UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Collective responsibility and democratic practice Tian, Jie (TJ) 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2014_spring_tian_jie.pdf [ 771.19kB ]
JSON: 24-1.0167412.json
JSON-LD: 24-1.0167412-ld.json
RDF/XML (Pretty): 24-1.0167412-rdf.xml
RDF/JSON: 24-1.0167412-rdf.json
Turtle: 24-1.0167412-turtle.txt
N-Triples: 24-1.0167412-rdf-ntriples.txt
Original Record: 24-1.0167412-source.json
Full Text

Full Text

 Collective Responsibility and Democratic Practice   by   Jie (TJ) Tian        M.A., Tsinghua University (China), 2005 B.A., Beijing University (China), 2002     A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate and Postdoctoral Studies  (Philosophy)  The University of British Columbia (Vancouver)  April, 2014    © Jie (TJ) Tian, 2014ii   Abstract Questions of responsibility lie at the very heart of most social and political controversies. Responsibility attribution becomes rather complicated when it comes to social activities, in which members of a group share agency in collective actions or practices, even though these members may not participate directly in the outcome of such actions and the contribution of any single member to that outcome is negligible. The general tendency to individualize and to psychologize human actions makes it difficult to understand how collectives, such as companies, families and nations, can or should take responsibility for their actions. Philosophical attempts have been made to make sense of the concept of collective responsibility. Following these philosophers, I intend to develop some guiding principles for investigating collective responsibility and to explain how these principles reshape our general understanding of responsibility and public life.   These principles of understanding collective responsibility have the potential to affect the practice of collective decisions making. Political decision-making by citizen participants in a democratic process is taken as a paradigm case of collective actions. I will defend a deliberative model of democratic practice as the most promising one in assisting citizen participants in recognizing that they will be invested with shared responsibility of the political community. A deliberative model not only helps citizen participants gain deliberative capacity and become more responsive to reasons, it also helps participants to see the shared nature of collective decisions.   iii  Preface Jie (TJ) Tian is the sole author of this thesis. She was responsible for the design of the research program, the research itself and the composition of the thesis. This thesis is original, unpublished work.   iv   Table of Contents  Abstract ii Preface iii Table of Contents iv List of Tables vii Chapter 1: Introduction 1 Chapter 2: Making Sense of Collective Action 9 2.1 Some general ideas related to reductive individualism ...................................9 2.2 Agent-collectivism ........................................................................................12 2.2.1 Peter French on collective action .........................................................13 2.2.2 Gilbert on collective action ..................................................................14 2.2.3 Remarks and criticism on agent-collectivism ......................................16 2.3 Intention-collectivism ...................................................................................21 2.3.1 Bratman on shared intention ................................................................21 2.3.2 Bratman’s shared intention in contrast to Gilbert’s joint commitment23 2.3.3 Criticism on Bratman’s theory of collective action .............................26 2.4 Participation-collectivism .............................................................................28 2.4.1 Kutz on group participation .................................................................29 2.4.2 Remarks and criticism..........................................................................31 2.5 Collectivity in group action ..........................................................................33 2.5.1 A minimal threshold account of collective action ...............................34 2.5.2 Effectiveness of collective action ........................................................35 v  Chapter 3: The Inadequacy of Individualistic Approach to Collective Responsibility 37 3.1 Soft individualism .........................................................................................37 3.1.1 The individualistic doctrines of responsibility .....................................39 3.1.2 Translation of collective responsibility ................................................43 3.1.3 Three options to make sense of collective responsibility by soft individualis ...........................................................................................50 3.2 Cases of imperfectly translatable collective responsibility ...........................54 3.2.1 Over-determination ..............................................................................55 3.2.2 Collective inaction ...............................................................................57 3.2.3 Discursive dilemma .............................................................................60 Chapter 4: An Alternative View of Collective Responsibility 65 4.1 Revisiting collective action and inadequate individualism ...........................65 4.2 Plural understanding of collective action and responsibility ........................72 4.2.1 Context-sensitivity ...............................................................................73 4.2.2 Respondent-dependence ......................................................................74 4.3 The function of collective responsibility ......................................................75 4.4 Being responsible and holding responsible...................................................80 4.5 Collective responsibility being political .......................................................85 Chapter Five: Democratic Practice as Paradigm Case of Collective Action 90 5.1 Vote-aggregation as democratic practice ......................................................91 5.1.1 Lack of rules for aggregation ...............................................................93 5.1.2 Lack of motivational effect ..................................................................95 5.1.3 The failure to be democratic ................................................................97 5.1.4 Citizen responsibility in aggregative democracy .................................99 vi  5.2 Public deliberation as democratic practice .................................................102 5.3 Group deliberation enhances collective responsibility ...............................109 5.4 Case discussion ...........................................................................................113 5.5 Replies to some possible criticism ..............................................................118 5.5.1 Failure to generate normativity ..........................................................119 5.5.2 Against value pluralism .....................................................................122 Bibliography 128   vii  List of Tables  Table 1 ...................................................................................................................... 61 Table 2 ...................................................................................................................... 94  1   Chapter 1: introduction The writing of this dissertation has been animated by a concern with how best to position responsibility in public life, in the multiple situations in which individual intentions and actions interact and overlap with one another in a community. One feature of the dominant understanding of the concept of responsibility is a strong emphasis on the individual. The relevant concerns in this regard have to do with the free will of individual moral agents1. In public life, however, the concept of responsibility invites a broader and more complicated examination. The approaches recommended by individualistic theories of responsibility do not provide a great deal of insight into collective responsibility in public life. Moreover, such approaches create some undesirable consequences for understanding responsibility in collective and public life. The particular aspect which this dissertation takes as its central is the disappearance of collective responsibility. Roughly speaking, the disappearance of collective responsibility refers to a situation where the actions of a group of individuals result in consequences for which no single individual party can be held responsible. This phenomenon has its variants in many parts of public life, and is conceptualized in different ways, common formulations of which include "the social trap," "tragedy of the commons" and "negative externalities."                                                    1 Although philosophical work on moral responsibility is often intertwined with discussions of free will, there are a number of recent exceptions. However, broadly speaking, it is reasonable to assert that a tendency towards individualism has characterized the work of moral philosophers.  2  Before I outline the scope of this dissertation, I offer a vignette drawn from the novel Enduring Love by Ian McEwan.2 In doing so, I hope to create a vivid picture of this dissertation's problem prior to engaging in philosophical reflection. The vignette, then, comes at the very beginning of the novel, where a hot-air balloon pilot is struggling against powerful gusts of wind. A child is constrained in the basket. If the pilot fails to regain control, the balloon will almost certainly be blown into high-tension electrical lines nearby.   Some people, fortunately, spot the problem and converge on the balloon to help the pilot to secure it. They are eager but uncoordinated, as explained in the narrative: “there may have been a vague communality of purpose, but we were never a team.”3 At first this lack of team organization does not matter. Each one tries, after his or her own fashion, to bring the balloon down to ground. The wind keeps on gaining power, however. They are lifted above by the wind, just dragging and on the wrong edge of the control. If they can act together and all hold on to the ropes, they can rescue the child and, having done so, return to safety. However, any cohesion among the people fails to materialize. The novel narrator in the novel thinks to himself:  “But there was no team, there was no plan, no agreement or to be broken…someone said me, and then there was nothing left to be gained by us. A good society is one that makes sense of being good. Suddenly, hanging there below the basket, we were a bad society; we were                                                  2 This story is also employed by Christopher Kutz in his paper “The Collective Work of Citizenship” (Legal Theory, 8(2002), pp471-494) when discussing collective actions. I was inspired by Kutz and read the original story in Ian McEwan’s book. My summary of the story has the larger aim of filtering its narrative through the lens of collective responsibility.  3 Ian McEwan, Enduring Love, 12 3  disintegrating…The moment I glimpse a body falling away — but whose? —and I felt the balloon lurch upward, the matter was settled; altruism had no place.”4  The story ends dismally, with the would-be rescuers dropping off, one by one, out of fear. Only one is left hanging, with a little more breavity, rising, then falling.   The failure of the collective, including the failure to form a collective when necessary, instead of altruism, is the theme in the story which I intend to investigate in the dissertation. How can we as individuals but nevertheless social−beings, become better at taking responsibility for collective action in public life?   My dissertation is nowhere close to a complete answer to that question. The proposal here is a rather modest and perhaps a little convoluted one. I intend to establish the following starting points for elaboration in the study that follows. First, collective responsibility, by which I mean the responsibility taken by individuals as contributors to collective action, is to be determined by contextual and relational factors, in addition to the individual wills or acts which constitute the collective action. Second, different group structures and collective decision-making mechanisms affect the competence of individuals in taking collective responsibility. Lastly, using democratic decision-making as a paradigm case of collective action, group deliberation is superior to vote aggregation in enhancing collective responsibility taken by individuals in political life.                                                    4 Ibid, 16 4  The numerous sub-topics which descend from the central concern of this dissertation, such as, for example, collective action, moral responsibility and deliberative democracy are highly contested in the available scholarly literature to date. The line I try to walk in my dissertation is highly self-constrained, in the sense that my proposal does not directly engage with some relevant debates over the big issues relating to these concepts or propose to have a solution to these controversies. For instance, the metaphysics of action, the issue of free will, and the ontology of collective entity are all relevant topics to the question I attend. However, this dissertation leaves questions under these topics open ended. The goal in what follows is to show a way that some approaches of understanding collective action, responsibility and democratic practice can rub up against each other, and thereby arrive at a plausible answer to the question raised earlier: How can individual social−beings like us become better at taking responsibility for collective action in public life?   In discussing the concept of collective responsibility, I make my argument in response to two kinds of individualistic skepticism, which I call, respectively, reductive individualism and soft individualism. In chapter one, I respond to reductive individualism, which is the hard skepticism toward collective action. Reductive individualism questions the very existence of collective action, and argues that talking about collective action is merely metaphorical, an indirect way of ascribing actions to the individuals within a collective. A collective agent lacks a reality independent of its constituent individuals. Therefore, to ascribe actions to a group is either sloppy shorthand for referring to the decisions and actions of individual human beings or mere nonsense.  5  If reductive individualism is correct, we would lose the means by which to make sense of collective action. The suggestion seems to be that any investigation into collective action, with further refinement and the addition of great detail will ultimately dissolve into the study of individual actions. In this chapter, I discuss three approaches which have been adopted by philosophers to make sense of collective action: agent-collectivism, intention-collectivism and participation-collectivism. The overall goal of this chapter is to establish the plausibility of collective action, and to show that collectivity comes in various kinds and degrees. So far as responsibility is concerned, an account of collective action needs a minimal threshold, an enhancement of its capability of taking responsibility through further structural integration.   In chapter 3, I tackle the second kind of skepticism, soft individualism, according to which individuals are the proper subject for normative assessment despite an acknowledgement that collectives are considered capable of forming intentions and performing actions. In other words, soft individualism argues that although collective action is possible, responsibility can only be attributed to individuals, because the assignment of responsibility understood by soft individualism is subject to roughly three principles: principle of individual difference, control principle, and autonomy principle. I argue that theories of soft individualism face a significant challenge: to translate responsibility for collective action into the responsibility of individual members based on individualistic doctrines of responsibility. For soft individualism to be valid, this translation would ideally be perfect in the sense that the reduction of collective responsibility to individual responsibility has no residual elements. Every responsibility there is to be taken for a collective action can be represented as individual responsibility 6  based on an individualistic doctrine of responsibility. In the second part of chapter 3, I provide three cases of collective responsibility that cannot be imperfectly translated based on the individualistic doctrine: over-determination, bystander effects and discursive dilemma. In light of these cases, I wish to show that individualistic doctrines of responsibility and the related principles are inadequate for helping us dealing with collective responsibility. Responsibility in the context of collective action invites a rather different understanding.   Given that it has been established that individualistic doctrines of responsibility are inadequate for assisting us to understand collective responsibility, Chapter four sets forth two necessary features for a persuasive account of collective responsibility: context-sensitivity and respondent-dependence. The aim here is to demonstrate the necessity of an alternative view of responsibility which can better guide us to navigate public social life. This chapter serves as a transitional chapter, which leads my dissertation from moral philosophy into political practice.   In chapter five I use democratic practice as a paradigm case of collective action, and aim to show that as a collective decision-making mechanism, group deliberation is better than vote aggregation for enhancing collective responsibility. The first part of chapter five lists three normative defects of vote aggregation, and the consequential hindrance to citizen responsibility. I then discuss group deliberation as a democratic practice, its related normative advantages compared to vote aggregation, as well as its strength in enhancing collective responsibility. I also sketch a hypothetical scenario of how group deliberation plays out as a practice of political decision-making. The chapter 7  concludes with an attempt to respond to two possible criticisms: the skepticism toward the normativity that group deliberation might generate and the worry that group deliberation would generate any normativity at all.   Articulated in its strong form, the goal of this dissertation is to provide a possible political solution to the problem of collective responsibility. The theoretical inadequacy of ethical doctrines in dealing with collective responsibility is not necessarily resolved through the formulation of superior explanatory mechanisms. The function of normative concepts such as collective responsibility is to assist us in appreciating the moral context in which we live, and to better conduct ourselves in such a context. My thesis, ultimately, can be read as offering a normative justification for the superiority of deliberative democracy over vote aggregation. Though the practice of deliberative democracy does not, as the theory is currently developed, issue in a morally tenable, well-functioning account of collective responsibility in a community, it serves as a plausible starting point for the formation of such a conception.  The weak reading of this dissertation's goal locates its central aim as bringing ethicists and democratic theorists into conversation, to explore their shared, but as yet unremarked interest in collective agency There is a possible mutual misunderstanding between normative philosophy and political science in relation to this topic The former has difficulty in seeing normative values being generated by actual political practice; whereas the latter has little faith in useful guidance provided by philosophical investigation in political life. This dissertation attempts to correct this misunderstanding: some moral problems are likely to be less challenging when the right kind of political 8  institutions and social practices are implemented, especially those capable of generating well-considered collective judgments. Meanwhile, some political arrangements acquire their legitimacy by enhancing our moral practice.     9  Chapter 2: Making Sense of Collective Action To lay down the foundation of collective responsibility, the very first issue to address, as far as I see it, is the idea of collective action. It is a notoriously hard topic. What I hope to achieve in this chapter is not to develop or defend any particular account of collective action. The goal is to show various ways people have proposed to make sense of collective action, and to present an overall plausible picture of it by combining these proposals together. The role of this chapter is to undermine the hard skepticism against collective action, which I call reductive individualism, and to draft a possibly incomplete yet plausible account of collective action, which can be used in following chapters when discussing collective responsibility.   I begin with a detailed explanation of skeptical attitudes toward collective action, which I take to be a serious challenge to the very foundation of group collectivity. I then present three major approaches adopted by various philosophers in establishing the collectivity in group intentions and actions. Lastly, I will propose an account of collective action, which delineates some ways through which one can locate collectivity in human intentions and actions, and illuminates certain features of collective action revealed by this plausible view on collective action.   2.1  Some general ideas related to reductive individualism Philosophers of psychology such as John Searle have argued that the intentional states as “mental states” exist only in entities with minds that have relevant mental 10  capacities, such as, for example, the state of consciousness5. In order to enact intentional states, it would seem self-evident that the foundational requirement is biological---to have a brain. Intentional states are, apparently, related to the biological materials that consist in the individual brain. Consciousness and mental states are nothing but “variable rates of neuron firings in different neuronal architectures”6. Intentions should be understood primarily as brain events. Following this line of reasoning, anything without a brain does not have a mind, and therefore cannot have mental states such as intentions.  A collective entity does not have a biological brain of its own, nor do the constituent members physically share this biological foundation. A group of people, say, do not share one brain in any serious non-metaphysically sense. It can be deduced that collectives as a whole lack mental capacities, do not have emotions or desires, cannot form beliefs or intentions, and cannot, as a result, perform intentional actions. In other words, there exists no intention or action over and beyond individual agents. I call views of this sort reductive individualism, or hard skepticism towards collective responsibility. The basic underlying idea is that the brain and only the brain serves as the necessary condition for an entity to be regarded as an agent, for only entities with brains can form beliefs, have desires and perform actions.   People do, however, ascribe mental predicates, to groups, when, for example a Philosophy department decides to open a new position or Detroit considers filing the                                                  5 See the work by Mind/Brain-identity theorists, such as J.J.C. Smart and U.T. Place. I use John Searle as my main interlocutor; for he also has interesting work done concerning collective action. See Searle, J (2002) “The Problem of Consciousness”, Consciousness and Language, Cambridge University Press.  6 Searle, “The Problem of Consciousness”, 9 11  largest ever American municipal bankruptcy. This way of talking saturates everyday language use. For cases like this, reductive individualists argue that to ascribe mental predicates to groups is an indirect way of ascribing such predicates to the constituent members of that group. Any talking involving collective agent is merely metaphorical, a way of summarizing its individual members. A collective agent lacks a reality independent of its constituent individuals. Therefore, to assign responsibility to a group is either sloppy shorthand for referring to the decisions and actions of individual human beings or mere nonsense. The rationale behind this claim, in the sketchiest form, is that responsibility is predicated on action, and action is predicated on related intentions. Given that groups cannot be said of being entities with intention, they cannot be appropriate bearers of responsibility.   