Available online at www.sciencedirect.

com

Safety Science 46 (2008) 11371148 www.elsevier.com/locate/ssci

Critical infrastructures and responsibility: A conceptual exploration

Koos van der Bruggen *,1
Delft University of Technology, Faculty of Technology, Policy and Management, Department of Philosophy, Jaalaan 5, 2628 BX Delft, The Netherlands Received 13 December 2006; received in revised form 8 May 2007; accepted 14 June 2007

Abstract In this article some considerations are presented as a starting point for judging responsibility for infrastructural systems from a moral point of view. Infrastructural systems are essential for present day high-tech society. Without good working infrastructural systems people would hardly be able to survive. The importance and relevance of good working critical infrastructures for states and other public authorities are illustrated with some examples. Next, the meaning of the concept of responsibility in relation to infrastructural systems is developed. Outcome responsibility and remedial responsibility developed in philosophy of law and political philosophy are elaborated as useful concepts for dealing with responsibilities in infrastructural systems. This analysis is applied to the players on the infrastructural eld: governments and other public authorities, non-state institutional actors and last but not least individual agents. In a last paragraph some conclusions are drawn, which can be applied for apportioning responsibility for infrastructures in practice. 2007 Elsevier Ltd. All rights reserved.
Keywords: Critical infrastructure; Outcome responsibility; Remedial responsibility; Operation; Management; Safety

1. Introduction In ethical literature a lot has been written on (moral) responsibility. It certainly is one of the key issues in ethical theory. In this article I will present some considerations that can be a starting point for judging responsibility for infrastructural projects and activities from a moral point of view. First, attention will be given to infrastructures. What are the so-called critical infrastructures? What is their importance? What is their function in present day society? What place do they have and what place should they have in social and political arrangements? Next, the meaning of the concept of responsibility regarding the management and operation of critical infrastructures will be further developed. What does responsibility mean in relation to critical infrastructures?
*

Present address: Leiden University/Campus The Hague, Lange Houtstraat 5-7, 2511 CV DEN HAAG. Tel.: +31 70 3021070. E-mail addresses: losoe@tbm.tudelft.nl, kvanderbruggen@campusdenhaag.nl, k.vanderbruggen@hetnet.nl Tel.: +31 15 27 85143.

0925-7535/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.ssci.2007.06.003

1138

K. van der Bruggen / Safety Science 46 (2008) 11371148

Whose responsibility is at stake? Do all players in the eld have the same responsibility? What are the practical or theoretical problems in dening and applying responsibility for infrastructures? Outcome responsibility and remedial responsibility will be elaborated as useful concepts for dealing with responsibilities in infrastructural systems. A third paragraph will be devoted to the players on the infrastructural eld. These are in the rst place governments and other public authorities. Furthermore, non-state institutional actors: infrastructural companies and organizations. And last but not least the individual actors in these elds. What are their roles and their responsibilities? In a last paragraph some conclusions will be drawn, which can be applied for apportioning responsibility for critical infrastructures in practice. 2. Critical infrastructures Infrastructures can be described as the basic facilities, services, and installations needed for the functioning of a community or society, such as transportation and communications systems, water and power lines, and public institutions including schools, post oces and prisons (http://www.thefreedictionary.com/ infrastructure). This is a rather broad denition, including schools, prisons, etc. So, the concept is not limited to technical systems alone, but it leaves no doubt that the technical basis is essential for many infrastructural systems. In this article attention will be limited to the so-called critical infrastructures that have a (strong) technical component. The safety and reliability of technical provisions are important if not necessary conditions for the functioning of most infrastructural systems. This implies a special responsibility for the people and the institutions that are in charge of the task to design and apply the technical artefacts that are part of an infrastructural system. The concept of critical infrastructures has become a key concept in recent discussions. Since a few years there is a Journal for Critical Infrastructures. This journal denes critical infrastructures as networks for the provision of telecommunication and information services, energy services (electrical power, natural gas, oil and heat), water supply, transportation of people and goods, banking and nancial services, government services and emergency services. (http://www.inderscience.com/browse/index.php?journalCODE=ijcis). This is a limitative denition, which covers infrastructures so vital that their incapacitation or destruction would have a debilitating impact on defense or economic security, as it is described in a report for the US government (Mote et al., 2003; Critical Infrastructures: What Makes an Infrastructure Critical? Report to Congress). Western societies have become more and more dependent of these critical infrastructures, but at the same time we are witnessing a paradoxical development (De Bruijne, 2006, p. 11). De Bruijne shows that on the one hand critical infrastructures are increasingly interconnected into a single system of unimaginable proportions. But on the other hand these large-scale infrastructures have become institutionally fragmented. There is no longer any single organization (be it public or private) that can compel others to act. This fragmentization has also occurred within the separate infrastructural sectors, such as telecommunication and energy, where many competitive actors are playing a role. Moreover, these actors are no longer acting within one country, but cross-national: German ICT-operators have taken parts of the market in France, and Dutch railways are active in Poland. Another development that De Bruijne describes is the interconnectedness of critical infrastructures (De Bruijne, p. 12). The functioning of all infrastructures becomes more and more dependent from ICT-performance and from the electricity system. The threatening slogan of the labour unions for Dutch railways-workers in the beginning of the 20th century is more true than ever before: this whole cogwheels system will fall down if you want it to happen. Nowadays, a simple computer failure can lead to the laming of railway trac, so safety cannot be guaranteed anymore. This happened more than once in the Netherlands in the past few years. Disturbances in the ICT-system or in the energy net indeed can lame a society, not only at a national scale but also across borders. Well known examples are the fall-out in great parts of the North-East of the United States and in Canada in August 2003, due to a chain reaction in powerplants. The same happened in parts of Germany, The Netherlands, Belgium, France and even Spain, Portugal and Italy in November 2006. A

