Logic of Liability

October 23rd, 2006

Views: 663

Posted by ChrisG at 2:28 pm

I reproduce below a section of something I’m writing on the conceptual problems involved in thinking about moral and legal responsibility in a way adequate to the contemporary world – in which decision-making processes, often conducted under massive pressure from economic and political actors with an interest in short-term benefits, have consequences that are reaching further and further into the future.

Responsibility of course tends to be seen in law as something established after the fact – and so Minority Report gained its impact against the background of this standard context, as an attempt to retain assumptions about how legal accountability works that runs on a reversed temporal logic. But there are other, much richer understandings of responsibility that inform ethics, and which can perhaps be mobilised to change the way we think about the relationship between action now and future consequences. The next extract I post will look at some of these alternatives.

Ok, avanti…

Liability and Reciprocity
Liability for direct harm is the most familiar form under which responsibility can be legally attributed. To legally assign responsibility, it is necessary that a direct causal role in harming can be imputed to an agent. For example, responsibility might be assigned on the basis of a direct causal connection between someone’s actions and specific consequences, or alternatively, on whether or not they successfully perform their publicly designated function in a particular social context. If you see me assault someone in the street with whom I am arguing, you are justified in holding me responsible for their injuries. Similarly, you would have been justified in holding a nearby police officer responsible for preventing the assault had one been present whilst the argument was escalating towards violence (Pellizzoni 2004, p. 547). The causal role of the agent here is direct, either in the sense that they are seen as the source of the harm, or as directly contributing to it by omission of action.

Alternatively, agents can be held responsible for certain outcomes of their actions further down the line, if it can be shown that they should have foreseen the risks of acting as they did, but did nothing to ensure public safety (Honoré 1999). For example, a manufacturer may be held legally liable for harm caused by a product once they can be shown not to have taken into account the risks posed by the harmful chemicals it contains. Liability for harm, like direct causal responsibility, can be imputed to both individuals and organisations. In either case, for a legal judgement about responsibility to be possible presupposes that, firstly, the agent actually has a legal responsibility to behave in certain ways, and secondly, that the causal link between the agent and the harmful act can be established. So there must be something about agents that means they can be held legally responsible for their actions. There must also be in operation some standard of proof that allows for the possibility of a truthful connection between agent and harm to be established.

These assumptions underlie what could be called the ‘thinnest’ concept of responsibility in law, that of legal liability. If liability can be attributed for an act, then it requires only that the act be caused by someone with the capacity to be a moral agent, and that the assumption that the fact that harm has been done is morally significant, i.e. is harm done to an entity who has rights. A right is an entitlement which, it is considered, must be fulfilled in order to secure and protect the intrinsic value of a moral agent. Rights can only be exercised by or on behalf of moral agents, and recognised by other moral agents. Whatever the social and institutional relationships between individuals, be these professional, political, familial, etc. these entitlements mark the limits beyond which actions taken by other individuals cannot cross without harm being committed and liability incurred. These entitlements, in classical social contract theory, are held to be the basis of natural law, which establishes a condition of equality between moral agents that obtains independently of socio-legal arrangements, and to which they should conform. In Locke’s formulation:

And, being furnished with like faculties, sharing all in one community of Nature, there cannot be supposed any such subordination among us that may authorise us to destroy one another, as if we were made for one another’s uses, as the inferior ranks of creatures are for ours.

(Locke 1988, 2nd Treatise, ch. 2, §6)

Given that the distribution of these entitlements is wholly equal, then their existence imposes an equal and reciprocal responsibility on every person to respect the rights of all other individuals. I acknowledge my responsibility to protect your rights because you reciprocally acknowledge your responsibility for mine. This concept of the universally binding reciprocal recognition of natural rights forms the basis for the legal equality of individuals, and also establishes what is involved in the breaking of the contract when harm is done to another individual, namely, an infringement of the rights of that individual and a transgression against their intrinsic worth as a moral agent. For an attribution of liability to be justified therefore requires that all individual agents be morally responsible for respecting the rights of all others.

To establish this relationship between fundamental legal responsibility and liability requires that we select and adopt a specific perspective on what is morally significant about human beings. To reach these conclusions we must abstract away from individuals’ particular concerns and attachments with the world and others, and view the situation from a ‘higher’ position above these entanglements, one in which the most morally fundamental aspect of being human is revealed. This higher position is one in which our basic, immutable interests, become known through rational reflection on the meaning of the concept of individual agency. These interests derive from the special dignity ascribed to the uniquely human ability to decree for oneself a rule of conduct and follow it out of free will (Kant 1993, pp. 40-1).