The phenomenon of collective action is, as far as I see it, more rich and complicated than what reductive individualism is willing to entertain. In the following, I will introduce three ways to make sense of collective action proposed by different philosophers, namely explaining collective action based upon collective agent, collective intention and participation in collective, which are the views labelled as agent-collectivism, intention-collectivism and participation-collectivism. For each view, I provide explanation and some criticism or deficit. The goal of presenting these views of collective action is to support an observation I am to make in later part of this chapter, which is that collectivity in group action comes in degrees depend on how well-integrated and sufficiently-interacted group members are. These theories of collective action, from my point of view, represent different developmental stages of collectivity in groups. The hope is to show that collectivity in group action can be developed and enhanced, and also 12  to show that we need a minimal threshold to start constructing collectivity in group action. Both points will be further elaborated in the later section after we take a close look at various views on collective action.   2.2  Agent-collectivism Reductive individualism and its objection to collective responsibility can be addressed in a number of ways. Philosophers argue for many possible sources that give rise to collective action. The first kind I will introduce is the view that collective actions might be performed by a collective agent with collective intentions. One early proponent of this view is Peter French, who criticizes the notion that biological existence is essentially associated with the concept of a person, and argues that corporations can be seen as “full-fledged moral persons” and “subjects of right”7 Margret Gilbert makes a more recent contribution to this theoretical work. Her notion of a “plural subject” is an idea of “collective agent”, which can be considered as a view of agent-collectivism.  Agent-collectivism refers to the views that groups and collectives are entities capable of forming intentions and performing action just like individual agents, because collectives share similar features with individual agents. Agent-collectivists might even agree that this renders collectives suitable for responsibility ascription. These features include being able to form intentions, having deliberative capacity, being responsive to reasons, having the ability to carry out plans and to perform actions. Certainly, not all groups or collectives can be regarded as agents of this sort. Careful philosophical                                                  7 P. A. French, “The Corporation as a Moral Person,” 208 13  treatment is needed to find out which collectives meet the basic conditions to qualify as functioning agents, and what these conditions are likely to be.   2.2.1 Peter French on collective action Peter French starts his investigation on these conditions by pointing out that there are two types of collectives: aggregate collectives and conglomerate collectives8. The aggregate collective is merely “a collection or a group of people”9. It refers to a group of people who share some common features. Such common features are most likely random, such as left-handedness. Moreover, there is no varying membership serving as the foundation that sustains the existence of the group. In other words, the identity of the group cannot survive the change of particular individual members. For instance, the group of “left-handed people in the philosophy department” would not be the same group if any member of the group of left-handed persons in the department opts out from the group or the group includes a new member who is not left-handed. What is true of an aggregate collective is generally reducible to what is true of its members. More importantly, in an aggregate collective “there is no established decision procedure for determining group actions and often no strong bonds of solidarity”10, and they are therefore “not intentional agents in and of themselves”11 according to French. Due to this lack of decision-making procedures and organizational structure, the group has no emergent or independent agency beyond its individual members. Far from a well-integrated organic whole, the group is a rather loosely-connected sum of individuals.                                                  8 French, “Types of Collectives,” 37 9 Ibid. 10 Ibid, 43 11 Ibid, 41 14  Therefore, holding these aggregate collectives responsible is equivalent to distributive ascription of responsibility to each of the individual members.   In contrast, a conglomerated collective, whose “identity is not exhausted by the conjunction of the identities of the persons in the organization”, often has what is Peter French calls Internal Decision Structure (IDS). With this IDS installed in a group, its individual members would have to follow a set of rules and principles in order to contribute to the collective decisions. These rules and principles direct, and even define the individual members' decisions and behaviours. According to French, the existence of IDS even makes possible the subordination of individual desire to the needs and mission of the collective. Through the interaction and cooperation among individual members emerges the irreducible group agency, which in its turn justifies the attribution of collective responsibility12. The main collective subjects in French's theory are modern corporations, where people exercise their power in an interpersonal network and follow certain structural norms which they would not have followed as biological individual human beings.   2.2.2 Gilbert on collective action  Margaret Gilbert's account of the plural subject is a more recent theory of agent-collectivism. Gilbert has been arguing against reductionism from a very early stage of her discussion on collective responsibility. She claims that collective agency is a result of mutual obligations among individual members and that it is irreducible to their individual                                                  12 One feature of French's theory of moral responsibility is that it very specifically focuses on moral blames.  15  intentions. There is a collective agent behind every acting-together, large or small, temporally-limited or enduring. An often-discussed hypothetical scenario illustrates her point rather well: James and Paula are out on a walk in New York, which is a joint activity. Various things can happen during the course of this joint activity. For instance, James may be a naturally fast walker and starts to move ahead. Gilbert thinks that Paula is entitled to respond in various ways: to call after him, to demand that he slow down, or to rebuke him for going too fast. Gilbert intends to show that Paula, in virtue of being a part of this joint activity, has a very special standing against James which other people do not. In another word, Paula is in a special position to demand that James acts in a manner appropriate to their joint activity, and to rebuke him if he acts in a manner inappropriate to it13. Meanwhile James holds the same position against Paula. The standing here, as Gilbert points out, is a function of their joint activity. Participating in a joint activity gives rise to rights of some sort held by participants against one another as well as obligations towards one another.   As shown in the discussion of James and Paula walking together, Gilbert’s account of joint activity has significant normative implication. The collectivity here goes beyond the intention which happens to be shared by a group of people or a collective decision-making mechanism upon which they agree. Participants of a joint activity, as Gilbert proposes, need to be viewed as a plural subject so that their acting together becomes more than the mere summation of individual actions. This plural subject                                                  13 Gilbert, Political Obligation, 103. 16  possesses agency of a different kind from the agency of its individual members, for it is now a plurality of members, “a single body”14.   The essential condition that holds individual participants together as a plural subject is their commitment to the group. According to Gilbert, participants have to mutually express their readiness to partake in a joint activity, and commit themselves to the necessary goals, decisions, and planning for this activity. Consequently this joint commitment places normative constraints upon group members. Individual members are obligated to take this commitment into consideration and to coordinate their future actions with it. Entering into a shared commitment obligates individual participants to allow their wills to be bound by it and to sees it as sufficient reason for acting in a consonant manner. Gilbert insists that the parties to a joint commitment are “surely liable to feel that they owe each other conformity to the commitment”15, and that they “have the standing to demand certain actions of other parties”16. The recognition of one member's owing another his part of the joint activity may even trumps considerations relating to one's own inclinations and personal interests.   2.2.3 Remarks and criticism on agent-collectivism Both Peter French and Margret Gilbert's accounts of collective actions start from the understanding of groups being agents independent of their constitutive members. Agent-collectivist philosophies believe that reductionism is far from a satisfactory explanation of genuine collective actions. They tend to support the view that, in order to                                                  14 Gilbert, “Rationality in collective action”, 16 15 Gilbert, A theory of political obligation, 135 16 Ibid. 17  make sense of collective actions, we need, first of all, to see the performer of actions as an agent, or, to be more precise, as a group agent. They point out that some emergent features of a group are generated by the relations and interaction among its individual members. These emergent features, such as French’s Internal Decision Structure of a group or Gilbert's joint commitment, give rise to group agency which is ultimately irreducible to the individual level. This new agency acquires independence from its composition, and serves as a starting point to discuss normativity at the group level.   Criticisms and concerns follow when agent collectivists are pressed to offer more details about group agency. Two fundamental issues are how to form a group agent and how to quit a group agent once a member of it. These two issues have to be answered adequately, for they clarify the very beginning of a group agent, where individuals becoming members, and the termination of this group agent, where members become individuals once more.   A frequent explanation for the beginning of a group agent invokes agreement among individuals. The general idea is that those who act together in a certain way must first enter an agreement to do so. The upside of this idea is that if we take agreements to be foundational to group agents, we would have some solid understanding of important aspects of a group agent: its members, its goals, plans, actions, as well as its corresponding obligations and responsibilities. The downside of the agreement hypothesis is its over-exclusiveness. It fails to count for those activities which should be regarded as collective actions, for many collective actions do not happen with explicit agreements in place. Most cases of collective actions such as tango dance, family dinner, 18  or even war17 occur in the absence of any explicit agreement into which participating parties enter. Less demanding criteria are needed to include those cases of collective actions which happen out of spontaneity without any previous agreement. Gilbert proposes that a group agent, or a plural subject in her word, occurs when “each understands what a joint commitment is and expresses all that is needed in his or her part to bring such a commitment into being”, and this behavioural expression of mutual commitment becomes a common knowledge. These conditions do not necessarily result in an explicit or well-formed agreement at the same time. An agreement might be a product of it, yet is not required. Generally the formation of a group agent is a gradual process that extends in time, as its group members eventually become ready to commit jointly. It can be hard to isolate specific moments, so she proposes various factors may be indexes to identify group agents: a sense of belonging among individual members, a shared belief and value system; a cultural or community convention, etc. Without explicit agreement in place, the group members might not be well aware of what the joint commitment is, nor can they spell out the particular goals of this collective body. Nevertheless, a more general and loose standard better assists us in appreciating the variety of possible collective actions.   Compared to the loose standard adopted by Gilbert to identify the beginning of group agency, the standard which she holds for its termination is rather strict and demanding. As Gilbert argues, once a member of a group, one is bounded by the joint commitment and therefore owes his or her share of the collective action to other                                                  17It is debatable whether a war or a quarrel can be understood as collective action. M. Gilbert gives elaborated discussion on this issue in her book Political Obligation. She argues a quarrel ought to be seen as collective action, and her account of joint activity enables us to do it.  19  members. One can only be freed from this normative bond, according to Gilbert, in three ways: satisfaction of the joint commitment; implicitly agreed and mutually recognized fade-out of the joint commitment; and lastly non-unilateral recession. Notice that termination in all three cases has to be done collectively---as Gilbert observes, “a joint commitment must be terminated jointly.”18 This means that one member's contrary acts or personal fiat does not grant his freedom from the group, for one cannot unilaterally terminate the joint commitment.   Now let us re-examine the picture of collective action offered by agent-collectivists such as Gilbert, especially the entrance conditions and exit condition for an individual to become or stop being a member of a group. Admission to a group requires only that the individual have some implicit behavioural expression of the readiness for joint commitment. It is not expected that he has complete knowledge of what he commits himself to, either the overarching goals or the specific plans of the group. Entering into the commitment is not necessarily a matter of “deliberation or forethought or consciously noted” 19 . When the plural subject is a large population, members may remain anonymous. Yet once a person becomes a member of the group, his actions becomes the business of the others. He is entitled with the standing to demand certain actions of other parties, while, at the same time being burdened with correlative obligations. No one party is in a position unilaterally to decide on a joint action in a group. Consequently, when quitting the group, one's own withdrawal is not effective unless it is proposed to the whole group, and is known, perceived and accepted by other group members. This                                                  18Gilbert, A Theory of Political Obligation, 143. 19 Ibid, 145 20  understanding of collective actions sets up a rather low entrance point for people to become a part of a collective, whereas it demands immense effort to quit the collective. This asymmetry between the entrance and termination of becoming a group member might problematize Gilbert's account of collective action.  Agent collectivism has numbers of merits: it explains the survival of a group when particular members of the group change; it brings out the normative dimension of collective action, which other accounts would not deliver; and its objection to reductive individualism is effective by upholding a solid idea of group agency, emergent and independent from individual agents.   However, I see two shortcomings of agent-collectivism theories, such as Gilbert’s account. Firstly, a reasonable account of group agent needs a better constructed, more or less symmetrical membership standard for inclusion and exclusion of members of a group. Without sound criterion to assist us identifying the constitutive members of a group agent and its dissolution, agent-collectivism is yet to provide a solid foundation upon which to build a theory of collective action, and thereby offers little insight about collective responsibility. The secondly, more serious weakness of agent-collectivism, related to the lack of robust standard of membership, is that the asymmetry between inclusion and exclusion might be hard to fix due to the difficulty of preserving two virtues of a reasonable account of collective action at the same time: to accommodate the complexity and diversity of phenomena of collective action on the one hand, and to demand normatively binding relations among individual members on the other hand. If agent collectivism were to prioritize the explanatory scope of their theory and treat 21  activities such as street rioting as collective action, it seems that the account has to do so at the cost of strong normatively binding force collective action has over its members. Some group activities take place where participating members hardly form any obligatory relations with one another. If agent collectivism were to insist that group members are morally committed to one another in a collective action, this account would to fail to represent many cases of group activity as collective action, and thereby lose a great deal of explanatory power.   2.3  Intention-collectivism The second approach, intention collectivism, seems to place a less demanding requirement on collective action. It argues that the collective actions are performed by individual agents out of shared intentions. Micheal Bratman is a major proponent of the theory. He points out that the key component of collective actions is shared intentions, which consist of the interpersonal structure of participants' intentions as well as the resultant bargaining and planning among them. This interrelatedness of participants' intentions, instead of the existence of a well-formed collective agent with collective intentions, gives rise to collective action.   2.3.1 Bratman on shared intention Collectivity, according to Bratman, does not necessarily rely on some emergent group agency independent from its participants. Moreover, to look for some sort of normative force arising from mutual obligation is erroneous. Bratman points out in a 22  footnote to his article “Shared Intention”20 that a strong sense of group agency as well as the resulted binding force of normative group obligation is neither sufficient nor relevant for a shared intention. The binding force of a joint commitment does not guarantee intentions of individual members to act accordingly on their parts. Theories of collective actions of this sort state that the task of understanding collective action is not to investigate group agent, but rather the shared intentions among group members.   Bratman, as an intention collectivist, thinks investigation on the collectivity in a group action should start from the primitively collective participatory intentions individual member has. The right question to ask is what happens when individual intentions concern something that is to be done by more than one person. In other words, we need to understand the nature of individual intentions taking the form of I intend we do action A.   Bratman holds a planning theory concerning intentions in general, which sees intentions as “guiding intentional actions, and as coordinating forms of planning central to our temporally extended agency and to our associated abilities to achieve complex goals across time and inter-personally”21. Intention, in Bratman’s picture, is practical attitude marked by its role in future-oriented planning. Now to apply this understanding of intentions to group level, Bratman thinks in the shared case intentions need to carry out the similar roles as it does at the individual level, which is to structure collective actions in order to achieve future group goals. In his influential article “Shared Cooperative                                                  20 See Bratman, M. E. (1993). Shared intention. Ethics, 104(1), 97–113. 21 Bratman, “Dynamics of Sociality,” 3 23  Activity”, Bratman provides a rough and preliminary picture of shared intentions. In order for a group of people to share intentions, each person needs to recognize the other as a constitutive participant in the shared intentions. Also, each person intends the other person's relevant intentions to be effective in carrying out joint activity. Moreover, each intends to fill in and execute his individual plans to participate in the joint activity in ways that mesh with each other. The cooperative activity based upon shared intentions has certain features. For instance, each member of the cooperative activity attempts to be responsive to the intentions and actions of the other, knowing the other is doing the same. If some member needs help and support to play his role in the joint activity, other members are prepared to provide such help.   2.3.2  Bratman’s shared intention in contrast to Gilbert’s joint commitment We can revisit Gilbert's walking together case and examine how Bratman's theory would play out for the scenario. In Bratman's understanding, Paula and James, by walking together, are engaged in a joint activity for the following reasons. Paula and James know each other as co-participants in this joint activity. Each knows his or her role in the joint activity, and makes a personal plan and intends to carry out the individual part. When one party encounters certain difficulty, for example Paula cannot keep up with James, the other party is ready to offer support and to help complete this joint activity, meaning James is willing to slow down for Paula.   Bratman's approach to collective actions differs from agent collectivism such as Gilbert's in one major aspect: Bratman refrains from imposing strong normative 24  dimensions upon the participating parties. Participants of a collective action are not regarded as having formed any group agent. Parties do not owe it to one another to perform their parts of the joint action. In other words, no obligation or entitlement is generated for individuals who partake in the shared intentions. Due to the absence of normative bonds among the participants, the individuals involved in collective actions through a shared intention do not necessarily act in unity or form an agent. The collective actions, as Bratman demonstrates, do not need collective agents as performers. Collective actions are made possible by individual participation and coordination within the group.   One essential condition Bratman holds for shared intentions is what he calls “the meshing of sub-plans”. According to Bratman, for two people to participate in a joint action, they do not necessarily have to aim at the same goals or agree upon some group plan with specific steps laid out. Through a joint action, parties may try to achieve rather diverse goals independently. For instance, in the walking together case, James's goal is to get some fresh air while Paula wants to grab a coffee. For them to walk together, neither does James have to become a coffee drinker nor does Paula have to try to relax. What they need to do though is to ensure the compatibility among their sub-plans. For instance, James is willing to walk towards a coffee shop while Paula would not buy her coffee from the cafeteria downstairs. James and Paula's joint act is more coordination than cooperation, which is “cooperatively neutral”22 to use Bratman's term. Their joint act can be completed by its individual participants for different reasons and goals. Instead of normative constraints upon the participants emerging from mutual commitment in Gilbert's theory, Bratman thinks that the defining feature is the structural compatibility                                                  22 Bratman, Faces of Intention, 105 25  and consistency among the individual intentions and actions, which is the “meshing” of sub-plans.   In order to ensure the meshing of sub-plans, individual participatory intentions are interrelated in certain way. If it turns out that the meshing of the sub-plan, and consequently, the foundation for joint action happens by accident, there is little difference between some coincidence of individual acts and a genuine case of joint act. Therefore the meshing of sub-plans has to be caused by the interrelatedness among individual participatory intentions. This way when the sub-plans do not mesh, participants are disposed to make them compatible and consistent. Bratman points out that the mutual responsiveness among individual participatory intentions is one significant feature of shared intentions. When one party is having difficulty in carrying out his part of the joint act, other parties are willing to help and to compromise, to change their sub-plans accordingly, to ensure the completion of this joint act.   