K. van der Bruggen / Safety Science 46 (2008) 11371148

1139

disturbance in Germany here also leads to a chain reaction. These developments show that the functioning of infrastructural systems no longer is a case for national governments alone. This means that questions of responsibility also get more and more an international character. Without good working infrastructural systems as electricity, water supply, railways, and ICT-telecom people would hardly be able to survive in our high-tech society! This leads to an interesting paradox: the better infrastructural systems are working, the more dependent we are from these systems and the less we are able to cope with the ever more exceptional failures of the system (Rathenau Instituut, 1994). In many so-called Third World countries it is normal that electric power is interrupted a few times a week. People are prepared for it and go on with their daily work. But if in our society the electric power system is disturbed, life comes to a standstill, as was recently (2005) seen in Haaksbergen (The Netherlands) where electric power had fallen down during more than 48 h because of heavy snow and storm. It almost led to a state of emergency. Because of the crucial importance of critical infrastructures, it is no coincidence that in times of war often assaults are planned on infrastructural systems as electric power plants, airports or railways nodes. Remember in the Yugoslavia war the destruction of the famous Mostar bridge (belonging to the Cultural Heritage of Humanity) in 1993 and in the Kosovo war (1999) the bombing of television studios in Beograd. It is hoped to destabilize a country with such assaults. For the same reason the conict between Russia and Ukraine in the beginning of 2006 on gas prizes was seen as a very serious threat for the state of Ukraine and even for some EU countries such as Poland, because their gas was delivered via Ukraine. So, again, it leaves no doubt that infrastructural systems are essential for the daily life of citizens. But an even stronger statement can be made: nowadays, critical infrastructures belong to the constitutive elements of a society or even a nation as such. Is it still possible to speak of a state or a society if the most important infrastructures have fallen down? Schneider and Ja ger refer to Herbert Kruger who wrote that there is no state without infrastructures (Schneider and Ja ger, 2003, p. 15). And indeed, if state authorities cannot provide directly or indirectly by private or semi-public companies infrastructural services and products as energy or drinking water a state of anarchy and chaos is very near. Reasoning from the other way around it can be argued that society or nation building is greatly helped or perhaps even only made possible by the development of infrastructures. By using the metaphor of the sovereign (state) as the Leviathan (Hobbes) infrastructures sometimes are labelled as the nerves and blood circulation of the state. This can be illustrated with some examples. It is a well known fact that water boards are the eldest democratic political institutions in The Netherlands, elder than parliament. These water boards were needed to arrange the (infrastructural) water system. Nowadays, the famous Dutch polder model still reminds to the narrow relationship between infrastructures and a good functioning democratic system.2 The initiating and constitutive role of infrastructures in the development of political institutions can also be shown at the cross-national level of European cooperation and integration. Already in the 19th century international regulations regarding the Rhine were developed. And the starting point of European integration cannot be untied from the rst intergovernmental organization with an infrastructural task: the European Coal and Steel Community (ECSC). The ECSC Treaty creates a framework of production and distribution arrangements for coal and steel and sets up an autonomous institutional system to manage it. Although its remit is limited to the two branches of industry, the ECSC has had a crucial impact on major economic and political developments in Europe for almost fty years (http://www.eu.int/ecsc/results/indexen.htm). The success of ECSC contributed a few years later to the foundation of the European Community of six nations, which became the starting point of the present day European Union. A third painful illustration of the importance of (critical) infrastructures for the political system is the saying that whatever was wrong in the Germany of Hitler and the Italy of Mussolini trains were departing and riding on schedule. And it was the Nazis who developed the famous German highway infrastructure. History shows that a government that succeeds in taking good care for infrastructures, gets appreciation for this, despite its policy in other elds (such as human rights).

2 The Dutch word polder refers to a part of land that is won from the sea. It is lying below sea level and is protected by dykes. All inhabitants have to work together to protect the land from the sea. This working together and looking for consensus also is a characteristic of the political system in the Netherlands. So the name of polder model.