The justification of and motivation for reciprocal responsibility is therefore cognitive: it stems solely from rationally-derived knowledge of certain features of being human.

Accountability and Knowledge of Causal Relationships
To act with regard for the rights of another means being held responsible for giving a true account of one’s actions. A legal tribunal is a formal setting in which a true account of a series of events is sought according to a set of norms and standards that meet with wider social agreement, before a ruling on the case can be made. The existence of responsibility is not therefore dependent, in the first place, on some ‘voice of conscience’, but upon an institutional framework that holds agents accountable. We are held responsible by another subject, or rather by a general plaintiff such as the Crown in the UK or State or Federal Government in the USA, to whom we must give an account of ourselves. In societies where the legal order receives its final legitimation from monotheistic religious belief, the legal tribunal is mirrored in the sphere of religion by the practice of confession, which makes accountability to another, and to God as the ultimate Other, the foundation of all ethical relationships (Davis 2001, p. 3). In societies that explicitly or implicitly do without such a religious foundation, and instead operate with a concept of human dignity as the ethical basis of the legal order, one is accountable for one’s actions to the ideal of human dignity as such. The tribunal then represents the social contract as such, demanding an account of why the contract of reciprocal recognition was broken.

Before this tribunal, evidence and reasoning about cause and effect are presented within a specific epistemological framework. This is generally understood to be based on a common-sense understanding of causality, buttressed if necessary with scientific evidence from expert witnesses, with this evidence being generally treated ‘honorifically’ as a more secure extension of common-sense means of establishing causal relationships.¹ The form of rationality that frames evidence is generally taken to be that which conforms to a scientific schema, as opposed to, say, one that admits non-naturalistic causes. Within this framework, for an account of an act to show that it does not necessarily incur liability for harm, it must demonstrate that, on the basis of relevant knowledge of the causal context around the act, it would not have been possible to predict that the harm could have been caused.

Distortions produced by temporal presuppositions
Together, these sets of assumptions, which together yield what I shall refer to as the R-L (for ‘reciprocity-liability’) model of responsibility, create serious problems for our understanding of responsibility within the new institutional milieu I described at the outset. This is because neither is capable of coping with the significance of the new temporal context of action created by this changed milieu. On the one hand, this limitation has to do with the model of human subjectivity that is operative within the conceptual framework of reciprocal rights and classical social contract theory, and which provides us with a particular understanding of subjectivity as the basis of action and knowledge. On the other, it concerns a related problem, which has to do with the epistemological limitations of scientific reasoning.

To take the latter issue first, consider this scenario:

(A) A company develops a new and highly effective chemical method of pest control which becomes the market leader. Early tests on the product indicated that it was safe when used in typical agricultural settings. But thirty years later, it is discovered that residues of the chemical have been building up in the bodies of animals and humans across the areas in which it is being used, and that the occurrence of certain birth defects and degenerative diseases has increased in the period since its widespread adoption.

Here we have an example of a problem which has proven disturbingly common in relation to advanced chemical, genetic, bio-, nano- and nuclear technologies. To use the products of such technologies is in effect to introduce them into social and ecological relationships in which unpredictable and sometimes uncontainable contamination of the surrounding environment can occur. Crucially, these unintended consequences can remain latent for generations, and are often multi-causal (Adam 1998, pp. 165-166; Colborn et al. 1996; Koppe 2001).

In cases like (A), the standard approach to responsibility in terms of liability is seriously deficient. This is because of an inevitable gap in scientific knowledge about the effects of technology once it is out of the laboratory and being assimilated into the social and natural world. Legal liability for the effects of actions can typically be avoided if it can be shown that all possible measures for avoiding harm, based on then–current knowledge of the risks of action, had been taken. This means that if certain risks are not predicted at the time of acting, due to the state of scientific knowledge at the time, then the agent responsible for the act cannot be held responsible for the unforeseen effects (Pellizzoni 2004, p. 552).

