosophical Review, 93:375-405.
Bratman, M. (2006) Structures of Agency, Oxford University Press, Oxford.
Broersen, J., M. Dastani, J. Hulstijn, and L. van der Torre (2002) “Goal generation in the
BOID architecture”, Cognitive Science Quarterly Journal, 2:428-447.
Brundtland Commission (World Commission on Environment and Development) (1987) Our
Common Future. Oxford University Press, Oxford.
da Costa Pereira, C. and A.G.B. Tettamanzi (2007) “Goal generation with ordered
beliefs”, pp. 133-144 in Basili, R. and M.T. Pazienza (eds.) AI*IA 2007: Articial Intelligence
and Human-Oriented Computing. Springer.
Chichilnisky, G. (1996) “An axiomatic approach to sustainable development”, Social Choice
and Welfare, 13(2):231-257.
Christensen, P. and A. Glad (1996) “Mandatory course of driving on slippery roads does not
reduce the accident risk”, Nordic Road and Transport Research, 3.
Cottrell, B.H. (1988) “The effects of wide edgelines on two-lane rural roads”, Transportation
Research Record, 1160:35-44.
Dastani, M. and L. van der Torre (2002) “Specifying the merging of desires into goals in the
context of beliefs”, pp. 824-831 in Shafazand, M.H. and A.M. Tjoa (eds.) EurAsia ICT 2002:
Proceedings of the First Eurasian Conference on Advances in Information and
Communication Technology. Springer, Berlin.
10
Davidson, D. (1980) Essays on Actions and Events, Oxford University Press, Oxford.
van Driel, C.J.G., R.J. Davidse, and M.F.A.M. van Maarseveen (2004) “The effects of an
edgeline on speed and lateral position: a meta-analysis”, Accident Analysis and Prevention,
36(4):671-682.
Edvardsson, K. (2006) How to Set Rational Environmental Goals: Theory and Applications,
Licentiate thesis. Royal Institute of Technology, Stockholm.
Edvardsson, K. and S.O. Hansson (2005) “When is a goal rational?”, Social Choice and
Welfare, 24(2):343-361.
Elvik, R. (1999a) “Can injury prevention efforts go too far? Reflections on some possible
implications of Vision Zero for road accident fatalities”, Accident Analysis and Prevention,
31(3):265-286.
Elvik, R. (1999b) “The effects on accidents of studded tires and laws banning their use: a
meta-analysis of evaluation studies”, Accident Analysis and Prevention, 31:125-134.
Elvik, R. (2003) “How would setting policy priorities according to cost-benefit analyses affect
the provision of road safety?”, Accident Analysis and Prevention, 35(4):557-570.
Færdselssikkerhedskommissionen [Danish Road Safety Commission] (2002) Hver Ulykke ér
en for Meget. Trafiksikkerhed Starter med Dig [Every Accident is one too Many. Traffic
Safety Starts with You]. Trafikministeriet, Copenhagen.
Føllesdal, D., and R. Hilpinen (1970) “Deontic logic: an introduction,” pp.1-35 in R. Hilpinen
(ed.) Deontic Logic: Introduction and Systematic Readings, Reidel, Dordrecht.
Fosser, S., F. Sageber, and A.F. Sætermo (1996) “Behavioural adaptation to anti-lock brakes
– drivers keep a shorter distance”, Nordic Road and Transport Research, 8:21-22.
Giordano, L., A. Martelli, and C. Schwind (2000) “Ramification and causality in a modal
action logic,” Journal of Logic and Computation, 10(5):615-662.
Gov. Bill 1996/97:137 Nollvisionen och det Trafiksäkra Samhället [Vision Zero and the
Traffic Safe Society].
Gov. Bill 1997/98:56 Transportpolitik för en Hållbar Utveckling [Transport Policy for a
Sustainable Development].
Gov. Bill 2001/02:20 Infrastruktur för ett Långsiktigt Hållbart Transportsystem
[Infrastructure for a Long-Term Sustainable Transport System].
Griffin, J. (1986) Well-Being. Its Meaning, Measurement, and Moral Importance. Clarendon
Press, Oxford.
Gärdenfors, P. (1988) Knowledge in Flux. Modeling the Dynamics of Epistemic States. MIT
Press, Cambridge (Mass.).
11
Hansson, B. (1969) “An analysis of some deontic logics”, Noûs, 3:373-398.
Hansson, S.O. (1999) A Textbook of Belief Dynamics. Theory Change and Database
Updating. Kluwer Academic Publishers, Dordrecht.
Hansson, S.O. (2001) The Structure of Values and Norms, Cambridge University Press,
Cambridge.
Hansson, S.O. (2004) “Welfare, justice, and Pareto efficiency”, Ethical Theory and Moral
Practice, 7(4):361-380.
Hansson, S.O. (2006) “Coherence in epistemology and belief revision”, Philosophical
Studies, 128:93-108.