1140

K. van der Bruggen / Safety Science 46 (2008) 11371148

The importance of good working critical infrastructures as a fundamental (even constitutive) condition for states and other public authorities has not been given as much attention as other essential state duties, but it surely is endorsed by political theorists. One of them is perhaps surprisingly Adam Smith. He writes that besides defence and judiciary tasks the third and last duty of the sovereign or commonwealth is that of erecting and maintaining those public institutions and those public works, which, though they may be in the highest degree advantageous to a great society, are, however, of such a nature, that the prot could never repay the expense to any individual or small number of individuals, and which it therefore cannot be expected that any individual or small number of individuals should erect or maintain. (Smith, 1812 [1776], pp. 570571). Carefully reading these quotes one learns that there is some ambiguity in it. On the one hand Smith talks about a duty of the sovereign or commonwealth (= public authority), but on the other hand the reason is that no prot is to be expected from these activities. Implicitly, this means that if prot is a possible result of erecting and maintaining infrastructural works, these can also and perhaps in his liberal view should be undertaken by individuals or private organizations. Since the 19th century taking care of infrastructural systems has become more and more a state task. Schneider and Ja ger summarize that public responsibility for infrastructures is based on the following arguments (Schneider and Ja ger, 2003, p. 15; Hermes, 1998, p. 271): (a) Technical eciency aspects: This relates to, e.g. standardization or coordination. State commitment is not necessary here. A private monopoly is also an option, but if that does not work, state has to intervene. (b) Military, state and integrationist motives: National security is an important argument for state involvement in infrastructures. (c) The demand of state resource expropriation rights and scal motives: Direct involvement or ownership by the state can be protable for the state budget. (d) Correction of market failure and public control: State involvement is often justied with the argument that it makes infrastructural provisions accessible and payable for all groups in society. (e) Guarantee of extensive supply (socio-political motives): The state should take care of the supply of public infrastructural goods that are scarce. (f) Democracy political, legal and environmentalist considerations: Some basic rights and basic needs are more or less dependent from infrastructural systems (e.g. communication). State commitment can be necessary to guarantee these basic rights. Of course it is possible to amend or comment most of these arguments. Partly, they have an empirical character: they refer to how and why decisions have been taken in more and less recent history. But certainly there is a reference to normative arguments also: that management, organization and possession of infrastructures should be in the hands of the public sector (politicians and government), because of the crucial importance of these infrastructure for this public domain. There is one more argument to be added, which is only partial and indirectly captured by Schneider and Ja ger: guaranteeing the security and safety of infrastructures is in last resort also a responsibility for a public authority. In theory, but more important in practice many things can go wrong. Most failures belong to one of the following categories:  Malfunctioning of the system: electricity breakdown, delay or falling out of trains or buses, no water supply, pollution of water, breakdown of telephone or Internet systems, etc.  Accidents: res, explosions, collisions, oods, etc.  Shortage of resources: shortage of fuel, shortage of water.  Personal problems: shortage of labour force, not enough well educated sta, strikes.  Vulnerability for criminal or terrorist attacks: bombing of power plants, poisoning water sources, spamming, spreading computer viruses, assaults on planes and/or trains, etc. How responsibility for the maintenance, functioning, safety and security of critical infrastructures should be dened and apportioned to dierent actors will be elaborated in the next paragraphs.

K. van der Bruggen / Safety Science 46 (2008) 11371148

1141

3. Concepts of responsibility David Miller states that in moral and political philosophy few concepts are more slippery than that of responsibility (Miller, 2001, p. 455). In this paragraph some slippery aspects, which are relevant in the discussion on responsibility for critical infrastructures, will be elaborated. By way of introduction a story, told by H.L.A. Hart: As captain of the ship, X was responsible for the safety of his passengers and crew. But on his last voyage he got drunk every night and was responsible for the loss of the ship with all [others] aboard. It was rumoured that he was insane, but the doctors considered that he was responsible for his actions. Throughout the voyage he behaved quite irresponsibly, and various incidents in his career showed that he was not a responsible person. He always maintained that the exceptional winter storms were responsible for the loss of the ship, but in the legal procedures brought against him he was found criminally responsible for his negligent conduct, and in separate civil proceedings he was held legally responsible for the loss of life and property. He is still alive and he is morally responsible for the deaths of many women and children (Hart, 1968, p. 11, my italics) (Table 1). In this story the following meanings of responsibility can be discerned (Bovens, 1998, pp. 2426): 1. Being the cause of a situation or an event (the captain was responsible for the loss of the ship). 2. Having the capacity to initiate or prevent a situation or an event (he was responsible for his actions). 3. Being accountable or liable for something that has happened (he was found legally and criminally responsible). 4. Having to full a task or obligation (he was responsible for the safety of his passengers). 5. Having (or not having) a personal virtue (he was not a responsible person). What is the relevance of each of these aspects of responsibility in relation to infrastructural systems? Responsibility as task or obligation (4) surely has a high relevance. In this paper the position is taken in line with Schneider and Ja ger that taking care of critical infrastructures belongs to the tasks of public authorities: it is a public responsibility. But that is not the same as to say that public authorities should have a direct involvement in the operation and management of infrastructural systems. Of course also having the capacity (2) is an important aspect. It should be evident that people and organizations possess the capacities to perform their infrastructural tasks conform the valid guidelines and rules. Responsibility as (not) having a personal virtue (5) may become relevant as far as individual persons have to execute operational and or managerial tasks. Last but not least responsibility in the sense of accountability or liability (3) surely also is a relevant aspect in relation
Table 1 Denitions of responsibility and related concepts Responsibility Being the cause of a situation or an event Having the capacity to initiate or prevent a situation or an event Being accountable or liable for something that has happened Having to full a task or obligation Having (or not having) a personal virtue Outcome responsibility Being responsible for the good and the harm we bring about by what we do Remedial responsibility Having a special obligation to put a bad situation right Accountability A relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgment, and the actor may face consequences Liability Duty, obligation, debt, responsibility, or hazard arising by way of contract, tort or statute Strict liability Legal doctrine that makes persons responsible for damages their actions or products cause, regardless of any fault on their part