Though Bratman refuses to accept strongly normative concept of shared intentions, such as Gilbert's joint commitment, he does argue that shared intentions do place minimal, normative constraints upon individual participatory intentions. These constraints come from rational requirements of practical reasoning, such as consistency and mean-end coherence. Again, intentions, according to Bratman, play a role of future-directed planning. They are constrained by one's plan be “means-end” coherence, efficient solutions to problems, as well as the consistency among the plans. For instance, Paula could not have intended to not go to the coffee shop given she wants to buy coffee there. Similarly, in the case of shared intentions, James could not have coherently 26  intended not to walk towards a coffee shop while intending to take a walk with Paula, given that he knows Paula intends to get her coffee by taking a walk with him. To succeed in taking collective actions, participants ought to mesh their sub-plans because they are subject to intersubjective rational requirements of this sort. Only when participants' sub-plans do not violate one another but would, rather, can be successfully executed together can these participants engaged in a genuine case of joint act.   As mentioned by Bratman himself, his account of collective activity is a rather individualistic one in spirit, because it portraits shared activity “in terms of attitudes and actions of individuals involved”23. Bratman's understanding of collective action starts from the internal intentions of individual participants. Shared intentions are rooted in the specific way the individual forms his intention with respect to the joint activity. Participants are willing to, or constrained in some sense, to take other members' intentions as necessary inputs into their practical reasoning and form intentions correspondingly. Collective actions can only be motivated by shared intentions when individual participants' intentions become mutual causes.   2.3.3  Criticism on Bratman’s theory of collective action  I see one major problem in Bratman’s theory of collective action in relation to collective responsibility: it does not capture the responsibility-relevant factors of collective action. In other words, emphasis on compatibility among participating intentions does not assist us in understanding collective action normatively. The relevant                                                  23Bratman, Structure of Agency, 108 27  cases are those in which, when people act together, individual participants act under a certain amount of disinterest and uncertainty of the intentions and actions of others. When a person acts, especially with other people, he is causally related to many, possibly infinite, events and consequences. In many cases, a person's action coordinates with others and results in certain consequences, without necessarily intending the coordination and its consequences.   Providing examples of joint activity of this sort requires drastic revision of James and Paula's case. Suppose a routine has been formed between James and Paula such that every afternoon they take a walk together when James intends to get some fresh air while Paula wants to get some coffee. In one particular afternoon, what changed was that James, instead of just intending to get some fresh air, intended to harm the lady who works at the coffee place. It might be easier for James to do that if Paula went with him, because the potential victim would be less suspicious if they both enter that coffee shop. Paula was aware of James's plan, took no interest in his intentions, and decided to walk together with James simply because she wanted a cup of coffee. Bratman's theory of collective action as meshing of sub-pans runs into trouble in helping us understand Paula's role in this scenario. On the one hand, Paula's sub-plan of going to get her coffee meshes with Jame's sub-plan of harming the lady who works at that coffee shop. They both intend the joint action which is to walk to the coffee shop but for rather different reasons. According to a weak reading of Bratman's theory, Paula did participate in a joint action that resulted in the coffee lady being hurt. On the other hand, Jame's intention to harm the coffee lady did not give rise to Paula's intention of getting her coffee. The interrelatedness between James's and Paula's intentions, based on a strong reading of 28  Bratman's theory, is not the right kind to define their action as a joint action. Therefore Paula cannot be regarded as a participant.   Bratman's theory gives little instruction in dealing with cases as such where, firstly, agents intend a joint action for different reasons; and secondly, the joint action is causally related to some event or consequence C; and thirdly, consequence C is not intended or endorsed by some, or possible none of the agents who take part in the joint action. In these cases, some agents are causally involved in the joint action. However, their causal roles are not represented in their intentional content. This imposes difficulty in deciding the nature of specific collective action and the roles participants play in it.   2.4  Participation-collectivism Christopher Kutz offers a complicity theory of collective actions to include the cases which are left unaddressed by previous approaches. In these cases, agents “knowingly aid collective enterprises”, but do not seem to meet “the minimal criteria of intentional participation”24. The intentional structure of participants in such cases does not have essential features such as joint commitment or mutual responsiveness. According to Kutz, in order to identify individuals acting collectively, what we need is to see if “the members of a group overlap in conception of the collective end to which they intentionally contribute”25.                                                   24Kutz, Complicity,164 25Ibid 29  2.4.1 Kutz on group participation Let us examine a case discussed in Kutz’s work. A member of the mob starts throwing rocks at the prison guards. Soon, others join him, and they collectively storm the prison. In this case, the mob members spontaneously acted collectively, with no existing shared commitments or plans. However, they did share the goal of breaking down the prison and only through this commonly shared goal can one make sense of their individual behaviors. Yet they did not plan the collective action out beforehand. Perhaps none of the particular individual action performed is anticipated or known by other members, yet together these individual actions constitute the collective action. By the simple fact that mob members participated in a joint action intentionally and saw their individual behaviors as contributing to a collective end, we are entitled to say that each individual mob member participate in breaking down the prison complicitly.   The participants, in Kutz's picture, are in some general sense people who act concurrently with others. The core, minimalistic notion of collective actions requires only that “individuals act on overlapping participatory intentions.”26 Some participants do not intend the consequence, some participants might have rather compartmentalized knowledge of others' tasks and identities, and they can have very different conceptions of what they do while still acting jointly.   Most of the weight of Kutz's theory lies on the idea of “sufficiently overlapping” among participatory intentions. As Kutz himself, points out it is essentially a “pragmatic concept and always a matter of degree”27, because there are always differences in each                                                  26Kutz, Complicity, 75 27 Ibid, 90 30  agent's expectations and conceptions of the group act. It is, as I understand it, highly context-sensitive. A group act can be jointly intentional under one description and not jointly intentional under another. Return to James and Paula's case. “To walk to the coffee shop together” is a rather clear description of their joint act. “To hit the coffee lady with a mug” is rather remote from the proper description of their joint act. However, “to cause harm to the lady” could be a controversial description of James's and Paula's joint act, when James intended to do it while Paula knowingly causally contributed to it.   Kutz proposes that the participation, or complicity as he puts it, as a key criterion for collective action.   “So long as the members of a group overlap in the conception of the collective end to which they intentionally contribute, they act collectively, or jointly intentionally. I call this the minimalist conception of joint action.”28 It is not necessary for the agent to be a part of a collective agent, or to intend and plan on the collective harm, or to know the collective consequence, or to have the capacity of making a difference to be held responsible complicitly. We individuals are the inclusive authors of collective action, often not the exclusive ones, as Kutz nicely puts. We do wrong together even when we do not have much sense of togetherness. Because our individual actions reach beyond our intentions, and their effects may significantly overlap with those caused by agents we neither know nor care about, we are always                                                  28 Kutz, Complicity, 122 31  already complicit with others, and participate in collective actions that we never could have anticipated.  2.4.2 Remarks and criticism One merit of Kutz’s minimalist account of collective responsibility is its superior descriptive coverage, which he takes to be significant for a successful account of collective responsibility. A minimalist account enables us to tackle the diversity and variety of collective action in regard to collective responsibility. However, my worry is that Kutz’s complicity principle focuses on intentional participation and leaves responsibility originated from collective inaction unscrutinised. Some collective inaction is clear cases of moral failure, which invites responsibility assigning. To accommodate cases of collective inaction and to justify responsibility attribution based on that, Kutz’s complicity principle requires revision. More investigation is needed to assess the structure of the interaction of micro-motives in the cases of collective inaction to complete this revision.  Let me start again with one case of this type of collective inaction. A famous, or infamous, example of this kind is the Kitty Genovese murder in NYC, in which a girl was raped and killed in an areaway while more than 30 neighbors watched out the windows – and none of them called the police. The apparent failure of action produced a national wave of horror and countless discussions among social psychologists and ethicists. In such a situation, it is clear that there is certain an individual barrier against calling the police. One might have to deal with a series of hassles if one steps in and calls the police. 32  Each observer may have felt a strong impulse of social conscience at the time, but simply hoped that someone else would make the troublesome phone call first.   Intuitively there was a collective moral failure, which does not result from individual wrongdoing in any sense. The wrongness of the onlookers or bystanders in this case does not lie in their intentional participation in the actions of the rapist. Moreover the wrongness does not lie in a single individual inaction if someone had in fact made the call. It lies in their collective inaction.   I wish to draw some attention to one important fact concerning collective inaction of this sort: the individual beliefs and intentions behind it are often highly interrelated and interdependent. The interdependence can happen in many ways: One person might decide not to take any action because he believes that other will do; it is also likely that one person decides not to act because he believes that other will also refrain from doing it. There is no joint commitment of any kind, or shared plans, or intentional participation. Nevertheless the collective inaction results from individual beliefs or intentions which are commonly shared or mutually caused. I propose that collective inaction, the failure of performing the needed or necessary action collectively by a group of individuals, can also serve as the foundation of assigning collective responsibility. Given the discussion of collective action is often driven by the concern of collective responsibility, certainly in this dissertation and in most literature29, the minimalist condition for collective action, as I see it, should be even less than what Kutz’s complicity theory is committed to.                                                   29  There are some recent attempts made to separate the discussion of collective action and the understanding of collective responsibility. See Marcus Hedhl, “The Collective Fallacy”, Philosophy of the Social Sciences, 2013/07. pp1-18 33   2.5  Collectivity in group action As shown in previous discussion, cases of collective action vary a lot depending on the way individual members contribute to the action, and on the degree of their involvement. The ideal type of collective action is likely to be those in which its individual contributors identify themselves as members of a collective, share a common goal, decide to cooperate, form individual intentions accordingly, come up with cooperation plans, carry out the plans, and achieve a common goal together. Individual contributors of this collective action are mutually aware of and responsive to one another’s intentions and actions. This is a general idea of well-formed collective action. More conditions might be applied to complete the description.   Before ending this chapter, I intend to make two observations based on previous discussion of various views on collective action, both of which will assist in approaching collective responsibility later in my dissertation. As far as I see it, in relation to collective responsibility, two questions needs to be addressed when we talk about collective action: What are the threshold criteria to hold a collective responsible for its action? And what enhances the agential power of a collective, and thereby makes it better in forming action? To address these questions, I would propose the following two ideas: first, an account of collective action serving as the foundation of collective responsibility attribution needs a rather minimal threshold for what amounts to collective action; second, collectivity in group action manifests itself in various ways and degrees. Compared to investigating the definite criteria used to identify collective action, the more 34  interesting and relevant aspect of collective action, at least for this dissertation, is to understand how collectivity and responsibility can grow, develop and get further strengthened for a group.   2.5.1 A minimal threshold account of collective action Recall my concern about Kutz’s account of collective action, where he argues that sufficiently overlapping intentional participation suffices collective action30. The worry is that Kutz’ emphasis on individual intentional participation still leaves collective inaction unaddressed. However, responsibility attribution seems to be proper in many cases of collective inaction. Though I doubt if Kutz’s own account of collective action is sufficiently minimal, Kutz himself does provide convincing reasons as to why we should prefer a minimalist conception of collective action. I shall follow him here and argue that there descriptive and normative reasons to do so.   The descriptive merit of a minimalist conception of collective action is straightforward: it has the maximal coverage of diverse cases of collective action. Minimal non-demanding criteria enable us to include the phenomena of joint activity in our scrutiny for collective responsibility. The other reason to prefer a minimalist account of collective action is normative. Whichever standard we employ, joint commitment, shared planning or intentional participation, not all people doing things together can be effectively identified as collective action. As we discussed before, using the case of James and Paula walking together, it can be said that James and Paula did not collectively hit the coffee shop lady, but not so much that they did collectively participate in the harm                                                  30 Kutz, Complicity, 89 35  the lady received. The extent to which some joint activity is a collective action is essentially a pragmatic concept and always a matter of degree, given the ever-existing differences in each individual’s understanding of the group act. Under some interpretation, some joint activity can be collective action and thereby responsibility attributable, whereas not so under the other. Therefore, given the major normative concern for collective action is to attribute collective responsibility, we do have prima facie reasons to take up a minimal, least demanding account of collective action, which licenses us to start considering the possibility responsibility attribution for all kinds of possible collective action.   Out of consideration as such, I advance a revision of Kutz’s complicity theory, where I emphasize on the involvement of affected individuals, instead of intentional participation, as the minimal standard for collective action. The hope is to further minimize what is required to be considered for possible attribution of collective responsibility. In cases of collective inaction, responsibility attribution is still possible under my proposed account of collective action, when there is a sufficient involvement of affected individuals, even when the involvement is not intended by individuals or not even aware of them.    2.5.2  Effectiveness of collective action This is not to say that a preferred account of collective action allows us to attribute responsibility to affected individual who are involved. Who are affected by a collective action as well as how much they are involved in it are both questions open for further discussion and various interpretations. This leads to the second question a 36  conception of collective action needs to address: how can a group of involved individuals better exercise agential power in their collective action? I think it is reasonable to expect that the more these individuals integrate into a collective, the more agential power they can exercise at the group level, the better they become in taking collective action.   Recall various theories of collective action discussed in this chapter. Instead of seeing them as unsatisfactory theories of identifying collective action, another way to utilize them is to regard them as efforts in demonstrating different ways in which collective action can take place. For instance, the strong form of collective action can be performed by a well-structured collective agent, whose members normatively commit to each other in acting together. Whereas the weak form of collective action can be sufficient overlapping of individual participation, with or without mutual obligation or shared intention.   Individual members’ involvements in different collective action vary in degrees and structures. Well structured, high degree participation of individual members in collective action often leads to well-integrated collective agent, strong agential power of the group, and achievement of collective goals. Poorly structured or unstructured, low degree involvement of individual members in collective action leads to the opposite. From well-formed collective action to its minimalist form, groups have varied degrees of capacity of taking collective action. The lesson is that the effectiveness of collective action comes in degree, and collective action can be enhanced by better organizing group structure and improving individual participation.    37  Chapter 3: The Inadequacy of Individualistic Approach to Collective Responsibility  The goal of this chapter is to defend the concept of collective responsibility against another individualistic understanding moral responsibility, i.e. soft individualism. By soft individualism, I refer to the view that normative responsibility cannot be properly ascribed to collectives due to the individualistic nature of responsible agency even if, contrary to reductive individualism, collective action is possible. This chapter consists of four parts: Firstly, I introduce soft individualism, and, more importantly, the relevant individualistic doctrines they are committed in the theory of responsibility. Secondly, I discuss two types of collective responsibility: a perfectly translatable type and an imperfectly translatable type. With these two components in place, we shall see the inadequacy of individualistic approach when translating imperfect collective responsibility. Thirdly, I provide more cases of imperfectly translatable collective responsibility to further strengthen my challenge to soft individualism. Lastly, I sketch out some features of an alternative understanding responsibility is expected to have, which would equip us better in dealing with paradox, ambiguity and conflicting judgments about collective responsibility that individualistic approach cannot adequately address.  3.1  Soft individualism In chapter two, I have discussed different approaches to defend the concept of collective action against reductive individualism, the view that a collective cannot 38  perform actions because it does not have the necessary physical or biological foundation, i.e. a brain, to qualify as an agent. As discussed in chapter two, many philosophers have suggested that collectives or groups can be said to possess and exercise some degree of agency, and consequently performing actions together and being responsible for them. Though these responses to hard individualism are more or less successful, objections and skepticism about the collective responsibility can still be raised from a different direction. This skepticism argues that group agency, depicted by various theories of collective action, is too weak to have much normative implication. In other words, the group agency one may observe in collective action is not strong enough to serve as a normative cornerstone for collective responsibility. This line of argument is what I call soft individualism. The challenge put forward by soft individualists to collective responsibility comes from the normative perspective: though collectives do perform action under certain descriptions, they qua collectives, as distinct from their individual members, are in no way qualified as proper responsibility bearers. As H. D. Lewis once wrote, “[v]alue belongs to the individual and it is the individual who is the sole bearer of responsibility. No one is morally guilty except in relation to some conduct which he himself considers to be wrong... Collective responsibility is ...barbarous.”31 Lewis thinks that if we believe responsibility is shared in a collective, it would remove any properly moral distinctions between one course of action and another. “All will be equally good, equally evil”, as he puts it, because “we will be directly implicated in one another’s action, and the praise or blame for them must fall upon us all without discrimination”. Though Lewis' account of collective responsibility as barbaric is short on supportive arguments, one can reasonably assume that what Lewis had in mind is attributing                                                  31 Lewis, “Collective Responsibility,” 32 39  responsibility to someone simply because he is a member of a group of which some other members committed a crime. The phenomenon to which he refers to as evil can be understood in reference to popular notions of “guilt by association” or “universal sin”.  In the remainder of this section, I will offer a step-by-step analysis as to how soft individualism reaches its conclusion that collectives cannot be considered as responsibility bearers for their action, though they can be said to perform actions, or even have collective intentions. Its reasoning, as I see it, consists of two major parts: 1) individualistic doctrines of responsibility and 2) the possibility of perfect translation from collective actions to individual participants' actions. By the end of this chapter, I wish to show that neither of these two doctrine stands, and that an alternative understanding of collective responsibility, possibly of moral responsibility in general, is required.  