This demonstrates the presence of a basic flaw in the idea of liability. The idea of liability certainly makes sense in face-to-face, everyday contexts where responsibility is to be established. The problem is the change of ethical context that comes with the use of technologies that inherently have an extended temporal reach, and which have the potential to alter basic ecological processes or natural structures. Given this temporal reach and potential for greater-than-envisaged contamination effects, the difference between laboratory testing or controlled trials and the release of a technology ‘into the wild’ is so large as to mean that the problem of uncertainty is not an adventitious one that only afflicts certain uses of technology. Rather, it becomes a structural feature of modern technology itself, once ‘technology’ means chemical, genetic, nanoscale or nuclear engineering.

The idea of accountability, as we have seen, must assume some shared concept of truth in order to establish the causal link between act and consequence. But when this concept of truth is centred on the scientific ideal of linear causal explanation, then it necessarily operates with a causal context that is based on knowledge of the past. The problem is that predicting the effects of powerful new technologies on the basis of what is known about other technologies is often like generals’ predilection for fighting the last war, instead of the one they are fighting (Colborn et al. 1996, pp. 241-2). The problem is that futures are being constructed by technologies of which the past knows nothing.

The temporal reach of technology brings another problem. Because of the causal complexity of the ecological context in which it operates, and the potentially huge time lapse between its introduction and the emergence of symptoms, the precise ‘story’ of how these symptoms came about is difficult to establish: in (A), is the chemical the sole cause? Or is the problem caused by (say) the later introduction of phosphate-based fertilisers or some other technology, which then combined dangerously with the pesticide? Given the temporal distances involved, causal linkages, and with them, ‘ownership’ of consequences can thus be very hard to establish.

We can say therefore that the R-L model of responsibility runs into trouble because of a mismatch between its basic temporal orientation, and the temporal reach of some of the social practices which it oversees. Fundamentally, the attribution of liability has to rely on the provision of a causal story that establishes, after the fact, authorship of an act, and the contributing role of recklessness if not deliberate intent. But the technological potential for creating latent effects, and the inability of predictions based on the past to deal with radically new situations, makes such a model of accountability inadequate for the contemporary context.

Let us now consider the assumptions about human agency with which the R-L model operates. As noted above, the definition of humans as moral agents upon which the standard concept of responsibility rests is individualist and cognitivist, and as such is comparable with Kant’s understanding of moral autonomy. What is basic to this position is that, in order to be capable of being considered a moral agent, the individual has to be considered as being capable of cognitively intending to do otherwise than she actually does in any given situation (Chisholm 1966, pp. 30-44). When harm is done to a moral agent, then whoever is held accountable is seen as responsible because she could have avoided doing harm, either through not making up her mind to deliberately do it, or by not choosing to act recklessly. It is this moral agency that is seen as constituting the agent as worthy of a special respect, and therefore as conferring rights and reciprocal duties. This unique moral status may be seen as necessarily implying a special metaphysical status for moral agents as opposed to other entities (Ingarden 1970, pp. 18-19). The capacity to make a single judgement here and now as to the rightness or wrongness of an action, together with the power to then act accordingly is the basis of moral respect.

This creates two problems:

(1) this concept of action has similar weaknesses to those that afflict the concept of accountability. To hold that responsibility can only be established on the basis of the power to choose to do otherwise is widely taken to imply that the individual can only beheld responsible for acts that were within the individual’s power at the time of acting (Chisholm 1967, pp. 415-16). If the risks of a course of action, such as producing the pesticide in (A), are unknowable due to inherent epistemological limitations, then the idea that the avoidance of such risks is ever within the power of an agent is simply inapplicable. The idea that responsibility can only be linked to an instantaneous decision to either do X or not to do it on the basis of what is ‘in the agent’s power’ cannot effectively be applied in the new context that we are concerned with here.

(2) To avoid harming a moral agent is justified on the basis of their possession or exercise of certain rights, i.e. justified entitlements claimed or exercised by someone capable of moral agency, or by someone else appointed by them for this purpose. It is on this basis that rights are interpreted as part of the terms of a social contract, and thus as the object of a reciprocal obligation. However, members of future generations do not yet exist, and can therefore neither claim anything for themselves, nor appoint someone else to exercise or defend their rights for them (MacLean 1983, pp. 183-4; Steiner 1983, p. 154). Nor does a logical relationship of reciprocity obtain between us and future generations. We can harm them, but they cannot, at least on the basis of the R-L model, harm us. By the same token, it is impossible for any who suffer harm but are not members of the ‘community of moral agents’, such as animals, plants and other natural elements of our wider ecology, to be taken into account once a social contract of reciprocity has been made the ultimate foundation of the legal order. Once the potentially disastrous consequences of present actions reach out beyond this ‘community’, this becomes more than just a problem with the coherence of a definition. The moral status of future generations itself, and thus the nature of our ethical relationship to them is in question, as well as our potential capacity for exploiting and polluting nature without limit.