1142

K. van der Bruggen / Safety Science 46 (2008) 11371148

to the management and operation of infrastructures. It should be clear who is accountable or liable when things go wrong (which is not necessarily the same as having causal responsibility (1)). A problem in all complex organizations and systems, and thus also in critical infrastructures, is that it is not always easy to nd out who is or who are responsible. This is known as the many hands problem. The problem has been described by Dennis Thompson in 1980: Because many dierent ocials contribute in many ways to decisions and policies of government, it is dicult, even in principle to identify who is morally responsible for political outcomes (Thompson, 1980, p. 905). Thompson is writing about responsibility in political processes, but this is also a problem in managing and operating infrastructural systems. Bovens (1998) makes a comparison with developments in law to treat or even to solve the many hands problem. In law legal bodies or legal personals (which are not identical with natural persons) exist who have legal rights and duties. He defends the analogy that these organizations then also have moral rights and duties: indeed there is such a thing as (moral) corporate accountability (Bovens, 1998, p. 53). In his view not all moral categories are applicable to organizations, but some such as responsibility as accountability or liability are (Bovens, 1998, p. 57). Accountability and liability are related but not identical concepts. Accountability can be dened as a relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgment, and the actor may face consequences (Bovens, 2006, p. 7). The forum to which accountability is directed can be of dierent characters: a parliament, a board, colleagues or mass media. Liability is a broad term including almost every type of duty, obligation, debt, responsibility, or hazard arising by way of contract, tort or statute (Encyclopedia Brittanica online). A special form of liability is strict liability. Strict liability is a legal doctrine that makes persons responsible for damages their actions or products cause, regardless of any fault on their part (http://injury-law.freeadvice.com/strictliabilty.htm). Strict liability can be described in if, then terms: if someones activities lead to damage for others (be it on purpose or not on purpose), then that person is unconditionally obliged to restore the damage. Zandvoort is an advocate of applying the principle in discussions on technological risks: Actors are unconditionally required to repair or fully compensate for any damage to others that may result from their actions, unless those who suer the damage had given their informed consent to the activities causing the damage (Zandvoort, in press, p. 3). According to Zandvoort the principle of self-determination is the most important ethical basis for strict liability. Self-determination implies the right to be safeguarded for the consequences of another ones actions (Zandvoort, 2000, p. 249). In line of these arguments I will further elaborate the concept of responsibility as accountability or (strict) liability with two concepts that were developed recently in the elds of philosophy of law and political philosophy: outcome responsibility and remedial responsibility. I will argue that these concepts are useful tools in elaborating and applying the dierent kinds of responsibility for the operation and management of critical infrastructures. 3.1. Outcome responsibility : Outcome responsibility means The concept of outcome responsibility has been developed by Honore being responsible for the good and the harm we bring about by what we do. By allocating credit for the good , 1999, p. 14). outcome of actions and discredit for bad ones, society imposes outcome responsibility (Honore The article in which he elaborates this concept starts with a provoking sentence: Being responsible in law and in ordinary life is not the same as being at fault or to blame. This view makes outcome responsibility relevant for judging the operation and management of critical infrastructures, because it is not in the rst place blaming and punishing that matter when the functioning of infrastructures is at stake. What matters is that critical introduces infrastructures work as they should do. In other words: it is the outcome that matters. Honore the concept of outcome responsibility as an argument from a moral point of view in order to apply given states that a system of outcome responcertain conditions such a severe rule as that of strict liability. Honore sibility can be defended as fair if some necessary conditions are met. It must be impartial, reciprocal and over , 1999, p. 26).3 If that is the case, outcome responsibility is the basic type of a period benecial. (Honore
3

This conditions guarantee that outcome responsibility does not lead to an absolute utilitarian interpretation.