3.1.1 The individualistic doctrines of responsibility Commonsense morality and folk psychology, explicitly or implicitly, assume a set of doctrines when understanding responsibility. Some of them are further articulated and advocated by philosophical theories in libertarian tradition, which directs the investigation of responsibility upon the relationship between free will and determinism. I shall briefly introduce some of these principles32, and explain how they give rise to the skepticism held by soft-agent individualism against collective responsibility in the rest of this section.                                                  32 The three principles, individual difference principle, control principle and autonomy principles are roughly categorized and discussed by Kutz in his book Complicity (see p3-4). Here I offer an extended detailed discussion of these principles based on Kutz’s observation, though it is possible that Kutz disagrees with some parts of my development of his proposal. 40   Firstly, the individualistic doctrine of responsibility holds the principle of individual difference that a person is only responsible for the harm if something he did makes a difference to its occurrence. In other words, a person cannot be responsible for a consequence which he could not have changed even if he had acted differently. If some harm would occur regardless of what a person has done, or if that person could not have done otherwise to prevent the harm from happening, it is not justified to assign responsibility to him. If a particular future state were to occur regardless of a person's action or choices, his action and choices do not stand a relation to this future state and carry no relevant consequences. Therefore we cannot assign responsibility to him.  This principle can be put with more philosophical sophistication. The principle of individual difference can be roughly understood as a combination of the thesis of “moral entails casual” and the principle of counterfactual causation. Both doctrines are highly complicated, the full discussion of which goes way beyond the scope of this chapter. I will offer a succinct discussion here and hope to demonstrate their connection with the principle. The thesis of “moral entails casual” states that for any X, if an agent is responsible for X, it is in the virtue of the fact he caused X. The basic idea of counterfactual theories of causation is that the meaning of causal claims can be explained in terms of counterfactual conditionals of the form “If A had not occurred, C would not have occurred”. Or more precisely, using David Lewis's definition, C causally depends on A if and only if, if A were to occur C would occur; and if C were not to occur C would not occur, where A and C are two distinct possible events33. The individual difference                                                  33 See Lewis, D. (1973a):  Counterfactuals. Oxford: Blackwell. 41  principle can be seen as built upon these two ideas: based on counterfactual condition, if the occurrence of the participation of an individual member or its absence does not change the occurrence of consequential harm, it can be said that the participation of the individual member does not cause the harm; based on the doctrine of “moral entails causal”, if she does not cause the harm, consequentially, she cannot be responsible for it.   The discussion of causation leads to the second principle of individualistic doctrine: the control principle, which claims that a person is only accountable for events over which he has control, and whose occurrence he could have prevented. The control principle lurks in the background of the discussion of moral luck and gains immense attention from philosophers. The very reason why moral luck imposes difficult challenges upon the concept of responsibility is that people are generally committed to the control principle and believe that responsibility assessment ought not depend upon the factors that lie beyond the agent's control. Things such as the unintended consequences, the particular circumstance a person is situated in, or the formation of one's personality and character are beyond his control, yet play significant roles in determining the course of his actions and their consequences. These things excuse us from being responsible for what is done. As Thomas Nagel observes,   “Prior to reflection, it is intuitively plausible that people cannot be morally assessed for what is not their fault, or for what is due to factors beyond their control... Without being able to explain exactly why, we feel that appropriateness of moral assessment is easily undermined by the 42  discovery that the act or attribute, no matter how good or bad, is not under the person's control.”34  The essence of control principle is that what is not under our control is not subject to our agency, for we could not have brought any change to them even if we have acted differently. Where the agent's control is absent stands luck. Legitimate responsibility attribution should be based upon what is up to the person, what he decides and controls, and, most importantly, what reflects his agency.   Built upon the principle of individual difference, the control principle furthers the conditions of responsibility. Even when an agent is causally related to a consequence, and does make a difference to it, the control principle states that responsibility can be properly ascribed only if the agent has control over causing the consequence. If an agent could not have done otherwise in relation to the consequence or in some sense determined to cause it, the agent cannot be held responsible for it.   The control principle is closely related to the autonomy principle, which states that a person is not responsible for the harm caused by another agent, unless he has coerced that agent into taking an act. “No one can be responsible, in the properly ethical sense, for the conduct of another”35. As far as I know, this principle is not articulated systematically in philosophical literature. Not only because it is intuitively strong, also has it flowed consistently from the previous principles we've discussed: a given responsibility can only be properly assigned based upon what a person individually                                                  34 Nagel, “Moral Luck”, 58 35 Lewis, “Collective Responsibility”17 43  makes a difference to and what he has control over, or in other words upon individual agency, it is unjustified to blame another people for what this person does.   These principles may or may not be all applied in every instances concerning responsibility attribution. They are most found in ordinary intuition and daily practice about responsibility. People appeal to these soft individualistic principles consciously or unconsciously, explicitly or implicitly when understanding responsibility, however, they are a standard, regulative element of our ethical practices of allocating responsibility. To reject these principles would be rather controversial practice, for it would imply holding people responsible for what is not caused by them, for what is not under their control, or for what is done by other people. It would risk being “barbarous” as Lewis argues36.   3.1.2 Translation of collective responsibility  Soft individualism builds its skepticism toward collective responsibility based upon its commitment to these individualistic doctrines of moral subject, and argues that the proper subjects for responsibility attribution are those who can exercise sufficient individual agency upon the events. As I see it, this commitment leaves soft individualists with two possible ways to understand the concept of collective responsibility, depending on whether collective responsibility can be perfectly translated into individual responsibility. I shall further explain this statement below.                                                    36 See Lewis, “Collective Responsibility,” 3–6 44  When the responsibility resulting from collective actions can be fully reduced to the individual responsibility of its participating members, I call it perfect translation of collective responsibility. In the case of perfect translation, collective responsibility is no more than aggregation of the individual act of its participating members, for which each bears personal responsibility. Examples of perfect translation of collective responsibility can be the intentional harm inflicted by one group of people upon another. Charity, for instance, can be seen as a collective action performed by the group of donors, with the consequence of improving millions of receivers’ life condition. Each donor could have acted differently, to refrain himself from contributing to the overall charity, so that he would exclude himself from this collective action and therefore bear no moral relation to it. If each individual donor had not done his part in donating to charity, the goodness benefiting poor population would not have occurred. In other words, each individual donor had sufficient control over his decision of making the donation and thereby improving particular receiver’s life quality or not. In a case as such, roughly speaking, the collective action can be seen as the accumulation of the act performed by individual donor, and the collective consequence can be seen as the summation of the benefits every individual in need received. The consequential goodness of collective action can be fully reduced, therefore perfectly translated, into individual contribution of each participating member.   Here I borrow and substantially revise Derek Parfit's thought experiment on Harmless Torturers to illustrate the translation problem, though I do not share his consequentialist view on responsibility. In Reasons and Persons, Parfit invites his readers to compare the following two scenarios.  45   Scenario A: The Bad Old Days. A thousand torturers have a thousand victims. At the start of each day, each of the victims is already feeling mild pain. Each of the torturers turns a switch a thousand times on some instrument. Each turning of a switch affects some victim's pain in a way that is imperceptible. But, after each torturer has turned his switch a thousand times, he has inflicted severe pain on his victim.   To illustrate it simply, Torturer #1 turns on Switch #1*1000 times on Victim #1 Torturer #2 turns on Switch #2 *1000 times on Victim #2 … Torturer #1000 turns on Switch #1000 * 1000 times Victim #1000  Scenario B: Things have now changed. Each of the thousand torturers presses a button, thereby turning the switch once on each of the thousand instruments. Consequently, each torturer turns each of the a thousand switches once. The a thousand victims suffer the same severe pain. But none of the torturers makes any victim's pain perceptibly worse.   To put it simply, Torturer #1 turns on Switch #1 on Victim #1+ Switch #2 on Victim #2+…+Switch #1000 on Victim #1000; Torturer #2 turns on Switch #1 on Victim #1+ Switch #2 on Victim #2+…+Switch #1000 on Victim #1000; … 46  Torturer #1000 turns on Switch #1 on Victim #1+ Switch #2 on Victim #2+…+Switch #1000 on Victim #1000; In both cases, the translation of collective responsibility can be perfect. The first scenario is a rather clear case of perfect translation of collective responsibility. Each torturing group member is responsible for the severe pain received by each victim. Whereas the second scenario takes some analysis to see how the translation is done. Each torturer in scenario B turned the switch one time and inflicted imperceptible amount of pain that accumulated to severe pain that harms the victim together in the end, though there is no particular person can be identified as the victim for a specific torturer. Seemingly, no torturer can be said responsible for the severe pain on any victim, for his contribution to the suffering of each victim is mild enough to be ignored. Parfit invokes the “type-token” distinction to translate collective harm into individual responsibility. Or as he puts it, an act can be wrong because it is one of a set of acts that bring out the harm collectively37. Unlike in the scenario A where collective harm is the summation of the pain inflicted by individual token-torturer upon individual token-victim, the collective harm in scenario B is the summation of the pain caused by individual type-torturer to individual type-victim. In other words, each participating torturer does turn switch enough times to severely harm one hypothetical victim. By being a token member of the torturer type, each torturer, though without severely harming one particular actual victim, is responsible for harming one type-victim through a “perfectly translatable” collective action. Parfit seems to think that the translation between collective responsibilities can be sufficiently translated into individual ones once we stop ignoring the trivial and imperceptible harm caused by individual action.                                                  37 Kutz, Complicity, 130 47   However, with some revision, Parfit’s thought experiment can be used to demonstrate what I call imperfect translation. The imperfect translation of collective responsibility is when the collective responsibility cannot be fully reduced into individual responsibility. It happens either when there are residual elements of collective responsibility after reduced into individual responsibility, or there is a complete disconnection between collective and individual responsibility. Examples of imperfect translation of collective responsibility are common in daily moral practice. For instance, the collective action of causing global warming consists of countless individual contribution in producing Carbon dioxide in various ways, consciously or unconsciously, by participating members. None of the individual contribution alone would cause global warming. One individual could not have acted differently to prevent global warming. Individuals take cross-continental flights, drive a SUV car, or keep the light on when leaving the room. However, none of these individual acts, in an obvious sense, causes global warming. It is the quantitative accumulation of individual contributions that crosses the qualitative threshold of global warming and triggers the environmental catastrophe. In cases like this, the recognition of collective responsibility borne by individuals takes more than recognizing the trivial and small harmful effect of individual action. To understand the relation between individual and collective responsibility here needs further analysis and diagnosis of collective actions.    Let us revisit the example of Harmless Torturer with some revision, to make clear the imperfect translation of collective responsibility: Some factors stay unchanged. There are a thousand torturers. Each turning of the switch causes some imperceptible pain upon 48  an individual victim. We change one detail to better illustrate the imperfection of the translation. A victim can easily put up with 999 times of imperceptible pain, but the pain accumulates in a way that 1000 of them will permanently paralyze a person. If one torturer among the one thousand refrains himself from turning the switch and all the other torturers do not, the victim will just suffer from minor, almost imperceptible pain for 999 times. If none of the torturers decides against turning the switch, it will result in the paralysis of the victim. Now again, consider the following two scenarios:  Scenario A': The Bad Old Days. A thousand torturers have a thousand victims. At the start of each day, each of the victims is already feeling mild pain. Each of the torturers turns a switch one time on some instrument. Each turning of a switch affects some victim's pain in a way that is imperceptible. Each torturer turns the switch one time in a day. By the end of each day, one thousand victims each received some imperceptible pain once from one thousand torturers.  Again, the simplified version goes,  Torturer #1 turns on Switch #1*1time on Victim #1 Torturer #2 turns on Switch #2 *1 time on Victim #2 … Torturer #1000 turns on Switch #1000 * 1 time Victim #1000  Scenario B': In the Bad Old Days where things have now changed. There are one thousand torturers and one victim. Each of the one thousand torturers presses a button once a day, thereby turning the switch once on the instrument and bringing some imperceptible pain upon the victim. By 49  the end of the day, the victim suffers from 1000 imperceptible pains and gets paralyzed.   Torturer #1 turns on Switch #1*1time on Victim #1 Torturer #2 turns on Switch #2 *1 time on Victim #1 … Torturer #1000 turns on Switch #1000 * 1 time Victim #1 (his paralysis)  If we were to embrace the individualistic doctrine of responsibility and try to translate the collective harm in both Case A' and B' into individual responsibility, the responsibility we place on individual participants in Scenario A' and B' would have to be equivalent, because the actions they take are identical as far as responsibility is concerned: each of them has control over the one switch turning; each of them alone could have made difference to some imperceptible pain upon the victim, and each of them only knowingly and intentionally causes the imperceptible pain. What can be translated away from the resulted collective harm is the 1000 imperceptible pain, not the paralysis of the victim.   One possible proposal is to attribute the responsibility of paralyzing the victim to the last switch turner. The rationale is if everyone knows that 999 shocks causes no serious damage, and that it is only the last one that results in substantial harm, then everyone is exempt from causing harm up until the last switch-turn, and the last switch turns bear the responsibility. Thus scenario B’ still stands as a case of perfectly translatable collective responsibility, where responsibility will be fully reduced, though 50  not evenly distributed, among individuals. Note that this proposal adds an extra cognitive condition to original case, where every participant is in full awareness of the situation concerning how many switch-turnings bring what effect as well as how many switches have been turned. Suppose this additional knowledge is absent or not as specific as it can be, where participants have only a general idea of how the collective action is instituted, and each tries to do their individual part of the action with the intention of turning the switch once regardless of the ordering of doing it, soft individualism would still suffer from its inadequacy of translating mentioned earlier.    3.1.3 Three options to make sense of collective responsibility by soft individualism We may take a better look at soft individualists' position on collective responsibility given the translatability of collective responsibility. As I see it, soft individualists have three options to respond to collective responsibility: firstly, a soft individualist may directly deny the very possibility of imperfectly translatable collective responsibility. This soft individualism grants that collective entities are not fully reducible to its constituent individual members, whereas it insists on an individualistic account of collective moral responsibility, that is, “one which ascribes moral responsibility only to individual human beings as opposed to collective entities”38. On this view, “collective entities...have collective moral responsibility only in the sense that the individual human persons who constitute such entities have individual moral responsibility.”39 It denies cases of imperfect translation, and argues that in the most                                                  38 Miller, The Moral Foundation of Social Institutes, 17 39 Ibid 51  fundamental level all collective responsibility is reducible to individual members' responsibility. Nothing is to be lost in translation, since the responsibility resulted from collective action X can be exhaustively distributed as the individual responsibility that each participant bears for performing his part in collective action X. To understand collective responsibility is no more than to list its constituent individual members' responsibilities with full and complete details. I will show why efforts of this sort fail later in this chapter by providing three instances of imperfectly translatable collective responsibility.   For those soft individualists who commit to this individualistic doctrine of responsibility, while at the same time accepting the possibility of imperfect translation of collective responsibility, there are two more options they may take in explaining collective responsibility. One is to ascribe the harm to, basically, the misfortune embedded in the way things naturally are. In face of the imperfectly translatable collective responsibility, one has to see the “severe limitations on the power of the individual to modify social conditions”40, and recognize that the success or failure of collective action is largely influenced by factors out of one's control. It is rather possible that a group of individuals, each of whose action is morally impeccable, bring out some consequential harm when they act together. The wrongness, if there is any, lies in some emergent properties of the group action, such as the condition under which it is performed, the structure of the group and the procedure with which group decision is made, none of which cannot be properly attributed to any single individual. Not every harm has its correspondent responsibility taker. The harm resulted from an imperfectly                                                  40 H.D. Lewis, “Collective Responsibility,” 28 52  translatable collective action can be properly treated as tragic misfortune falls upon the victim.   We may revisit Case A' and B' in previous discussion in light of this line of reasoning. Soft individualists of this sort would argue that the individual participants in scenario A' and B' carries essentially identical responsibility, i.e. the responsibility of causing imperceptible pain. None of the individual participant should be held responsible for the paralysis of the victim in Scenario B', which though results from the accumulation of imperceptible pains. The way the torturing scenario is set up enables this accumulative effect, and can be best seen as the misfortune or unlucky position the victim is placed in.   The other possible way for soft individualism to deal with imperfect translation of collective responsibility is to postulate that a responsible collective consists of non-responsible parts, i.e. the irresponsible individual participants. David Cooper, following Karl Marx as he claims, holds such a view and argues that it is possible to hold collective X responsible for Y without thinking that each member of the collective bears the responsibility. It is rather essential, according to Cooper, to see that the collective blame that we ascribe cannot be analyzed in terms of individual blame. Some particular cases of blame can only attach to groups. “Different standards [of responsibility] may be applied to the behavior of individuals and to the behavior of groups...a group has fallen below some standard; and it is not necessarily the case that any individual has fallen below some standard, or, at any rate, below the same standard we are applying to the group.”41                                                  41 David Cooper, “Individual and collective responsibility”, 139; Frank Jackson holds similar view. See Jackson, F. (1987), “Group morality”, Metaphysics and morality: essays in honour of JJC Smart, pp91–110. 53  While committing to the individualistic doctrine of responsibility, this line of reasoning manages to explain the imperfect translation of collective responsibility by advocating a disconnection between collective responsibility and its individual members' responsibility. Again, to apply this approach to scenario A' and B', individual participants in both scenarios take equivalent responsibility, whereas the collective in scenario B' takes the responsibility for the paralysis of the victim and the collective in scenario A' does not.   To sum up, soft individualists have three ways to explain collective responsibility: 1) to accept the phenomena of imperfectly translatable collective responsibility and treat it like any other tragic events happening in nature; 2) to accept imperfectly translatable collective responsibility and treat it as something disconnected from individual responsibility, and therefore having no individual bearer; 3) to deny imperfect translation of collective responsibility and argue collective responsibility can be effectively translated into individual responsibility at all time, which can be properly dealt with based upon the individualistic doctrine of responsibility. I would argue that neither 1) nor 2) is a desirable answer that a satisfactory theory of responsibility would look for when making sense of collective responsibility. A more preferable theory of responsibility would be able to better equip us to deal with the issue. It is not good enough to either refuse to admit that there is something that can be done, collectively by all the participants together, to avoid the collective harm, or postulating some responsibility of a collective that has no effective taker and therefore no input whatsoever for its individual member. If a theory of responsibility different from soft individualism can make sense of collective responsibility better than these two approaches, I shall argue that we have some 54  justification to turn away from soft individualism and adopt this alternative theory. However, before I move on to this alternative theory of responsibility, I want to provide more evidence and robust cases that demonstrate imperfectly translatable collective responsibility in a more strong and more systematic fashion, so as to show that 3) is hardly a real option for soft individualists.   