Hansson, S.O. and D. Makinson (1997) “Applying normative rules with restraint”, pp. 313332 in Dalla Chiara, M.L. et al. (eds.) Logic and Scientific Methods. Kluwer Academic
Publishers, Dordrecht.
Hansson, S.O. and E.J. Olsson (1999) “Providing foundations for coherentism”, Erkenntnis,
51:243-265.
Hårsman, B., C. Persson, and S. Pädam (2002) Metod för Mål-/Medelanalys I
Infrastrukturplaneringen [Method for Goal/Means Analysis in Infrastructure Planning].
Inregia AB, Stockholm.
Heal, G. (1998) Valuing the Future, Economic Theory and Sustainability. Economics for a
Sustainable Earth. Columbia University Press, New York.
van Herten, L.M. and L.J. Gunning-Schepers (2000) “Targets as a tool in health policy. Part I:
lessons learned”, Health Policy, 52:1-11.
Hubin, D.C. (1994) “The moral justification of benefit/cost analysis”, Economics and
Philosophy, 10:169-194.
Hughes, J., A. Esterline, and B. Kimiaghalam (2005) “Means-end relations and a measure of
efficacy,” Journal of Logic, Language and Information, 15:83-108.
Hughes, J., P. Kroes, and S. Zwart (2007) “A semantics for means-end relations,” Synthese,
158:207-231.
Kallberg, V.-P. (1993) “Reflector postsigns of danger?”, Transportation Research Record,
1403:57-66.
Kahneman, D., P. Slovic, and A. Tversky (eds.) (1982) Judgment under Uncertainty.
Cambridge University Press, Cambridge.
Kommunikationsministeriet (2008) Riktlinjer för Trafikpolitiken och Utvecklings- och
Finansieringsprogram för Transportnätet fram till 2020 [Directions for Traffic Policy and
Development and Financing Program for the Transport System until 2020], Statsrådets
trafikpolitiska redogörelse till riksdagen, Helsinki.
12
Levi, I. (1980) The Enterprise of Knowledge. MIT Press, Cambridge (Mass.).
Levi, I. (1984) Decisions and Revisions, Cambridge University Press, Cambridge.
Levi (1986) Hard Choices. Decisions Under Unresolved Conflict. Cambridge University
Press, Cambridge.
Levi, I. (1997) The Covenant of Reason. Rationality and the Commitments of Thought.
Cambridge University Press, Cambridge.
Lewis, C.I. (1946) An Analysis of Knowledge and Valuation. Open Court, LaSalle (Ill.).
Lewis-Evans, B. and S.G. Charlton (2006) “Explicit and implicit processes in behavioural
adaptation to road width”, Accident Analysis and Prevention, 38(2):324-334.
Luce, D. and H. Raiffa (1957) Games and Decisions. Wiley, New York.
Mattsson, B. (2005) Kostnads-Nyttoanalys. Värdegrunder, Användbarhet, Användning [CostBenefit Analysis. Value Bases, Applicability, Application]. Räddningsverket, Karlstad.
Mele, A.R. and P. Rawling (eds.) (2004) The Oxford Handbook of Rationality. Oxford
University Press, New York.
Millgram, E. (2000) “Coherence: The price of the ticket”, Journal of Philosophy, 97(2):82-93.
Nihlén Fahlquist, J. (2005) Moral Responsibility in Traffic Safety and Public Health,
licentiate thesis, Royal Institute of Technology, Stockholm.
Nihlén Fahlquist, J. (2006a) “Responsibility ascriptions and Vision Zero”, Accident Analysis
and Prevention, 38:1113-1118.
Nihlén Fahlquist, J. (2006b) “Responsibility ascriptions and public health problems. Who is
responsible for obesity and lung cancer?”, Journal of Public Health, 14(1):15-19.
Nihlén Fahlquist, J. (2007) Moral Responsibility and the Ethics of Traffic Safety, doctoral
thesis, Royal Institute of Technology, Stockholm.
Olsson, E.J. (2002) “What is the problem of coherence and truth?”, The Journal of
Philosophy, 99:246-272.
Olsson, E.J. (2005) Against Coherence. Truth, Probability, and Justification. Oxford
University Press, Oxford.
Owens, D.A. and R.A. Tyrrell (1999) “Effects of luminance, blur, and age on nighttime visual
guidance: a test of the selective degradation hypothesis”, Journal of Experimental
Psychology: Applied, 5(2):115-128.
Parfit, D. (1984) Reason and Persons. Clarendon Press, Oxford.
13
Peden, M., R. Scurfield, D. Sleet, D. Mohan, A.A. Hyder, E. Jarawan, and C. Mathers (eds.)
(2004) The World Report on Road Traffic Injury Prevention. World Health Organization,
Geneva.
Popper, K. (1957) The Poverty of Historicism. Routledge, London.
Prior, A.N. (1954) “The paradoxes of derived obligation”, Mind, 63:64-65.