K. van der Bruggen / Safety Science 46 (2008) 11371148

1143

responsibility in a community: more fundamental than either moral responsibility as generally understood, , 1999, p. which requires fault, or legal responsibility, which requires either fault or special danger (Honore 27). The last sentence in the paragraph where he elaborates the concept of outcome responsibility again is pro , 1999, p. 31). It does not voking: Finally, it is outcomes that in the long run make us what we are. (Honore seem strange to replace us in this sentence by critical infrastructures. The concept of outcome responsi for applications at the individual level, but applications at broader levels bility may be developed by Honore are legitimate and defensible and perhaps even more suited than at the individual level, as is shown by Miller (2004). Miller elaborates the concept of outcome responsibility at the level of nations. In line with the denition of , he describes outcome responsibility as follows: When we say that an agent is outcome responsible Honore for the consequences of her action, we are attributing those consequences to her in such a way that, other things being equal, the resulting benets and burdens should fall to her. If the consequences include harm to others, then outcome responsibility may, depending on the case, entail liability to compensate for that harm (Miller, 2004, pp. 244245).4 According to Miller outcome responsibility is a narrower notion than causal responsibility (Miller, 2004). This means that someone may be causal responsible, but not outcome responsible. This can be the case if outcomes arise in bizarre and unpredictable ways (p. 245). Miller gives this example: If I toss an orange pip over the edge of a cli and in doing so precipitate a rock fall, I am causally responsible, but not outcome responsible, for the damage that ensues. So outcome responsibility indicates that a person or an institution knows or should know what possible consequences of their actions are. Outcome responsibility is related to foreseen and unforeseen consequences of actions. Of course it is a point of discussion of what is the dierence between bizarre and unpredictable consequences on the one hand and unforeseen consequences on the other hand. Let me try to make the dierence clearer by giving some examples in the sphere of infrastructures. In The Netherlands heavy earthquakes are very rare phenomena. When building a nuclear plant the earthquake resistance of the plant does not have to be that strong as when it would be built in e.g. California. Should an earthquake or something comparable happen anyhow (e.g. because a meteor has struck the region), this could be seen as a bizarre and unpredictable situation, in which there is no outcome responsibility as dened by and Miller. But this does not mean that there is no responsibility at all. In these cases there may Honore be remedial responsibility (see below). Miller points out that outcome responsibility is less stringent than moral responsibility as that term is usually understood. To be morally responsible for the consequences of an action, you must be outcome responsible, but the converse does not hold (Miller, 2004, p. 246). Moral responsibility has to deal with the assessments of praise and blame, meeting or breaking obligations, etc. That is not necessarily the case when there is only outcome responsibility. Be it remarked here that Miller uses the concept morality (as usually understood) in a rather narrow sense by coupling it to the assessments of praise and blame. It can be defended that the concept of outcome responsibility is also a moral concept. If someone is outcome responsible for a result he or she is morally obliged to compensate the damage, even if he or she is not to blame for a fault. In relation to the operation and management of infrastructures outcome responsibility is an important concept, because the outcome (result, product, safety) is the most important criterion for judging infrastructural systems. And in that respect it often is less relevant who is, e.g. causal responsible for a malfunctioning infrastructure. Of course it has to be examined who has caused the malfunctioning, but rst of all it is important that the outcome is realized! And if this should not be possible, the option of remedial responsibility becomes relevant. 3.2. Remedial responsibility The concept of remedial responsibility has been elaborated by Miller (2001): To be remedially responsible for a bad situation means to have a special obligation to put the bad situation right (Miller, 2001, p. 454). In
4 This formulation implies that the harm can be compensated. In the case of irreversible harm, there still is outcome responsibility, but by denition the harm cannot be compensated. Dependent on the character of the damage the outcome responsible person or organization can be compelled to help, e.g. survivors or to pay a nancial debt.

1144

K. van der Bruggen / Safety Science 46 (2008) 11371148

practice this means that agents should be held remedially responsible for situations when, and to the extent that, they were responsible for bringing those situations about (Miller, 2001, p. 455). At rst sight this concept seems closely related to that of outcome responsibility. But according to Miller remedial responsibility does not always follow immediately from outcome responsibility. This is most obviously in the case where that agent cannot now remedy the harm in question (Miller, 2004, p. 266).5 This is the case if the government of a country has because of bad policy contributed to the poverty of its inhabitants, but does not have the possibility to prevent a famine. In such a situation the international community has a remedial responsibility to help these people, although there is no outcome responsibility. Also in the case of natural disasters as an earthquake or a tsunami no one is to be blamed (except perhaps for not building earthquake resistant houses). But there will not be many people who will say that in these cases there is no obligation to help victims and to restore damage. This moral obligation does not coincide with moral responsibility that is directly linked to accountability and liability. So, remedial responsibility entails more than restoring the damage that is caused under the umbrella of outcome responsibility. Because of that remedial responsibility can be seen as a variant of responsibility as a task or obligation. What is the practical relevance is of the principle of remedial responsibility? Applied to infrastructures remedial responsibility means that non-functioning or malfunctioning infrastructures need to be restored or replaced as soon as possible, regardless of who caused the damage. This is also the case when because of bizarre and unpredictable consequences a failure of the critical infrastructure has occurred. So, even after an earthquake in the Netherlands there is the remedial responsibility to restore the damage and to help the people who are dependent of this infrastructure. It must be realized that this is a very demanding interpretation of responsibility, which perhaps not always can be realized. For instance, how realistic is it to ask the public authorities or the energy company for restoring the energy supply, if in times of war all power plants of a country have been damaged or destroyed. It should be recognized that there are diculties in applying the concept of remedial responsibility in practice. But applying the concept to the management and operation of infrastructures shows the importance of infrastructures for society. The next question is who should be the bearers of this responsibility. 4. Responsibility for the management and operation of infrastructures Until now attention has been paid to the meaning of the concept of responsibility, and more specically outcome and remedial responsibility, in the context of infrastructural systems. But who is or should be outcome and/or remedial responsible for the functioning of critical infrastructures? How far does the responsibility of public authorities as politicians and civil servants go? And what does it mean for the responsibility of other actors, such as private companies, because these undoubtedly have their own (part of) responsibility? And last but not least: what do outcome and remedial responsibility mean for individual actors? 4.1. Responsibility of public authorities As infrastructural systems are so important maybe even constitutive for a society it is almost evident that in last resort public authorities are responsible for infrastructural projects and activities that take place within their jurisdiction. A generally shared conclusion in most societies indeed is that the good functioning of critical infrastructures is a public responsibility. Politicians and broader the public sector can and must be held responsible for the operation and management of infrastructures. More specic this public responsibility for infrastructural projects is at stake in dierent kinds of arrangements.