As we have seen, the R-L theory of responsibility is suited for certain contexts of action, defined by specific relationships between agents and patients, which must be taken as including a temporal dimension. However, when the institutional element of these contexts changes due to the evolution of technologies that fundamentally alter the ‘timescape’ of action (Adam 1998, pp. 53-6), then this theory of responsibility effectively becomes an incentive for irresponsibility. The use of a rights-based model as the basis for policy practices such as future-discounting, where future persons are defined as being of inherently lower moral status than those alive now, is a good example of how this occurs.² If liability, the attribution of agency, and the protection of rights cannot function adequately in contexts where the temporal and ecological reach of action has expanded, then we need to interpret responsibility anew so as to avoid relying on the excessive thinness of the R-L model.

Notes:
¹ Haack 2005, p. 67. This is, however, arguably no longer the case in the USA, where ever since the 1993 Daubert vs Dow Pharmaceuticals ruling, it has been possible for a judge to determine what is ‘sound science’ and what is ‘junk science’. Debate continues over the significance of this ruling for the judicial understanding of what standards of rationality are to be applied in court. See the papers collected in the 2005 Supplement to vol. 95 of the American Journal of Public Health, and especially Lakoff 2005.

²A growing literature addresses the morality of future-discounting and cost-benefit analyses in general. See for example (Jacobs 1991, pp. 68-70, 81-82; O’Neill 1993, pp. 52-3; Parfit 1983a, pp. 31, 37)

References:
Adam, Barbara. Timescapes of Modernity: The Environment and Invisible Hazards. London: Routledge, 1998.
Chisholm, Roderick. “Freedom and Action.” In Freedom and Determinism, edited by Keith Lehrer, 11-44. New York: Random House, 1966.
———. “He Could Have Done Otherwise.” Journal of Philosophy 64 (1967): 409-17.
Colborn, Theo, Diane Dumanoski, and John Peterson Myers. Our Stolen Future: Abacus, 1996.
Davis, Winston. “Introduction: The Dimensions and Dilemmas of a Modern Virtue.” In Taking Responsibility: Comparative Perspectives, edited by Winston Davis, 1-27. Charlottesville and London: University Press of Virginia, 2001.
Haack, Susan. “Trial and Error: The Supreme Court’s Philosophy of Science.” American Journal Of Public Health 95, no. Supplement 1 (2005): 66-73.
Honoré, Tony. Responsibility and Fault. Oxford: Hart Publishing, 1999.
Ingarden, Roman. Ãœber Die Verantwortung. Stuttgart: Reclam, 1970.
Jacobs, Michael. The Green Economy. London: Pluto Press, 1991.
Kant, Immanuel. Groundwork for the Metaphysic of Morals. Translated by James W. Ellington. Indianapolis: Hackett, 1993.
Koppe, Jana G., and Jane Keys. “Pcbs and the Precautionary Principle.” In Late Lessons from Early Warnings: The Precautionary Principle 1896-2000, edited by Poul Harremöes, David Gee and Malcolm MacGarvin, 64-75. Copenhagen: European Environment Agency, 2001.
Lakoff, George P. “A Cognitive Scientist Looks at Daubert.” American Journal Of Public Health 95, no. Supplement 1 (2005): 114-20.
Locke, John. Two Treatises of Government. Cambridge: Cambridge University Press, 1988.
MacLean, Douglas. “A Moral Requirement for Energy Policies.” In Energy and the Future, edited by Douglas MacLean and Peter G. Brown, 180-96. Totowa, NJ: Rowman and Littlefield, 1983.
O’Neill, John. Ecology, Policy and Politics. London; New York: Routledge, 1993.
Parfit, Derek. “Energy Policy and the Further Future: The Social Discount Rate.” In Energy and the Future, edited by Douglas MacLean and Peter G. Brown, 31-37. Totowa, NJ: Rowman and Littlefield, 1983a.
Pellizzoni, Luigi. “Responsibility and Environmental Governance.” Environmental Politics 13, no. 3 (2004): 541-65.
Steiner, Hillel. “The Rights of Future Generations.” In Energy and the Future, edited by Douglas MacLean and Peter G. Brown, 151-65. Totowa, NJ: Rowman and Littlefield, 1983.