Prior, A.N. (1958) “Escapism”, pp.135-146 in A.I. Melden (ed.) Essays in Moral Philosophy,
University of Washington Press, Seattle.
Rahwan, I. and L. Amgoud (2006) “An argumentation based approach for practical
reasoning”, pp. 347-354 in Stone, P. and G. Weiss (eds.) Proceedings of the Fifth
International Joint Conference on Autonomous Agents and Multiagent Systems AAMAS
'06. ACM, New York.
Rawls, J. (1971) A Theory of Justice. Oxford University Press, Oxford.
Resnik, M. (1987) Choices. An Introduction to Decision Theory. University of Minneapolis
Press, Minneapolis.
Rosencrantz, H. (2008) “Properties of goal systems: consistency, conflict, and coherence”,
Studia Logica, 89:37-58.
Rosencrantz, H., K. Edvardsson, and S.O. Hansson (2007) “Vision Zero – is it irrational?”,
Transportation Research Part A: Policy and Practice, 41(6):559-567.
Ross, A. (1941) “Imperatives and logic”, Theoria, 7:53-71.
Rubenstein, A. (1998) Modeling Bounded Rationality. MIT Press, Cambridge (Mass.).
Rumar, K. and D.K. Marsh (1998) “Lane markings in night driving: a review of past research
and of the present situation”, UMTRI-98-50, University of Michigan Transportation Research
Institute, Ann Arbor.
Sagberg, F., S. Fosser, and I.F. Sætermo (1997) “Investigation of behavioral adaption to
airbags and antilock brakes among taxi drivers”, Accident Analysis and Prevention, 29:293302.
Shapiro, S., Y. Lespérance, and H.J. Levesque (2005) “Goal change”, pp.582-588 in
Pack Kaelbling, L. and A. Saffiotti (eds.) IJCAI-05. Proceedings of the Nineteenth
International Joint Conference on Artificial Intelligence. Professional Book Center.
Shogenji, T. (1999) “Is coherence truth-conducive?”, Analysis, LIX(4):338-345.
Segerberg, K. (1992) “Getting started: beginnings in the logic of action,” Studia Logica,
51:347-378.
Sen, A. (2002) Rationality and Freedom. Belknapp Press of Harvard University, London.
14
Simon, H. (1983) Reason in Human Affairs. Basil Blackwell, Oxford.
Simon, H. (1997) Models of Bounded Rationality. Volume 3, Empirically Grounded Economic
Reason. MIT Press, Cambridge (Mass.).
Smith, S.C and M.P. Todaro (2006) Economic Development, 9th ed. Pearson/Addison Wesley,
Harlow.
Spohn, W. (2002) “The many facets of the concept of rationality”, Croatian Journal of
Philosophy, 2:247-262.
Starmer, C. (2000) “Developments in non-expected utility theory”, Journal of Economic
Literature, 38(2):332-382.
Steinset, B.K., P.O. Wanvik, I. Handorsen, A.K. Midtgaard, R. Muskaug, and B. Skaugset
(2002) Nasjonal Handlingsplan for Trafikksikkerhet på Veg 2002-2011 [National Strategy
Plan for Road Safety 2002-2001]. Vegdirektoratet, Oslo.
UNCED (1993) The Earth Summit. The United Nations Conference on Environment and
Development (1992: Rio de Janeiro). Introduction and commentary by Stanley P. Johnson.
Graham & Trotman, London.
UNDP (2000) Human Poverty Report, 2000. United Nations, New York.
Wetmore, J. (2003) Systems of Restraint: Redistributing Responsibilities for Automobile
Safety in the United States Since the 1960s. Cornell University, New York.
Wilde, G.J.S. (1994) Target Risk: Dealing with the Danger of Death, Disease and Damage in
Everyday Decisions. PDE Publications, Toronto.
Wooldridge, M. (2000) Reasoning about Rational Agents, MIT Press, Cambridge.
von Wright, G.H. (1951) “Deontic logic”, Mind, 60:1-15.
von Wright, G.H. (1956) “A note on deontic logic and derived obligation” , Mind, 65:507509.
von Wright, G.H. (1968) “An essay in deontic logic and the general theory of action”, Acta
Philosophica Fennica, 21:1-110.
von Wright, G.H. (1998) “Deontic logic – as I see it”, Paper presented at the Fourth
International Workshop on Deontic Logic in Computer Science, Bologna.
Zeitler, U. (1997) Transport Ethics: An Ethical Analysis of the Impact of Passenger Transport
on Human and Non-Human Nature. Aarhus University, Aarhus.
Zhang, D., and N.Y. Foo (2001) “EPDL: a logic for causal reasoning”, pp.131-138 in
Proceedings of the Seventeenth International Joint Conference on Artifical Intelligence,
IJCAI 2001, Seattle, Washington, USA, August 4-10.
15
Zimmer, M. (2005) “Surveillance, privacy and the ethics of vehicle safety communication
technologies”, Ethics and Information Technology, 7(5):201-210.
16