Miller gives this example: . . . showing that there are instances in which nations that lack sucient resources to keep all of their members above a poverty line are themselves responsible for that lack, by virtue of policies pursued in the recent past, or features of the national culture that have prevented sucient resources from being generated, does not entail that other nations should not now step in (Miller, 2004, p. 266). If remedial responsibility is accepted in these cases it will always be a point of discussion how far this remedial responsibility goes. What may and should be asked of people with remedial responsibility. Should St. Martin of Tours have given the beggar only half of his mantle?

K. van der Bruggen / Safety Science 46 (2008) 11371148

1145

 The government (or a local/regional public authority) is partially or totally the owner of the infrastructural system. Until the 1980s the common situation in most Western European countries was that public authorities (local, regional and/or national) were the owner of almost all infrastructural systems. At the end of the 19th and in the beginning of the 20th century most infrastructural projects were initiated or taken over by local, regional or national authorities. To give some examples from the Netherlands: regional organizations (most coinciding with provinces) for electricity supply were founded. And the dierent private railways companies came together in one state owned national railways company. This development was strengthened after the Second World War, when all over Western Europe the concept of the welfare state (or the Rhineland model) was adopted. This model is based on a rather strong government intervention in economic aairs. In the Netherlands this welfare state model led to the above mentioned polder model. In this model great governmental inuence on critical infrastructures was the generally accepted policy line. Of course there were specialized organizations that took care of the daily operation and management tasks, but these were owned and ruled by public authorities. The people who worked for these organizations were civil servants. And although politicians were not involved directly or often even indirectly in the management and operation of critical infrastructures, they were and should be approachable and responsible for these matters. If something went wrong, politicians were accountable for it.  The government has made explicit legal, nancial or policy arrangements with the owner of the infrastructural system. In the past 1520 years the Rhineland model lost ground in favour of the so-called Anglo-Saxon model, which favours only limited government intervention (Bolkestein, 1998). A consequence was the liberalization of most infrastructural systems. It was no longer thought necessary that the local, regional or national government should own and control the performance of the infrastructural system directly. Private companies bought the rights and/or the ownership of infrastructural services and systems. Also old players showed up in a new appearance of privatized or semi-public state companies. But this new situation does not imply that politics should no longer bear any responsibility for the management and operation of infrastructural systems. The state and other public institutions still are or should be important actors in a direct and an indirect way. Instead of the formal ownership of infrastructural systems by public authorities new relations have grown based on legal, nancial and policy arrangements with the (new) owners of the infrastructural system. These arrangements are meant to guarantee the good functioning of the infrastructural systems. On the one hand the arrangements describe the obligations of the owners to maintain the operation and management of their infrastructural systems. On the other hand also the obligations and tasks of the public authorities such as nancial and legal support are described. These public tasks give the governmental institutions a direct responsibility in those situations where they are no longer the formal owner of the infrastructure. An example from the Netherlands is TenneT. TenneT is the transmission system operator (TSO) that manages the Dutch transmission grid. We do this by making our transmission grid available impartially for electricity transmission and by ensuring the necessary balance between supply and demand in the Netherlands. Seen that the transport of electricity doesnt stop at our borders, TenneT cooperates with European TSOs. Since October 2001, the State has been the sole shareholder in TenneT (http://www.tennet.nl/wie). Also in other important infrastructural sectors government often have a majority or even all of the shares. This guarantees government the possibility to inuence the policy of the infrastructural company, for instance and if the arrangements that are made with the owners do not have the aimed results. In the words of Honore Miller: government or broader, the public sector accepts its outcome responsibility for infrastructural systems.  The government has obligations to the people who are dependent on the infrastructural system. Even if a public authority is no owner or does not have any formal arrangements with owners of infrastructural facilities, politicians still have an outcome responsibility for the adequate operation and management of the infrastructures, if only because infrastructural provisions are essential for the good functioning of society.