3.2  Cases of imperfectly translatable collective responsibility In this section, I will offer three often discussed types of collective actions that result in imperfectly translatable collective responsibility. As mentioned before, the common feature of these collective actions is that the resulting collective responsibility cannot be fully reduced to the individual responsibility of their participants according to the individualistic doctrine of responsibility. There are either residual elementary in collective responsibility or a complete disconnection between collective and individual responsibility for various reasons: the collective action is intended only by some, or perhaps none, of the participating members; or none of the participating member has exclusive control over the collective action; or little difference would have been made to the consequence even if a participating member had acted differently, etc. Three typical instance of imperfectly translatable collective action will be discussed in this section: Over-determination, collective inaction and discursive dilemmas. The aim is to block soft individualists from claiming in principle all collective responsibility is perfectly translatable at fundamental level.  55  3.2.1 Over-determination In many contexts, some outcome can only be achieved by collective efforts made by a group of collective members or a subset of it. A single participating member's effort does not make much, if any, difference to the outcome. The success or failure of this collective action lies largely, if not exclusively, upon the accumulated actions of other members of the group.   The title of Johnathan Glover's article nicely captures the issue in question, “it makes no difference whether I do it or not”42. Johnathan Glover offers a thought experiment case “The Desert March” in this article, which later is heavily discussed by Derek Parfit and Christopher Kutz in their own work on collective responsibility. I am presenting Kutz's version of it, for it is concise and sufficient for our discussion here.   “Two groups are traveling across a desert landscape. We are in Red Group, which consists of ten thousand people; Blue Group is a hundredth our size at one hundred people. The two groups are separated by a significant distance. Each member of both parties carries a quart of water, roughly the amount, it is universally acknowledged, and anyone needs to survive the rest of the journey. In fact, of course, a few drops less than a quart would also suffice, but not much less. Overnight, bandits creep into Blue Group’s camp and take all of their water. Blue Group radios us to request water from us, enough to sustain them for the trip, or roughly 25 gallons. Fortunately, we have an empty water tank mounted on a cart that is capable of holding that much, and so a call goes                                                  42  See Glover, J., & Scott-Taggart, M. J. (1975), “It Makes no Difference Whether or Not I Do It”, Proceedings of the Aristotelian Society, Supplementary Volumes, 49, 171–209 56  out for each member of Red Group to tip into the tank a very small amount of water, less than a third of an ounce. Though it is frightening, to be sure, to give up any water in the middle of the desert, the knowledgeable among us assure us that this small amount of water will not make the difference between our living and dying. Taken together, however, our contributions will make that difference to the members of Blue Group.”43 Let us assume that saving Group Blue is a desirable outcome, and given the circumstance, Group Red is in a position to make a collective relief effort. Failure to carry out this effort would be seen as an ethical failure of Group Red, one way or another. Now the question is how this collective responsibility of Group Red is translatable into individual group members in that group. Suppose Group Red fails to carry out this collective action, because some or all members of Group Red fail to give up water for Group Blue, and as a result some members in Group Blue die of dehydration. The following defense made by individual member of Group Red is deemed justifiable by soft agent-individualist's line of reasoning: given the size of the problem and population of both groups, the best each individual in Group Red can do makes very little difference. Giving up his a third of one ounce water is not sufficient to save any single individual in Group Blue. Provided that everyone else in Group Red does contribute their a third of an ounce water, not giving up his water is not necessarily going to leave any individual in Group Blue to die. Therefore, no single individual in Group Red has exclusive control over the survival of any individual in Group Blue. Holding any particular individual in Group Red responsible for the death in Group Blue is violating, firstly, the principle of                                                  43 Kutz, “Collective Work of Citizenship”, 474 57  individual difference because he could have made little difference to its occurrence; secondly, the control principle because he has trivial control over the death in Group Blue, and the occurrence of death could have not been prevented by him; thirdly and lastly, the autonomy principle, because this moral failure is made possible by the behaviors of other members in the group. Since no particular individual in Group Red can be held responsible for the collective failure, and since every individual member in Group Red is in equal contributory position to the collective action, no individual in Group Red can be held responsible for it. It can be concluded that soft individualism is left again with only two options here, as argued in the previous section, which are to either deny that Group Red, as a whole, is responsible for the failure of collective relief action, or to claim that Group Red, as a whole, is responsible for the failure, however, it does not give any reason for its individual members to act differently. What cannot be reasonably argued is that the collective responsibility of Group Red can be perfectly translated into individual group members' responsibility based upon the individualistic doctrine of responsibility. Facing cases of over determination, to ascribe collective responsibility to individuals, a soft-individualist would have to forge the principle of individual difference, and to admit that responsibility can be legitimately ascribed to an individual, even when his action could not have made any consequential difference or when this individual does not have exclusive control over the state of affairs.   3.2.2 Collective inaction A second type of collective responsibility results from collective inaction, where some action is reasonably expected from a group of individuals and each group member is in the position of performing that action; however, all group members refrain from 58  doing that because other members act alike. This sort of collective inaction is well documented in social psychology as an empirical phenomenon called bystander effects, also known as diffusion of responsibility.   Responsibility diffusion, mainly a theory in social psychology, explains the social phenomenon that individuals become be less likely to intervene in a situation if they witness it in the presence of other people than if they see it alone, because, according to Bibb Latané and John Darely, two social psychologists who first reported this, that bystander apathy is caused by a dilution of the individual's sense of responsibility in a larger group of people. This theory is further tested and confirmed by various social psychological experiments.   In their article “Bystander 'Apathy' ”44, Latané and Darley reported an experiment where subjects thought they were overhearing another student having an epileptic seizure. In some conditions, the students were told they were one of two subjects. In other conditions, they were told they were one of six subjects in the experiment. In the six-person condition, 31% of the subjects responded to calls for help. In the two-person condition, 85% of the subjects responded45.  Similarly, Latané and Rodin staged another experiment at Columbia University. They had students meet an experimenter, then divide up to fill out a questionnaire. Each student was assigned to a cubicle. Suddenly there was a crash and scream from one of the                                                  44 Latane, B. & Darley, J. M. (1969), “Bystander Apathy”, American Scientist, 57(2), 244–268 45 Darley, J. M. & Latané, B. (1968), “Bystander Intervention in Emergencies: Diffusion of responsibility”. Journal of Personality and Social Psychology, 8, 377–383 59  cubicles and a girl student's voice called out, "Oh My God, my foot! I can't move it! Oh....my ankle...I can't get this thing off me!" The "accident" was faked, but the subjects presumably thought it was real. When subjects were by themselves in a cubicle, 70% responded. When subjects were grouped in pairs, only 40% responded.46 It is consistent with the Darley and Latané hypothesis. Experiments as such in social psychology repeatedly show that the same effect occurs when people thought they were in a large group or actually were in a large group. An infamous example of this sort in real life is the murder of Kitty Genovese in NYC, which has been discussed in chapter two.   In cases of bystander apathy, some sort of moral failure can be intuitively identified. If we were to say there was a collective moral failure of the bystanders’ group, soft individualists would argue that it does not result from individual wrongdoing in any sense. The moral failure, if any, of the onlookers or bystanders in this case cannot be determined by their intention to participate in causing the harm in question; Moreover, had other bystanders in the group acted differently, the wrongness would not have occurred. The moral failure therefore does not lie in any particular individual's inaction. It lies in their collective inaction. Individual responsibility of a bystander, if there is any, can only be justified in a rather conditioned and limited sense: the moral failure of an individual bystander only occurs on the condition that other people act in the same indifferent and apathetic fashion.   In cases involving the bystander effect, soft individualism faces the challenge of imperfectly translatable collective responsibility. Soft individualism has to renounce the                                                  46 See Meyers, D. G. (2010). Social Psychology (10th Ed), New York: McGraw- Hill. 60  autonomy principle and concedes that we need to look into the interplay of individual intentions in particular situation when assigning responsibility of collective inactions. The intentions and action of others in a group, and how they condition and shape the intention and action of the individual in question should be taken into consideration when deciding the responsibility he or she bears. The solipsism embraced by autonomy principle does not serve soft individualism well in dealing with identifying collective responsibility in face of a group of bystanders.   3.2.3 Discursive dilemma  The third type of imperfectly translatable collective responsibility derives indirectly from a paradox in collective decisionmaking, which is first identified in jurisprudence and known as the “doctrinal paradox”. It happens when multiple parties in a court have to make a collective decision on the basis of received doctrine as to considerations that ought to determine the resolution of a case. Rather different decisions are arrived at, depending on what method of aggregating votes is adopted. Judges may make their individual decisions on the case and then aggregate their votes. Also, they can vote on whether relevant considerations are obtained and let those votes dictated how the case should be solved.   Let us consider a hypothetical example to better understand this paradox and the disconnection between the decisions of the group and of its individual members. Suppose that Bill Gates, Steve Jobs and Mark Zuckerberg are deciding on whether to nominate Aaron Schwartz as the head of Cyber-Liberty Association. Under relevant administrative doctrines, three members have to judge affirmatively that first, Schwartz is competent as 61  the association head; second, that Schwartz would accept the position once offered, and third, that an association head is necessary. In case of disagreements among members, the majority of two will carry the vote. Now imagine that Bill Gates, Steve Jobs and Mark Zuckerberg vote as follows on those issues and on the doctrinally related matter of whether to nominate Schwartz as the association head.   Competence Acceptance Necessity Individually Zurkerberg YES NO YES NO Gates YES YES NO NO Jobs NO YES YES NO Group YES YES YES ?? TABLE 1  Suppose each person decides on each issue in a perfectly rational manner, now there are two ways the group might make its decision in a case like this. The group might have each individual decide on the conclusion based upon their private judgment on the particular issues, which would leads to two of the three persons respectively deciding against nominating Schwartz for different yet exclusively determining reason, and therefore reach the collective decision of rejecting the nomination. The other way to collect individual judgments is to aggregate the decisions on each issue. As a consequence of this process, the majority view would be that Schwartz is competent for the position, that Schwartz is likely to accept it once offered the job, and that it is 62  necessary for the association to have a head. Aggregation of majority view on individual issue would lead to nominating Schwartz with certainty.   This paradox in collective decisionmaking is introduced to philosophers by Philip Pettit as the discursive dilemma. As Pettit points out, a discursive dilemma will arise “whenever a group of people discourse together with a view to forming an opinion on a certain matter that rationally connects, by the lights of all concerned, with other issues.”47, which often happens in public bodies, corporations, governments, various committees and boards. It will always be possible that collective decisions will lead to different or even opposite directions, depending on what aggregative procedures are adopted by the groups, as shown here. It can easily be the case that a group will take a position and act upon it, whereas no individual members of the group endorse the position. The divergence between collective decision and individual members' decision is convincingly demonstrated by the example of discursive dilemma as I see it.   In the case of nominating Schwartz, it seems the group and its members are facing a choice between two options: on the one hand, they may endorse the collective decision “No”, and consequently maximize the responsiveness of the collective decision to individual decisions, meaning the final collective decision highly correlates with individual decisions. However, as a group, they have to pay the cost of collectively endorsing an inconsistent set of propositions, meaning that the group says “yes” to every premise yet does not come to the rational and consistent conclusion. On the other hand,                                                  47 Pettit, P. (2011). “Groups with minds of their own” Social Epistemology: Essential Readings, Oxford University Press, 245 63  the group may preserve the rationality and consistency at the collective level while running the risk of collectively agreeing upon a conclusion which the majority of members would individually reject.   According to Pettit, the existence of discursive dilemmas suggests that “there is a type of organization found in certain collectives that makes them into subjects in their own right, giving them a way of being minded that is starkly discontinuous with the mentality of their members. This claim in social ontology is strong enough to group talk of such collectivities as entities that are psychologically autonomous and that constitute institutional persons.”   A further implication of Pettit's theory of discursive dilemma in socio-political deliberation and group decision making will be addressed in upcoming chapters. For the purpose of this section, we need to see where this leaves soft individualism given Pettit's demonstration that a collective can be a subject with its own mind. I see the challenge posed upon soft individualism by discursive dilemma more severe than by other two cases of imperfectly translatable collective responsibility. Collective entities, as Pettit argues, forms themselves around a group purpose or goal, reasons with consistency and rationality at the collective level, and reaches a conclusion which the group commits to and acts upon. In many cases, the group commitment “we ought to do X” is not necessarily held by individual members. When the size of a group increases and its decision procedure becomes more complicated, the problem for soft individualism gets more severe as members have little control over or play little causal role in determining the collective’s actions or their consequences. The more fundamental challenge comes 64  from the dualism between the intentions individuals have as members of a group while motivated by this group commitment and the intention individuals have independently of their membership of the group. It seems there are two sets of mutually independent intentions held by individual members, and therefore, perhaps two kinds of agency exercised by individual members. The translation from collective responsibility to individual responsibility could be impossible, rather than merely imperfect.   To sum up this section, my goal is to present some substantial cases of imperfectly translatable collective responsibility, in order to falsify the view held by some soft individualists that perfect translation from collective responsibility to individual responsibility is always possible based upon the individualistic doctrine of responsibility. With these cases of collective actions and responsibility in mind, I hope the case has been convincingly made that the individualistic understanding of responsibility will not always succeed assist us in effectively and appropriately distributing collective responsibility among participating members. Confusion, conflicting opinions and paradoxes are inevitable due to the inadequacy of the individualistic approach. In the next chapter, I want to offer some consideration of what kind of understanding of responsibility can better equip us in dealing with collective action. The goal is not as ambitious as to systematically provide normative standards for collective responsibility. Rather, I want to propose some features this alternative understanding of responsibility should possess, and possibly how to reach that understanding.    65  Chapter 4: An Alternative View of Collective Responsibility The earlier chapters discuss two main issues: some actions are performed by collective agents, for collective goals or in collective contexts; and there are considerable difficulties in translating responsibility from collective action into the responsibility of individual members based on individualistic principles. I shall briefly revisit these two issues before setting forth an alternative view on collective responsibility, with the hope that we shall come to appreciate the desirability and necessity of an alternative view when presented with the gravity of this challenge put forward by the combination of these two issues. Following that, I discuss the alternative view of collective responsibility. The plan is not as ambitious as to present a completed picture of this alternative understanding of collective responsibility. Instead, I attempt to sketch out two necessary features such alternative account of collective responsibility must have: sensitivity to multiple contexts and dependence on respondents. Lastly, I aim to explain why the issue of collective responsibility, given this alternative understanding, can be treated as a political problem and may be approached with a political solution.   4.1 Revisiting collective action and inadequate individualism Chapter two discusses various types of collective action: some are performed by collective agents; other of these actions foreground the interconnectedness of participants' intentions and still other actions stem from collective contexts. Briefly explained, collective action refers to the phenomenon of people doing things together. The togetherness cannot be meaningfully explained by analysis, however detailed, of the actions performed by individuals in this collective doing because collective action is 66  either performed with group identity, or results from complicated and entangled interpersonal webs of intentions and action, or is made possible in some collective context. Investigations restricted in focus to the mental states and bodily movements of individuals in the group fail to reveal the interpersonal character of collective action. One person cannot tango; the best he can do is to move around in certain patterns conventionally associated with the tango next to his tango partner. Likewise, one person cannot engage in a quarrel, win a soccer match, or cause global warming. We fail to grasp the significance of collective action when we divide it into discrete individual actions. Moreover, some individual actions become incomprehensible and even absurd when not understood in relation to collective action. Eleven adults chasing and kicking one ball, for instance, is nonsensical and comic unless they are participating in a football game. To gain a better understanding of collective action, one needs to zoom out from what happens at the individual level, and bring back the collective contexts, and take the interaction and interdependence among individual participants into consideration. Individual actions are often performed as parts of a collective action to achieve certain collective goals. In turn, collective goals and contexts give individual actions the significance which would not otherwise have been possible.   If one fails to see the collective nature of actions, there will be at least three aspects to which one would fail to attend when explaining human social activity. On a basic level, one would fail to appreciate the aggregate magnitude of individual actions when not seeing them collectively. Environmental crises, for example, result from an aggregation of individual actions. It might be problematic to describe one person dumping a bag of garbage into a running river as having a destructive impact on local 67  environments. But it quickly becomes obvious pollution when a group of people all do the same thing. The significant consequence of this collective action, namely polluting the river together, motivates a different assessment of what each individual is doing in this case. Following this line of reason, the dumping of the contents of a garbage bag into the river can be properly seen as participating in harming the local environment. Although the individualistic doctrine of responsibility would have little difficulty in explaining this collective action as an aggregated outcome of individual actions, to situate individual actions with seemingly trivial consequences against the magnitude of their accumulated harm makes more salient the morally problematic side of the action.   Secondly, one would have a hard time explaining actions at an individual level without knowing the collective goals behind them, or as I call it the sense-making function of a collective goal. Sharing a plan or a common goal among a group of people teleologically explains the actions of the constitutive members in that group. “We want to win this soccer game” gives rise to “I want to kick this ball far” in the context of soccer playing. The latter would have a rather different sense, or even make no sense at all, without the former. Moreover, individual doings in collective action depend upon the intentions and actions of other participating parties. “I am passing the ball to my teammate” is caused by “you are blocking my way”. The dynamic interplays among intentions and actions of individuals in collective action are central to explaining specific instances of individual actions. Provided that individual actions are highly dependent upon the collective goals and contexts, the explanation for individual action cannot be completed by describing solely what happens at the individual level.  