1146

K. van der Bruggen / Safety Science 46 (2008) 11371148

The legitimacy of a public authority that is not able to guarantee these provisions is at stake, even if other public tasks are performed well (although that will not be the case very often). In such a situation the capacity to take responsibility for infrastructural systems has to be restored as soon as possible. If the public authority is a city council or a regional authority, higher authorities (regional, national, perhaps even international) have to take their responsibility by assuming power until the lower authority is capable again to take the responsibility. Things are even more serious of course if a national government is not able to guarantee good working infrastructures. If this is only temporarily and can be explained because of extraordinary circumstances, e.g. a ood or an earthquake, the situation can be restored with international support. But if the failing has a more structural character, more drastic measures may be necessary. One solution can be that internally another (stronger) government is chosen or appointed. From a legal point of view this in fact is the only option. Another more controversial option is an intervention in the failing state by foreign countries. This intervention can be economical, political and in last resort military. In political, ethical and legal circles a lot of debates were held in the past few years about the question under which circumstances a (humanitarian) military intervention can be justied. Until now there is no international consensus. But if there is a kind of agreement growing, then it concentrates on the opinion that intervention could be allowed only if the human rights are severely violated (e.g. by genocide). The malfunctioning of infrastructures is of course not comparable with such a terrible event. A last remark: most of the time it is far more easy to say that politicians have or should have outcome responsibility than to say what that responsibility entails in practice. Agreement exists about one very specic task as a responsibility for public authorities: inspecting and evaluating the way private or semi-public infrastructural organizations are performing their management and operational tasks. Most countries have created a structure of evaluating, controlling and inspecting infrastructural organizations. It is an institutionalized way to perform a duty that is based on the concept of outcome responsibility. 4.2. Responsibility of infrastructural companies Another important group of actors on institutional level that bear responsibility for infrastructures are the organizations and companies that manage and govern the daily operations of these systems. These can be publicprivate cooperation institutions, prot and non-prot organizations. The outcome and remedial responsibility of involved organizations is more direct and more specic than that of public authorities. The responsibility is immediately related to the task of managing and operating infrastructures in daily practice. In fact this means no more and no less than assuring and guaranteeing that the critical infrastructure does what it should do: delivering electricity, delivering clean water, guaranteeing telecommunication-trac, driving trains and buses according to appointed time schedules, etc. And this should not only be guaranteed when the sun is shining, but also in bad weather (both literally and guratively). Related to this task is the obligation to prevent failures from happening. And if accidents should happen, the damage needs to be restored as soon as possible. Outcome responsibility includes remedial responsibility: restoring and correcting failures of the system as soon as possible as well as to have emergency measures prepared so that services can temporarily be taken over, for instance generators for electricity in essential places (like hospitals) or buses that replace trains. This remedial responsibility also exists if the infrastructural company does not have any accountability for an accident or incident that has caused the failure of the system. In most cases the infrastructural organizations have to perform these tasks themselves. It is not seen as an executive task of the political authorities. An exception is the employment of police or military forces in the case of a security threat. But this happens only if there is an acute threat that cannot be countered by internal security measures of the organization. In most institutions and companies executive tasks are allocated or delegated to lower levels in the organization. In this paper no attention will be paid to these dierent organizational, legal and nancial arrangements that exist on this level of infrastructural institutions. All these dierent arrangements are ways to take care of outcome and remedial responsibility. The main point of the argument is that this outcome and remedial responsibilities exist, irrespective of the way how tasks and responsibilities are shared and/or distributed within an organization. In practice this can lead to a situation in which it is not clear who is responsible

K. van der Bruggen / Safety Science 46 (2008) 11371148

1147

for a failure or that more people or groups are co-responsible. This is the earlier mentioned problem of the many hands. But this many hands problem can never be used as an excuse for the malfunctioning of infrastructures. Of course that does not mean that the many hands problem is not relevant. Every organization should try to prevent, solve or reduce that problem. 4.3. Responsibility of individual persons Often the problem of many hands is literally linked to real hands, or better people of esh and blood! The third layer in the ladder of responsibility for critical infrastructures is the responsibility of individual persons, be it that most of these persons are part of either public authorities or infrastructural organizations. Independent from views on the (im)possibility of collective or shared responsibility, fact is that in a public authority or in an infrastructural organization individual persons do have their own responsibility as a politician, a civil servant, a designer, an engineer, a manager, a user but also as a consumer. The kinds of individual responsibility dier according to the function or role that individual persons are performing. An engineer does have another responsibility than a manager, a driver or a politician. Again, an example can make this clear. Suppose there is a derailment of a train. Who on a personal level can be involved? At least the following persons and roles can be identied: the train engine driver, the engineers who are responsible for the construction of the train, the engineers who are responsible for the maintenance of the train, the engineers who developed the rail track, the people who are responsible for the maintenance of the rail track, the trac controller, the director/manager of the railway company, the director/manager of the rail track company, the civil servants of the Railways Department, the Minister who has railways in portfolio. But although it can be stated that all of these persons possibly have a personal responsibility, this does not yet mean that they all have the same responsibility. Some are liable and because of that responsible because of their action or omission that caused the accident; others perhaps have an administrative or organizational accountability: they have organized the train planning or the personal duties. And in last resort the Minister has outcome responsibility or perhaps better in this case political responsibility. In some cases there is at least at rst sight no doubt about the individual responsibility for accidents. If someone as has happened in 2005 in the Netherlands throws a heavy lorry on the rail track, this person and nobody else is responsible for the accident that follows. But even in these clear cases there is room for a second thought. How could this happen? Why did this person have access to the rail track? Could or should nobody have stopped him? These questions certainly arise if these kinds of incidents take place more than once and thus cannot be qualied as bizarre and unpredictable. In that case there is an outcome responsibility for the authorities, and this again leads to executive responsibilities at the individual level. So, whatever responsibility public authorities and infrastructural companies have in managing and operating infrastructures, in last resort individual responsibilities are extremely important. Because of that it is a primary task of infrastructural companies and public authorities to carry out a policy that leads to putting the right people on the right place. And the right people themselves have the obligation to perform their duties optimally. Arrangements have to be developed to evaluate and control the performance of individual technicians, sta members, etc., and to punish or even replace them if they fail. This should lead to a chain of responsibilities that can be dierent in various infrastructural systems (such as water management and telecommunications systems). 5. Conclusions In this article it has been argued that critical infrastructures are of great importance for each modern society. A good working infrastructural system can even be viewed as a conditio sine qua non for modern states and societies. Because of that public authorities at the national, regional and local level do have a responsibility for the operation and management of infrastructures, also if the management and/or ownership of infrastructures are in private hands. The divergent multitude of dierent individual and corporate responsibilities in the operation and management of infrastructures make it dicult to decide who is responsible if accidents or incidents happen. Practice