68  Lastly, the collective context enables individuals to perform actions which would not otherwise have been possible. In other words, the collective context has an empowering effect on individuals. Actions that are impossible for any single individual to take can be performed by the collective, of which the individual is a member. Ccollective contexts expand the range of possible actions which an individual can perform as a member or representative of a group. For instance, in the role of German President, Johannes Rau could properly issue an apology for the Nazi Holocaust. Moreover, in the collective context, one's individual action can bring out more than what the action by itself stands for. Wearing headscarves in Turkish public schools is not just a style of dress. It could be an act of anti-secularism or a way of expressing religious identity, which invites further interpretation. Actions as such need to be understood in relation to their collective contexts, and hence collective contexts need to be included in our description and understanding of these individual actions. By intentionally engaging with collective contexts effectively, individual actions acquire additional significance in certain collective contexts.   The central mistake of reductive individualism is to regard collective action as essentially no more than an aggregated outcome resulting from the confluence of individual intentions and actions. According to reductive individualism, aspects of the individual level adequately describe both individual and collective kinds of actions, since what happens at the collective level is fully reducible to individual intentions and actions. As a result, actions are characterized by mental content in individual minds or by the consequences of individual behaviors. What is missing in this reductive individualistic account is the interpersonal relations between the individual intentions and actions, 69  which, as I argued previously, are constitutive of collective action. Therefore reductive individualism inevitably falls short in offering a satisfactory account of collective action.   In chapter three, I aimed to point out that this failure is further amplified when soft individualism argues that a subject can be held responsible solely for individual action freely willed by him. Based upon the same rationale, soft individualists argue that only individuals, and never collectives, are proper bearers of responsibility, because the beliefs and desires of individual human beings are considered to be the ultimate explanations for and causes of actions, whether by individuals or collectives. In order to determine responsibility for a certain action, collective action included, we need to investigate only the facts about individuals involved: what the individual intends to do; whether he is coerced or manipulated; what this individual could and could not have done; whether he has adequate mental capacity for deliberation and reflection; etc. Responsibility is considered to be an intrinsic and inherent feature of the agent and his actions. We need not and must not concern ourselves with other factors, such as in what context the action is performed, how different people react to this action, or what is expected of the individual who takes the action. For these are the factors neither intended by the individual nor completely under his control. An individualistic understanding of action and responsibility tends to be univocal, uncontested, ahistorical, and one-dimensional due to the exclusion of these factors.    Skepticism towards the individualistic notion of responsibility has been raised by various philosophers. These skepticisms do not derive from the skepticism towards free will, which is what individualistic approach traditionally concerns itself with. For 70  instance, Heinrich Gomperz points out that “the statement, therefore, that men can be held responsible solely for individual conduct freely willed is certainly wrong; it mistakes a principal characteristic of individualistic ages for an eternal law of human nature”48. Some philosophers have started to challenge the commonly held view that choice or voluntary control is a precondition of moral responsibility, and argue that what matters in determining a person’s responsibility is whether the action can be seen as indicative or expressive of her judgments, values, or normative commitments 49 . Contemporary skeptics also concern themselves with some necessary conditions for the practice of individualistic moral responsibility, and challenge whether these conditions can be realized. For instance, in discussion of the cognitive conditions of moral responsibility, George Sher raises his skepticism towards this notion of responsibility and says,  “If agents are responsible only for acts they have knowingly and willingly chosen to perform, only for omissions in which they have knowingly and willingly chosen to produce, then no one is responsible for any act, omission, or outcome whose moral or prudential defects can be traced to his lack of imagination, his lapses of attention, his poor judgment, or his lack of insight.”50  Tamler Sommers, as a meta-skepticist of moral responsibility, argues in his book Relative Justice, that in actuality, the understanding of responsibility has proven to be a                                                  48 Gomperz, Individual, Collective and Social Responsibility, 243 49 See Robert Adams (1985), “Involuntary sins”. The Philosophical Review, 93(1), 3–31(1985); Scanlon T. M. (1998), What we owe to each other. Cambridge, MA: Harvard University Press; Smith, A. (2005). “Responsibility for attitudes: activity and passivity in mental life”, Ethics, 115(2), 236–271 50 Sher, Who Knew, 7 71  phenomenon that varies in different societies and cultural tradition. Differently situated human beings have regarded moral responsibility differently across cultures and throughout history. The variant understanding of responsibility cannot be explained away as the products of “irrationality, superstition, conceptual ambiguity or ignorance”51. It is more likely that we would never reach agreement about the criteria of moral responsibility, “even under the ideal condition of rationality”52. “[T]here is no set of conditions for moral responsibility that applies universally and therefore that no theory of moral responsibility is objectively correct.”53  Moreover, philosophers have recently started to question the commonly held view that choice or voluntary control is a precondition of moral responsibility, and argue that what matters in determining a person’s responsibility is whether the action can be seen as indicative or expressive of her judgments, values, or normative commitments54.   Following these philosophers, I venture to argue for a revisionary view that responsibility for collective action indicates and expresses the shared understanding, judgments and values among involved participants. Therefore, to investigate on the conception of responsibility, one needs to attend factors which go beyond the facts of individuals in separation. The failure to attend to relevant, yet non-individualistic determining factors makes it difficult for soft individualists to deal with problems such as                                                  51Tamler Sommers, Relative Justice, 4 52Ibid, 5 53Ibid. 54 See Robert Adams (1985), “Involuntary sins”. The Philosophical Review, 93(1), 3–31(1985); T. M. Scanlon (1998) What we owe to each other. Cambridge, MA: Harvard University Press; Angela Smith (2005). “Responsibility for attitudes: activity and passivity in mental life”, Ethics, 115(2), 236–271  72  collective responsibility evasion, for some collective harm can be left unaccounted by any individual member’s responsibility when we attribute responsibility based upon an individualistic doctrine of responsibility. In Chapter 3, I discussed three instances, over-determination, interdependence of onlookers’ behaviors and discursive dilemma, in all of which individual and collective responsibilities are clearly disconnected, and therefore one cannot be fully translated into the other based on the individualistic doctrine.   4.2  Plural understanding of collective action and responsibility Based on my revisionary attempt, I propose that the problem of attributing responsibility for collective action motivates us to consider the plural perspectives we take when perceiving action in collective contexts. I intend to propose that an action in a collective context is subject to plural descriptions when relevant factors are taken into consideration. Reductive individualistic descriptions of an action presents one kind of description, though not the kind relevant to collective responsibility. For instance, voluntarily shooting someone with the intent to kill can be classified a murder, an assassination, an act of self-defense, or an execution depending upon the contexts in which this action is performed, the role of the person who performs it and by whom this act is perceived. Reductive individualism would have a hard time in distinguishing different descriptions of an action as such, and fail to attribute responsibility of proper degrees and kinds. When it comes to collective action, those actions performed either by a collective agent, or for a collective goal, or in collective contexts, the description becomes even further complicated for there are more contextual and relational factors that come into play. Responsibility for collective action thus becomes contested, 73  uncertain, plural, and dynamic. Judgments about whether a person is responsible and for what he is responsible are shaped by the interplay of competing descriptions and assessments of the action.  A well-developed proposal concerning how to perceive a collective action with pluralistic lens is beyond the scope of this dissertation and should be left for my future research. For the time being, I will constrain myself and argue for two features which I take to be essential for an adequate account of collective responsibility: context-sensitivity and respondent-dependence.    4.2.1 Context-sensitivity By context-sensitivity, I mean the necessity of including situational factors where an act is performed, such as social norms, commonly accepted behavioral rules, and standard practices in a given environment. It is often the context which makes it possible for people doing things together to be understood as collective action of some sort, which acquires different meanings and interpretations, despite the fact that body movements and behavior manifestations are identical. Therefore the determination of what ought to be done is often context-sensitive.   For example, joking about the dead when delivering memorial speech at a funeral can be deemed offensive or humorous, and hence either blameworthy or approvable, depending on the local culture and social norms, even when the intention behind it is to lighten up hearts. In some social context, joking about the dead at a funeral can be accepted as a way of publicly expressing remorse over the death, and thereby be 74  agreeable to the context. In other context, possibly it is considered disrespectful and offensive, and thereby disagreeable. In latter situation, an agent is responsible to issue an apology, or at least to make some explanation of his behavior, even when he has the best of intention.   4.2.2 Respondent-dependence The idea that responsibility needs to be respondent-dependent is one I borrowed from Kutz in his book Complicity, where he argues that “the agent’s relations to others… do enter into the determination that an act was wrongful and the description of the wrong.”55 The responses that agents warrant for their actions depend upon “both prior moral and social relations among the parties, and the particular perspective of the respondent”56. In other words, the responses to an agent by others give the concept of responsibility its distinctive shape and content.   Following his idea, I take it that the understanding and assessment of a collective action, and thereby the determination of related responsibility, varies from person to person depending on the perspective they take as respondents. As Kutz puts it, the nature of it is “fundamentally relational”57.   Kutz uses an example in his book to demonstrate this point, and I shall briefly introduce it here. Suppose I carelessly break a neighbor’s vase at a party while dancing on his grand piano. Depending on different respondents, I am placed under plural                                                  55 Kutz, Complicity, 19 56 Ibid, 20 57 Ibid, 21 75  responsibilities. My neighbor is warranted in resenting my carelessness and asking for an apology; other guests at the party may expect a public display of contrition since I have disrupted their pleasant night with my loutish behavior, though they cannot appropriately feel personally aggrieved in the same way as my neighbor. Perhaps among the guests, the friend who had given this vase to my neighbor is warranted in responding differently from other guests at the party.58  This case shows that what a person is responsible for is closely related to whom this person is responsible to. An agent is susceptible to multiple and varied responses from different people because of his action. The content of responsibility is partly determined by the existing relation between the agents and other respondents. The particular role the agent plays and the position he occupies in a group serves as a major factor in deciding his responsibility. The idea of responsibility reflects what can be reasonably expected from the agent by others and by himself in interpersonal interactions. Given that different respondents are in different relations to the agent, the practice of respondents attributing responsibility and the agent taking responsibility reveals and expresses mutually reasonable expectations and understandings of the very nature of their relationship.  4.3  The function of collective responsibility  If the foregoing is correct, an adequate account of collective responsibility ought to be sensitive to the contextual and relational factors that motivate and shape the action,                                                  58 Ibid, p21 76  so that we are equipped with better descriptions of collective action. Responsibility must then be attributed accordingly. Factors often neglected by reductive individualistic approaches are well recognized by this revisionary view of responsibility and acquire normative significance, including considerations such as what individuals do as members of a collective, how their action is understood within a particular context, what relation they are in with one another, and what they could have done collectively, etc.   If this contextual and relational view of responsibility is correct, then the question “does the collectivity have the same moral status as a person?” is less relevant than we initially might have thought. The main task here is to find an account of collective responsibility that reflects the nature of the relationship among all affected parties in collective action. An adequate, well-functioning, effective system of responsibility in the collective context would convey to individual parties the knowledge which is necessary for them to be competent responsibility bearers, represents these interpersonal relationships in a moral community, and motivates individuals to recognize collective responsibility and to respond to it properly. Moreover, it would enable individuals to take up collective perspectives and to form new collective identities. The consideration of collective responsibility “poses deliberative problems for participants, forcing them to think through the significance of their relationship to the collective structure, and to act on the basis of that understanding.”59 An established system of collective responsibility guides individuals to scrutinize their understanding of the relationships in moral communities, regulates their behaviors in morally satisfactory ways relative to one                                                  59 Kutz, p202 77  another, and offers guidelines for individuals to orient themselves in complex webs of group interaction and social living community.  On this view, in order to be competent bearers of responsibility we need to accept that we can act wrongly even when we do not intend any harm, when we do not have much control over the state of affairs, or even when we are not aware of any sense of togetherness. Because our actions reach beyond our intentions and because their effects may overlap in damaging ways with those caused by other agents about whom we neither know nor care, we are always already complicit with others, and incur responsibilities we might not have imagined. As Kutz argued, “whether or not we are affected causally by others, or whether we are in a position to make judgments about their [others'] characters, we are nearly always in the process of evaluating others' attitude towards us”60. Being a competent responsibility bearer means that it is not enough for individual actors to have mental sanity in order to be able to make voluntary decisions or to be freed from all sorts of coercion. A competent responsibility bearer in the collective context needs, firstly, to be aware of his position in a group and of his role in collective action. He needs to be able to take seriously the perspective of other group members, and to perceive others' expectations upon a person in his position. Moreover, he needs to be able to deliberate over the action and perform his part based upon his understanding of the action and its context. It follows that he should be able to respond properly with emotion and action depending upon his respondents. Finally, he needs to commit himself, to some degree, to the collective actions in which he participates, and to be open to reassessing his participation and responsibility when new contextual factors are present.                                                  60Kutz, Complicity, 26 78   This view of collective responsibility has its roots in contemporary theories of moral responsibility. Peter Strawson holds an expressivist account of responsibility, according to which the practice of attributing responsibility reveals the natural patterns in our emotions and in our conduct toward one another. Our reaction manifests in our positive or negative reactive attitudes toward one another, such as resentment and gratitude. Although we modify our reactive attitudes if we discover that, for instance, the harm is an effect of another person’s mental disorder or delusion, our expression of reactive attitudes, which is basically our practice of responsibility attribution, is not explained merely by the agent’s will and conduct, but by the nature of our relationship with them. Kutz supports this reading of Strawson as he makes the following claims,   “The crucial point, not made explicitly by Strawson, is that our disinclination to express reactive attitudes to partly or wholly non-responsible agents is explained not merely by the quality of their wills, but by the nature of our relationship with them…Strawson emphasizes how our awareness of cognitive and affective limitations in non-responsible agents naturally precludes them from participating in the relationships characteristics of adult society. I [Kutz] would add to this description that our attribution of an incompetent will to such agents is just the projection of ourselves into a certain type of relationship with them…it should now be clear that the attitudes and expression of agents only warrant response given a certain understanding of the nature of relationship between agent and respondent. In Strawson’s very rough terms, the relationship must be either participatory or potentially 79  participatory. The agent to whom we respond must be someone with whom we will and could cooperate in social life…So the responses characteristic of accountability are warranted by the point and demands of relationship.”61  Strawson’s account is particularly helpful in explaining collective responsibility. The practice of taking and attributing responsibility, according to Strawson's view, takes place among moral agents who interact with one another in a moral community, rather than by socially isolated individuals. By assigning responsibility to himself or to others, an individual expresses his normative judgments relative to the moral and social norms accepted in the group or community.   Based on this approach, one can reasonably assume certain conditions that need to be in place for proper and effective responsibility attribution in a community. First, in the community ideally there is a basic framework of moral and social norms concerning responsibility, which is potentially agreeable to all community members. Social interaction and political institution should be arranged for community members to learn about and contribute to the discussion of underlying beliefs and values supporting this framework. Moreover, community members should be able to apply principles of responsibility attribution to particular events and actions. As a group, they need to collect information, acquire understanding, and make judgments concerning who is responsible and for what. Lastly, community members need to cultivate their competence in engaging                                                  61 Kutz, Complicity, 29. Kutz discussed Scanlon’s characterization of Strawson’s theory as “Quality of Will” theory, which implies that Strawson’s understanding of accountability rests exclusively upon the internal will, rather than on the external affiliations, of moral agents. Kutz thinks Scanlon is incorrect with this reading. I intend to agree with Kutz.  80  in the practice of assigning responsibility. A competent responsibility attributor needs to learn to take the perspective of the group as a whole, to communicate effectively with other members, and to be open and adaptive to the judgments and decisions of these members.  In the last section of this chapter, with all the previous ideas and arguments in place, I will explain what collective responsibility, a moral issue, implies for political practice, and why it may require a political solution.   4.4  Being responsible and holding responsible  The relational and contextual account of collective responsibility, which depends upon individual contributions, whether intentional or not, to the collective action, needs to come to grips with an often raised concern. This approach to collective responsibility may fail to distinguish the fair conditions of holding individuals responsible for their contributions to collective action and the moral nature of an individual being responsible. In her essay “On being responsible and holding responsible”, Angela Smith discusses this distinction, though she does not particularly concern herself with the issue of collective responsibility in that piece. In this section, I shall use Smith’s work as a point of reference, building some possible criticism towards my own proposal of collective responsibility. I hope to show that though Smith is correct in making this important distinction, it should not be read as an objection or criticism toward contextual and relational account of collective responsibility. Rather, when combined with proper political institutions and social practices, this contextual and relational account of 81  collective responsibility can possibly abridge the alleged gap between holding responsible and responsibility agency   Smith points out that some theories of responsibility tie attribution of responsibility too closely to the practice of “holding” people responsible. Philosophers at fault include R. Jay Wallace and Christine Korsgaard, who, according to Smith, hold the view that conditions of responsibility are to be construed as conditions that make it fair to adopt the stance of holding people responsible. These accounts of responsibility start with our judgment of who to hold responsible, rather than “that of the agent who is judged”. Following Strawson’s approach, these theories make the attribution of responsibility a dependent upon whether and when it would be fair or appropriate for a judge to react to the agent in either positive or negative ways. However, Smith thinks approaches of this kind are highly problematic, for our considerations regards to whether and when to hold an agent responsible are sensitive to a lot of things which have little or nothing to do with the agent’s responsibility. The question “under which conditions a person can be said to be responsible for something?” and the question “under which conditions it would be fair to hold a person responsible for that thing?” are different and to a certain extent independent questions. For the latter question concerns many factors such as the standings people hold in a community and the significance of the fault being conducted, and these factors are relevant to the holding responsible, but not to the determination of being responsible or not.   The account of collective responsibility I advocate is vulnerable to this criticism. If my argument is sound, then the understanding of responsibility relies on figuring out 82  individual contributions to collective action given the relation one has with others in a community and the context in which the action was conducted. As Smith points out, this way of investigating the question of responsibility, i.e. to start the investigation by asking what to consider in order to properly hold a person responsible, already deems the person responsible in relevant sense. It does not address the basic conditions of the agent being responsible. In the case of collective responsibility, Smith would be correct in pointing out that my approach presupposes involving parties in a collective action being potentially responsible one way or another in certain sense. Based upon that presupposition, collective responsibility is further determined by the relations an agent bears and the situation he is in. In other words, contextual factors might determine whether to hold a person responsible for a collective action. However, these factors have little to say if the person is responsible or not. In the case of collective action, the gap can appear even larger: whether one can be held responsible based upon his contribution to certain collective actions is intuitively different from whether one is responsible for the action.   Discussion of a familiar case will help vividly to elucidate Smith’s concerns. Imagine we are conducting a conversation with some character such as Adolf Eichmann concerning his responsibility for partaking in a collective action, which is holocaust in this case. He was held responsible for his crime against humanity as a major perpetrator of the genocide, given his contribution in organizing and managing the logistics of mass deportation of, the political position he was holding in Nazi Germany, and the historical and social context he was situated. However, Eichmann would not held himself responsible in the same way. His understanding of his contribution to the holocaust was 83  that he was just “following orders” and “doing his duty” as a patriotic citizen and a good soldier. Eichmann might argue that what others consider as conditions of holding him responsible for what Nazi Germany as a collective did has nothing to do with the conditions of him being responsible, given that he simply did not share these considerations. Eichmann might be an extreme case to illustrate the concern, but I hope basic idea of the criticism is clear: the worry is that we would fail to understand the real nature of someone being responsible for certain action if we merely focus upon the appropriateness of reactive attitudes or the judgments of responsibility we hold to that person.   The gap between being a responsible agent and holding responsible is, however, not as wide as one might think. As I mentioned earlier in this chapter, the practice of responsibility takes place among moral agents who interact with one another in a moral community, rather than by socially isolated individuals. By engaging in the practice of holding other responsible and taking responsibility for oneself, an individual expresses normative understanding and judgments relative to the moral and social norms accepted in the community. To be a responsible agent, one needs to understand that other members of the community stand ready to assess him against the shared moral expectation and normative norms among them. He needs to be able to assess others in the same way to understand why and what others in the community express through their practice of responsibility attribution, either with reactive attitudes or active blaming. What makes an agent a responsible agent, as far as I see it, is his competence and capability of participating in the practice of holding responsible. To be a responsible agent is, firstly, to understand appropriate moral rules and norms concerning responsibility in a community, 84  and secondly to stand ready to hold people, others or oneself, responsible, and lastly, to further adjust his conduct accordingly. To be a responsible agent is to able to consider the points of view of a responsibility attributor and to regulate his own conduct based upon these considerations.   From this relational and contextual view of collective responsibility it is admittedly true that a morally responsible agent needs always to act with the knowledge that his contribution to the collective action is a potential object of responsibility attribution by his fellow members of the moral community. He needs to be aware that it is always possible that his contribution bears more consequences and meaning than what he, as an individual alone, can know or even predict. Therefore, as a responsible agent, he needs to stand ready to engage with the practice of, constantly and ceaselessly, especially when it comes to individual contribution for collective action.   Let us bring back the case of Eichmann. What makes Eichmann a responsible agent, I would argue, is not some universally applicable, objective, static conditions of being responsible. What makes him a responsible agent is the fact that he is capable of having some kind of value and belief system, and engaging in the practice of responsibility with others in the moral community. In other words, although it is important whether the person who is held responsible recognizes the objective fact about his responsibility for certain action (if there is such an objective fact), or shares the conditions of appropriately responsible with others who hold him responsible, what qualifies him as a responsible agent is his capability of participating the normative practice of responsibility attribution in a moral community. The capacity to determine 85  appropriate moral responsibility to oneself or to others signifies the presence of a responsible agent.   Ultimately, then, the competence and appropriateness of one’s participating moral practice of responsibility does closely tie to the person’s moral agency. Furthermore, to improve a person’s competence and capability of holding responsible in a sense enhances moral agency. To bring the discussion back to the thesis of this dissertation, social structure and political institution makes significant difference in cultivating people’s competence of holding responsible, particularly in the context of collective action, and thereby in enhancing moral agency of the members of a collective. Hence, the moral concern of collective responsibility leads us to favor some political institution over the other.  4.5  Collective responsibility being political  As argued above, an individual’s responsibility for the actions of collectives of which she is a member cannot be determined solely by the investigation of facts about that person. Our practice of responsibility attribution is explained not only by the will or conduct of individual actors, but by the nature of our relationship with them. The understanding of what an agent has done is itself given by our social contexts and relations, and any notion of responsibility must itself be interpreted contextually and relationally. The practice of making normative judgments and attributing collective responsibility both relies upon and contributes to a unified, dynamic system of social life. A system of collective responsibility cannot be constructed without social interaction and 86  group decision-making, and the attribution of collective responsibility thus becomes a public project upheld by the group of individuals involved.   To understand collective action and to determine related responsibility is an ongoing process, where new contextual particulars and respondents’ perspectives should be constantly brought into consideration, shaping and reshaping the results. Collective action generally happens in complicated contexts, with various numbers of participants and affects many people in different, sometimes unpredictable ways. Therefore, the determination of responsibility attributable to particular individuals for participating in collective action requires an effective determination mechanism, which can include sufficient contextual and relational factors of the action in question and consolidates diverse views on the attribution of responsibility. This is why I argue that this alternative view of collective responsibility can and should guide our social and political practice.   As I shall discuss in detail in my next chapter, different group decision-making mechanisms generate different pictures of collective action and responsibility as a consequence. I shall argue that those collective decision-making mechanisms which cultivate well-functioning collective responsibility attribution are normatively more justified than those that do not. In other words, depending upon the different ways collectives structure and organize their actions and decisions, they possess varying capacities for establishing effective responsibility attribution systems, and, as a result, for sustaining social relations and promoting the public good of the collective. The more effectively a collective decision-making apparatus cultivates its well-functioning responsibility attribution, the more normatively desirable it is.  87   This makes the issue of collective responsibility into a political project, which requires our moral, social and legal institutions of responsibility to be organized based on this account of collective responsibility so as to stabilize and foster our communities of concerns and interest. The establishment of a fine account of collective responsibility motivates social, political and legal institutions to be made to respond to, and to reflect this concern.   Given what is required for a community to have a well-functioning system of responsibility attribution, the preferable decision-making mechanism which generates collective action and responsibility has to perform specific functions. Firstly, group decisions ought not to be made by isolated participants. Group decisions concerning collective actions and responsibility have to be made in a situation where participants converse with one another, and exchange information about issues of common concern. This kind of context allows collective decisions to be formed as more than a mere aggregation of individual preferences and judgments, but instead as decisions resulting from rich contexts, diverse perspectives and dynamic group rationality. I will further develop this point in last chapter by comparing aggregation and deliberation as group decision making mechanism.   Secondly, this decision-making mechanism ought to cultivate participants into becoming competent responsibility attributors as well as takers by enabling them to jointly and publicly share their understanding of responsibility through communication and interaction. Communication and interaction among individuals reveals their mutual 88  expectations, their relations and positions in relation to one another, as well as their interdependence as collective participants.   Thirdly, this mechanism needs to transform the individual participants into members of a collective when they partake in collective action. It should invite individuals to take on an identity as members of the collective and adopt preferences and perspectives which they should not otherwise be expected to hold.  The project of establishing a moral discourse concerning collective responsibility also has a practical goal at an individual level. With the established account of collective responsibility, one may see the collective responsibility as inputs to his practical and ethical reasoning. The consideration of collective responsibility “poses deliberative problems for participants, forcing them to think through the significance of their relationship to the collective structure, and to act on the basis of that understanding”. [source?] This assists individuals to deliberate better over their individual part of collective action, to recognize better their related responsibility and to prepare themselves for it.   Due to the fact that collective responsibility can be dismissed easily by invoking the individualistic principle of responsibility and therefore may not directly influence an individual’s practical reasoning, a collective decision mechanism ideally should manage to make clear the consideration of collective responsibility to individuals, help them to integrate collective responsibility into their practical deliberation concerning how to 89  conduct their moral and social lives, and thereby reduce these kinds of responsibility evasion.   In the following chapter, I will offer deliberative democratic practice as a paradigmatic example of collective decision-making, and argue that deliberative democracy is to be preferred over liberal democracy, because deliberative democracy is more capable of performing the functions mentioned above to cultivate a well-functioning social practice of collective responsibility.   90  Chapter Five: Democratic Practice as Paradigm Case of Collective Action  As discussed in chapter two, collective action takes different forms and structures. In some collective action, members share collective intention, plan together, and participate intensively, whereas in other cases, people act collectively in a loose fashion without necessarily sharing collective goals or actively participating in strict sense. Depending on different processes and structures of collective action, some ways of acting together enable participants to take collective responsibility more effectively than others. Following this idea, in this chapter I intend to propose that the problem of collective responsibility, first and foremost, invites a practical solution of structuring collective action, regardless of the various beliefs we might hold concerning the ontology of collective action or the nature of responsibility.   Taking collective responsibility, as argued in previous chapter, is a capacity possessed by different groups, which varies depending on their structures and organizational norms. In other words, depending upon particular group structures and collective decision-making procedures, participating members vary in their capacity for exercising collective responsibility. Given its contextual and relational nature, collective responsibility needs to be explained in reference to a particular social and political mechanism which, firstly, reveals to its members the contextual and relational information for the determination of responsibility; secondly, enables group discussion about responsibility among group members; and thirdly, cultivates both willingness and competence to take responsibility once it has been collectively identified. A well-91  integrated group which has met these conditions would be well positioned to ensure that its individual members achieve moral betterment through taking individual responsibility for collective actions.  This chapter takes democratic practice as a paradigm case of collective action. In this case, the responsibility of citizens for democratic political decisions exemplifies a particular instance of collective responsibility, the examination of which is a central task of this chapter. The conclusion I hope to draw is that a deliberative model of democratic practice is normatively superior to a model of vote-aggregation, because the former more effectively deals with the problem of collective responsibility. I will present my argument in four parts: first, I will discuss vote-aggregation as collective decision-making and provide a normative analysis; second, I will introduce the concept of deliberative democracy along with its underlying ideas; third, I will examine the capacity for dealing with collective responsibility in these two models and discuss the reasons for preferring a deliberative framework; lastly, I will respond to anticipated objections and challenges to my conclusions.   5.1  Vote-aggregation as democratic practice  People often collect information, form preferences, make decisions and take actions together. Many collective decisions and actions happen in political life, and, taken together, constitute democratic practices. The ways in which particular models of democracy generate political decisions depend upon adherence to specific procedures, principles and ideals of political participation. One prevailing democratic model is the 92  aggregation of majority votes in an election. The underlying idea of this model is straightforward. In order to reach group decisions concerning certain public issue among a population, each individual member of the group expresses a preference by casting a vote. The summation of these expressed preferences, namely the aggregated result of the majority votes, represents the collective decision.   This model of democratic decision-making upholds certain normative ideas about political life. First of all, it is built upon a strong egalitarian foundation. The political interests and judgments of all participants are considered to have equal validity, and therefore to hold equal entitlement in collective decision-making. Existing private interests of individual participants are taken to be the basis of political decisions without discrimination. In other words, vote-aggregation does not distinguish the quality of individual participation. It matters little who the voter is and how he or she comes to a particular decision. The electoral mechanism consequently has a high degree of inclusiveness. The threshold requirement for participation in political decision-making in voting is fairly low: the very act of casting a vote suffices. Moreover, built on previous merits of vote aggregation, the goal of politics is set to satisfy, rather than change or module, existing preferences. When facing political conflicts among different interest parties, vote aggregation provide a neutral method of arbitrating disputes by appealing to an abstract principle that few can reject in its entirety, the majority rule.   There are major concerns, however, with vote-aggregation as a democratic decision- making mechanism. Public choice theorists have repeatedly argued that the aggregation of individual preferences cannot be converted into a collective decision. The 93  practical coherence and normative implication of vote- aggregation are thrown into doubt by a model of majority-rule as ideal democracy. The criticisms come from three main directions. The first criticism concerns the practical coherence of the aggregative model. It is argued that there is no viable rule for aggregating individual preferences into public choice. Arrow’s General Possibility Theorem is taken to be a strong demonstration of this challenge. Second, even if acceptable rules are discovered for aggregating, the aggregated collective-decision lacks any motivational impact on individual participants, and therefore faces serious obstacles to implementation. Last and most important, vote-aggregation systematically fails to achieve the ideal of democracy. Democratic processes are intended not merely to ensure the satisfaction of individual preferences, but more fundamentally, to promote the development of individuality and the formation of well-deliberated preferences. In what follows, I will discuss these three classes of criticism in detail with the aim of showing why vote aggregation under majority rule cannot provide normative guidelines for democratic decision-making.   5.1.1 Lack of rules for aggregation The difficulty of treating the aggregation of individual preferences as the basis of collective preferences lies in the impossibility of finding adequate rules for aggregation. This claim is best demonstrated by Arrow’s General Possibility Theorem62.   In Social Choice and Justice, Kenneth Arrow builds a game-model in his analysis of international relations. The model aims to treat nations as unitary game-playing actors in game-theoretic scenario, since their preferences are derived from, and hence represent,                                                  62See K. J. Arrow, Social Choice and Individual Values, vol. 12, Yale university press, 1970.  94  the preferences of their individual citizens. What Arrow works out, according to Russell Hardin, is that “there is no general rule for aggregating to a social choice from individual preferences that can meet certain acceptable criteria.”63 These criteria all seem essential for an acceptable conversion of the ranking of individual preferences to the collective level. For instance, it is required that the collective preference between X and Y should depend only on how the individuals order these two alternatives.64 When the aggregative outcome of individual preferences fails to meet these criteria, it is highly doubtful that mere aggregation can render a reasonable and justified collective decision. Instead of a detailed discussion on the theoretical and mathematical aspects of Arrow’s theorem, I will present a voting scenario based upon Condorcet's paradox, which is essentially a straightforward instance of flawed aggregation to which Arrow’s Theorem refers.   Three voters 1, 2, 3, vote among three candidates A, B, C. Each voter's preference ranking of three candidates is shown in the table below:    Most preferred 2nd preferred Least preferred Voter 1 Candidate A Candidate B Candidate C Voter 2 Candidate B Candidate C Candidate A Voter 3 Candidate C Candidate A Candidate B TABLE 2                                                  63 Hardin, “Public Choice Versus Democracy”, 158 64The other criteria include unlimited domain, Pareto principle and non-dictatorship. For more details, see Arrow, Social Choice and Individual Values. 95   As each voter has his individual preference ranking of three candidates, a preference ranking of the collective cannot be produced from a mere aggregation under majority rules. Most simply, as the table shows, none of the candidates can be regarded as the leader, because the majority, namely two out of three voters would not agree to this. In other words, whichever candidate we put as the most preferred in the collective ranking, the resulting preference-ranking would not easily establish connection and consistency with its corresponding ranking of individual preferences. We can draw the conclusion that mere aggregation does not deliver much transitivity from individual preferences to collective ones. A more general lesson from Condorcet's paradox is that aggregation systematically fails to form rational collective choice from individual preferences, when there are multiple parties, each of whom selects more than three candidates. Arrow's theorem formalizes this discovery. The general conclusion is that there is no certainty that an aggregated outcome of individual preferences will correspond to the rational choice of a unitary collective agent. Even assuming a society, in which well-functioning rules and norms of trust are in place, and individual preferences are sincerely expressed and effectively registered, an aggregative democracy still fails to provide coherent “public will” and fails to create a solid foundation for political decisions.   5.1.2 Lack of motivational effect The second challenge aggregative democracy faces is the lack of motivational effect of collective decisions. Suppose we manage to discover a set of rules for the aggregation of individual preferences which produces coherent collective decisions 96  justified by various rational criteria. The new concern will be that the collective decisions cannot motivate individuals to act accordingly given that these decisions fail to directly accommodate existing individual preferences. This issue has been long identified as the main problem of collective action. The problem is well-documented in various forms and frequently discussed in the scholarly literature of political science and economics as the “free-rider” problem (Pareto, 1935), the “tragedy of the commons” (Hardin, 1968), and the “theory of public good” (Samuelson, 1954). In his 1965 book Mancur Olson explains that public decisions, which reflect collective rationality, fail to motivate individual members to take action accordingly. Using the model of the “prisoner’s dilemma” Olson demonstrates that it might not be in the interest of each individual in the group to take individual action to implement a public policy, even when the policy reflects the collective interest of a group. Therefore, the social action prescribed by collective rationality and decisions is likely to be distorted and hindered by the private interests of individuals, specifically by the individuals who are to put them into execution. This paradox has its moral analogue in collective responsibility, which is discussed in chapter three.   