1148

K. van der Bruggen / Safety Science 46 (2008) 11371148

shows that involved parties try to shift the responsibility to other organizations, as happened frequently in the Netherlands in the railway sector. Politicians blamed the railway company and vice versa. The same patron was repeated during a great breakdown of electricity supply in November 2005. In this article the position is defended that in last resort public authorities have the outcome and remedial responsibility for the infrastructural systems in their country, because these systems are vital for the functioning of society. This responsibility does in practice extend to critical infrastructures. If something goes wrong politicians almost always are asked for their responsibility as well for the outcome as for the remedial activities. Politicians cannot escape by telling that what has happened is not their responsibility but, e.g. that of the company that runs the infrastructural system. Of course this company has its own responsibility, but that does not eliminate the responsibility of politicians. By introducing the concepts of outcome responsibility and remedial responsibility the relevance of the direct (causal and perhaps moral) responsibility for incidents can become less important. It is the outcome and remedial responsibility for infrastructures that matters! Arguments and considerations should be directed at pointing out what political and institutional arrangements guarantee the best possible realization of outcome and remedial responsibility and because of that an optimal operation and management of infrastructures. This analysis can be applied in concrete infrastructural systems to decide what political and institutional arrangements will lead to a better, more reliable and more safe outcome for society.6 References
Bolkestein, F., 1998. The Model of Consent The Dutch Experience, Lecture of the President of the Liberal International. Helsinki, 1998. Bovens, Mark, 1998. The Quest for Responsibility. Accountability and Citizenship in Complex Organisations. Cambridge University Press, Cambridge. Bovens, Mark, 2006. Analysing and Assessing Public Accountability. A Conceptual Framework, European Governance Papers (EUROGOV) No. C-06-01 (January 16, 2006). <http://www.connex-network.org/eurogov/pdf/egp-connex-C-06-01.pdf>. De Bruijne, M.L.C., 2006. Networked Reliability. Institutional Fragmentation and the Reliability of Service Provision in Critical Infrastructures. Febodruk BV, Enschede. Encyclopedia Brittanica. <http://www.britannica.com/search?query=liability&submit=Find&source=MWBOX>. Hart, H.L.A., 1968. Punishment and Responsibility. Essays in the Philosophy of Law. Clarendon Press, Oxford. bertragungsHermes, Georg, 1998. Staatliche Infrastrukturverantwortung. Rechtliche Grundstrukturen netzgebundener Transport und U systeme zwischen Daseinsvorsorge und Wettbewerbsregulierung am Beispiel der leitungsgebundenen Energieversorgung in Europa. Mohr Siebeck, Tu bingen. , Tony, 1999. Responsibility and luck. In: Honore , Tony (Ed.), Responsibility and Fault. Hart, Oxford, pp. 1440. Honore Miller, David, 2001. Distributing responsibilities. The Journal of Political Philosophy 9 (4), 453471. Miller, David, 2004. Holding nations responsible. Ethics 114 (2), 240268. Mote, John, Copeland, Claudia, Fischer, John, 2003. Critical Infrastructures: What Makes an Infrastructure Critical? Report to Congress, Washington DC. Rathenau Instituut, 1994. Stroomloos Kwetsbaarheid van de samenleving; gevolgen van verstoringen van de elektriciteitsvoorziening. Rathenau Instituut, Rathenau Institute. Schneider, Volker, Ja ger, Alexander, 2003. The privatisation of infrastructures in the theory of the state: an empirical overview and a discussion of competing theoretical explanation. In: Wubben, Emiel M., Hulsink, Willem (Eds.), On Creating Competition and Strategic Restructuring, Regulatory Reform in Public Utilities. Edward Alger, Cheltenham, UK/Northampton, MA, USA, pp. 101 137. <http://www2.hu-berlin.de/CompPol/Mitarbeiter/JAEGER%20Veroeentlichungen/schneiderjaeger2003.pdf>. Here the website version has been used. Smith, Adam 1812 [1776]. Wealth of Nations. Ward, Lock & Co., London. Thompson, Dennis, 1980. Moral responsibility of public ocials: the problem of many hands. American Political Science Review 74, 905 916. Website European Union. <http://www.eu.int/ecsc/results/indexen.htm>: information about history ECSC. Website Journal for Critical Infrastructures. <http://www.inderscience.com/browse/index.php?journalCODE=ijcis>. Website TenneT. <http://www.tennet.nl/wie>. Zandvoort, H., 2000. Controlling technology through law: the role of liability. In: Brandt, D., Ernetic, J. (Eds.), Preprints of Seventh IFAC Symposium on Automated Systems based on Human Skill, Joint Design of Technology and Organisation, June 2000, Aachen, Germany. Duesseldorf, pp. 247250. Zandvoort, H., in press. Risk Zoning and Risk Decision Making. International Journal of Risk Assessment and Management.

For example: does public or private ownership of the railways, or splitting or not splitting up energy companies lead to optimal results?