Given the lack of general aggregative rules and the lack of motivational effect, social choice theorists draw the conclusion, correctly I would argue, that collective decisions converted from the aggregation of individual preferences alone do not exert much normative force due to its incompetence two aspects: first, lack of general aggregative rules can often mean the disconnection between individual existing preference and the outcome of political decisions. It is questionable whether the outcome bears reflects its original sources, i.e. political participants’ existing private interests. 97  Secondly, the lack of motivational effect impedes further participation in voting once participants come to the realization that their preferences are not effectively reflected in the outcome decision. This circumstance would reduce democratic participation and thereby further undermine the legitimacy of the outcome decisions. It seems to me that the explanation and legitimization of political outcomes in a complex society requires more than a recipe for converting individual views into collective ones. In the context of votes aggregation under majority rules, as Russell Hardin points out, our understanding of democracy “cannot be grounded exclusively in the democratic procedural values.”65  5.1.3 The failure to be democratic Last, yet perhaps most significantly, is the criticism that vote aggregation systematically fails in achieving certain democratic goals, and therefore does not offer a proper model of collective decision-making in democratic politics. The underlying idea of aggregative democracy is that individuals engaged in politics are participating in an activity which is “typically, if not always, an effort to aggregate private interests. It is surrounded by checks, in the form of rights, protecting private liberty and private property from public intrusion”66 . Aggregative politics is essentially a market-like mechanism, mediating the interaction among individuals in a society. The goal of political decision-making is to identify various private interests, to negotiate well-balanced decisions, and successfully to implement the resulting policies by exercising public administrative power. Individual participants are assumed to be concerned mostly with private interests rather than with the public good. The aim in participating in                                                  65 See Hardin, “Public Choice Versus Democracy,” 172 66 Cass Sunsterin, “Democracy and shifting preference”, 196 98  political decision-making is to register participants’ interests in collective decisions and to realize these interests through collective action. However, one major goal of democracy, as some deliberative democrats argue, lies in an improved process of collective decision-making. The goal of democratic practice is not to efficiently satisfy the interests of participating citizens. Instead, the goal is to empower citizens and to make them better participants in democratic society. Aggregative models of democratic practice fall short in achieving these goals in a number of ways. I turn now to a discussion of these problems.   To begin with, an aggregative model takes existing private interests as the starting point for collective decision-making. Political participants should be taken as they are, not as what they should be or might be. The pursuit of private interests in politics is not only necessary, but ultimately good for the polity. The goals of politics should be narrowly defined. The polity should not concern itself with making its citizens better human beings. The political practice is limited instead to the establishment of rules and regulations under which people can be left alone to pursue their private interests in whatever way they deem fit. The advancement of private interests amounts to the advantage of the public. Attempts to shift or reshape these existing private interests require justification, or else are considered to be problematic and illegitimate.   Moreover, in aggregative democracy, individual decisions, which are primarily made out of the consideration of one's private interests, ought to be made in the absence of external influence. In other words, for individual political decisions to serve as a legitimate basis for collective decisions, these decisions should be made in isolation, 99  where participating citizens are free from possible manipulation. The preservation of individuality, which lies in undisturbed private interests and individual decisions, contributes to the legitimacy of the resulting collective decisions.   Deliberative democrats find these features of aggregative democracy normatively problematic. It has been argued that the political decisions reached by aggregating private interests often suffer from important, unavoidable, and endogenous difficulties in manifesting the public will. The best an aggregative democracy can do is to reflect the majority of existing individual preferences. At worst, there could be a complete disconnect between the intentions of individual participants and the resulting collective decision due to the lack of effective aggregating rules. Collective decision-making of this sort is “inadequate to the task of producing normatively binding political outcomes”67, and the solution to this problem cannot be “simply extended or even refined aggregative democracy”68. A mere aggregation of individual preferences and opinions lacks the moral resources required to generate and sustain legitimate political decisions for public issues.  5.1.4 Citizen responsibility in aggregative democracy In this section, I shall provide a moral diagnosis of aggregative democracy in light of the concept of collective responsibility. I argue that there are at least three aspects in which aggregative democracy fails to make its citizens take responsibility for collective political decisions. Firstly, given the relational nature of collective responsibility, an aggregative model of democratic practice fails to enable communication among its                                                  67 Jack Knight and James Johnson, “Aggregation and Deliberation: On the Possibility of Democratic Legitimacy,” 277.   68 Ibid. 100  participants about how to assign responsibility for the outcomes. The second drawback is that an aggregative model cannot easily justify the attribution of responsibility among individual members for the consequence of collective action. Lastly, to take collective responsibility often requires the group to continuously act collectively. The aggregative model does not empower citizens to generate necessary new collective actions for the taking of responsibility. I would therefore argue that an aggregative model of democratic practice is incapable of determining the responsibility of its participating citizens, by failing to convey information concerning collective responsibility, to motivate responsibility for collective actions, and to empower further collective action.   Firstly, in aggregative collective decision-making, the political participants see collective action as nothing more than a vehicle to carry out their individual intentions. As ongoing discussion shows, political decisions made under majority rules unavoidably fail to reflect some, if not most private interests. By casting a vote, participants are not invited to recognize and reflect contextually and relationally on their roles as citizens in a political community. The decision is made solely out of consideration of one’s assumed private interests. Other necessary elements do not factor in, including the particular context in which political decisions are made, the reasonable expectations held by all the possible effected parties, and various relations citizens hold with respect to one another. However, these elements are relevant and necessary for the determination of collective responsibility. When it comes to the distribution of responsibility for collective action, individual participants would---rightfully, according to the individualist understanding of responsibility---excuse themselves from responsibility or liability that does not stem from their own intentions and actions. Aggregative democracy does not enable public 101  deliberation and communication concerning collective responsibility for political decisions. Moreover, this model refuses to recognize the significance of public discourse over collective responsibility.   Secondly, the aggregative mechanism does not assist in the formation of new identities for the members of a political community, which means participating members neither acquire a collective perspective of their decisions nor acknowledge their commitment to collective action. When political decisions fail to accommodate interests of their individual participants or fail to manifest their intentions, as correctly diagnosed by some critics, aggregative democracy loses the normative binding force to its citizens. As a result, there is great difficulty not only in motivating continuous participation from the constitutive members, but also in holding members accountable for the consequences of collective action.   Lastly, without forming collective will or establishing group identity, even if participants come to realize the outcome of collective decisions made through majority rule, aggregative democracy does not provide a platform to initiate or generate further collective decisions in response to specific outcomes. Without sustainable collective identity, a group of political participants would not be able to identify its responsibility for collective action, or to be motivated to act in response to its consequences.   This problem of aggregative democracy has long been manifesting itself through voters’ non-voting behavior. In aggregative democracy, individual participation, i.e. vote casting, is often considered irrational. Voting has costs for each voter, such as, for 102  example, the time they have to spend and the transportation they have to arrange. However, the likelihood that any voter’s individual participation will be critical in assuring the success of his or her choice of candidate is rather slim. Therefore, under the aggregative system, there is little reason to participate at all, and even less to put a lot of time and effort into learning how best to participate. The severe disconnect between the casting of individual votes and the collective outcome alienates the participants from participating in their collective decision-making. This alienation in turns typically results in large numbers of people not voting.  Voting under aggregative democracy, I shall argue, is not the proper kind of collective action for generating collective responsibility. Voters do not self-identify as contributors to the outcome, and therefore do not feel disconnected from any consequential responsibility, because the mere aggregation of individual decisions does not provide extra normative resources for citizens’ responsibility other than the responsibility they have as individuals. The outcome is no more than a confluence of individual wills and interests. Citizens are ready to employ individualistic principles and to excuse themselves from taking responsibility for political decisions, for neither do they intend the outcome, nor could they have made any difference in it.   5.2  Public deliberation as democratic practice Since vote aggregation falls short in producing political decisions with normative significance, and provides little democratic legitimacy, the normativity of collective responsibility ought to be derived from other component of democratic practice. John 103  Dewey is regarded as a classic proponent of this view. He wrote in his 1927 piece “The Public and Its Problem” that “majority rule is as foolish as its critics charge it with being.”69 The essential need of democracy is “the improvement of the methods and conditions of debate, discussion and persuasion. That is the problem of the public”70. The legitimacy of political decision is produced by “the sort of reasoned argumentation under fair procedure that defines deliberation as a critical idea”71 . The appeal to public deliberation among equal and rational agents evolves into a contemporary political ideal, i.e. deliberative democracy.   Although there is not yet a settled and commonly accepted account of deliberative democracy among its proponents, the essential idea is that political legitimacy should derive itself from the will of the people formed through the public reasoning and arguments among equal and rational political agents.  “[Political] outcomes are democratically legitimate if and only if they could be the object of a free and reasoned agreement among equals.”72 This conception of democracy is different in various aspects from the aggregative one. Here I will discuss only those aspects of deliberative democracy that are relevant to collective responsibility. Two major components are regarded as necessary or desirable in deliberative democracy, which are absent or insignificant in aggregative democracy. They are 1) communication and sharing of reasons and judgments among participants; and 2) commitment to collectivity and possible common good.                                                   69 John Dewey, The Public and Its Problems: An Essay in Political Inquiry , 154. 70 Ibid. 71 Ibid. 72 Joshua Cohen, “Deliberation and Democratic Legitimacy,” 347 104   Ideally, participants of deliberative democracy bring into public deliberation the sincere and informed judgments of relevant issues, argue for them by appealing to considerations to which other rational equal participants can reasonably agree, stay fair-minded and ready to accept the judgment of other participants, go beyond ones’ self-interest and re-orient themselves to common good and thereby arrive together at political decisions with democratic legitimacy. A decision-making procedure in which individual participants are bound by norms of equality, reasonableness and freedom should generate outcomes that are in principle acceptable by everyone. This kind of procedure is considered by various contemporary political theorists as the favored approach to reach political consensus against the social background of diverse, even conflicting systems of belief and values. The early proponents of deliberative democracy proposed to build rationality and normativity into the procedure of public discourse about “what we ought to do”. John Rawls’ Veil of Ignorance and Habermas’ Ideal Speech Situations are both efforts of this kind. The central idea here is that democracy is deliberative in so far as participants engage in public reasoning about common good. The engagement with public reason can be hypothetical, such as the thought experiment of the “veil of ignorance” proposed by Rawls. By restraining the nature and scope of public reasoning, reasonable and fair-minded participants will figure out common good and fair political arrangement. Habermas also believes public reasoning is essential to the truth and rational laws that promote common goods and social justice. Later deliberative democrats refrained from making specific claims about objective standards by which the use of public reason generates political justice. They maintain that, although decisions made under the conditions of deliberative democracy may not be just, these decisions are more 105  likely to be. More importantly, actual political deliberation among free and equal participants is necessary for the legitimacy of collective decision. Politics, in the picture of deliberative democracy, is mostly regarded as a medium through which individual participants carry out practical reasoning for the collective. Political participation is structured in a way oriented towards mutual understanding, where autonomous citizens can mobilize their rationality jointly and publicly. This shared deliberative setting would induce collective decisions independently of the respective prejudices of participants, and generate legitimate political outcomes rooted in rational consensus among its members. Participants in deliberative democracy are committed to shared values such as rationality and impartiality. The participants do more than simply expressing and registering their preferences and interests in a political arena. They have to be open and attentive to one another, to justify their claims and proposals in terms acceptable to all. The hope is that, through the process of group deliberation, the participants’ orientation moves from self-regard to an understanding of what is publicly justifiable, and they become aware of their dependence on one another and act with deliberation as citizens. Deliberative democracy, thereby, helps further to shape and to develop the existing relations of a well-functioning society.  There are three theses held by deliberative democrats that are critical to the discussion of collective responsibility in democratic decision-making: epistemic value of deliberation; self-transformation thesis, and citizen development.   First, one major value of group deliberation is its epistemic reliability. Deliberative democracy commits itself to the pursuit of common goods about which 106  participants can reason and make judgment. To discover the truth about justice and the common good which democratic decisions aim at, deliberation among equal and free participants is necessary because otherwise it is not possible for individual participants to be fully informed and empathize with the situations, judgments and interests of all involved in collective decision. By deliberating together, participants not only learn about the private interests of other people, but also about the deep value-belief systems held by one another. They share information and verify judgments against one another. It breaks down the compartmentalization of rationality. Interaction of this sort helps the group to arrive at a better informed decision. Meanwhile, individuals who participate in the formation of this decision are well aware of how this decision comes about. It is an idea deliberative democrats borrowed from Mill, who holds that deliberation on laws leads to more rational and informed decisions, because free discussion and open debate allow relevant information to be distributed, mistaken reasoning to be exposed, and all the reasons for and against laws to be debated and considered. Contemporary deliberative democrat Iris Marion Young argues that “through the process of public discussion with a plurality of differently opinioned and situated others, people often gain new information, learn of difference experiences of their collective problems, or find that their own initial opinions were founded on prejudice or ignorance, or that they have misunderstood the relation of their own interests of others.”73   The second thesis is concerned with citizens’ self-transformation in their participation in democracy. This thesis holds that participants of a deliberative democracy go through a series of self-transformations, in their beliefs, experiences and judgments,                                                  73 Iris Marion Young, Inclusion Und Democracy (Oxford University Press, 2002), 26 107  and thereby become better political participants. In deliberative setting, where individuals are more broadly empowered, “they would become more public-spirited, more tolerant, more knowledgeable, more attentive to the interests of others, and more probing of their own interests.”74 For increased democracy transforms individualistic and conflicting interests into common and non-conflicting ones.   It is essential to understand that the interests behind collective decisions are not “causally prior to practices”75. Instead, the formation of group interests depends upon the membership of the group, its internal discursive and structural organization and its position in an overall system of social actions. Preferences and judgments of participants are “as much formed as revealed in the exercise of choice. Individuals choose in order to become, and the nature of the opportunities given for the expression of choice affects the formation of will.”76 Through participation and deliberation, individuals are expected to go through some preferences and capacity transformation, thereby becoming more aware of their co-membership in a collective form of life and their a priori social bond.  Cass Sunstein makes a similar point in his article “Democracy and Shifting Preferences”77. He points out that the preferences individuals bring into collective decision are never acontextual, for there is no way to avoid the task of initially attributing what we deserve. Whatever position we take in terms of what one deserves pre-                                                 74 Mark Warren, “Democratic Theory and Self-Transformation,” The American Political Science Review 86, no. 1 (March 1, 1992), 24 75 Samuel Bowles and Herbert Gintis, Democracy and Capitalism: Property, Community, and the Contradictions of Modern Social Thought (Routledge, 2012), 149 76 Benjamin Barber, “Strong Democracy." Berkeley, Cal.: University of California Press (1984), 183 77 See Copp, Hampton, and Roemer ed., The Idea of Democracy. 108  politically, it makes us value our deserts more than we would if the initial deserts have been attributed differently. In other words, preferences and interests introduced by individual participants into public discourse are always shaped and continuingly so. The significant question is not so much to build political practice and decisions based upon existing preferences and judgment. Rather, it is to have working social institutions have a positive bearing upon the shaping and formation of individual preferences and judgment of political life. Deliberative democracy understands the shaping force of political life and takes an active interest in cultivating the right kind of political institution, which promotes participants’ competence in being better citizens.   Thirdly, the deliberative model has the advantage of what Philip Pettit calls a “development rationale”78. Group deliberation can clarify to participants that they are constituent parts of a group, independent of their own intentions and interests. Participants will be strongly motivated to take up a new collective perspective, and to influence one another more with group deliberation. As a result, deliberative democracy not only helps a decision-making group to gain more deliberative capacity and become more responsive to external rationality, but it also helps participants to recognize the shared nature of the decision-making with which they are involved and to which they contribute.   Citizens can and should be morally motivated by justice or the common good and have a willingness to abide by and participate in democratic decisions regarding these                                                  78Philip Pettit, “Responsibility Incorporated*,” Ethics 117, no. 2 (2007), 176  109  values. Proponents of the deliberative ideal assume that citizens can be moved by either their commitment to reciprocity and public reason-giving, or by a sense of justice and willingness to cooperate with others, or by their commitment to deliberative institutions and norms arrived at through deliberation. Whichever reason it is, deliberative participants are expected to acquire a normative commitment to collectivity, which otherwise would not be present without deliberative participation.   5.3  Group deliberation enhances collective responsibility In my last section, I will explain how group deliberation can enhance the individual taking of responsibility at a collective level. Deliberative democracy is advocated as a solution in political practice to cope with the diverse and possibly even conflicting values background in modern society. In modern pluralist societies, political disagreement often reflects moral disagreement, as citizens with conflicting perspectives on fundamental values debate the group norms that govern their public life. Any satisfactory theory of democracy must provide ways of dealing with this moral disagreement. A fundamental problem confronting all democratic theorists is to find morally justifiable ways of making binding collective decisions in the face of continuing moral conflict. Deliberative theory of democracy provides a different—and better—approach to this problem because it leaves open the possibility that the moral values expressed by a wide range of theories may be justifiable. Fundamental principle of deliberative democracy is that citizens owe one another justifications for the laws they collectively impose on one another. When it comes to collective responsibility, which is a highly contested yet significant concept in moral and political life, I argue that decision 110  formation mode promoted by deliberative democracy has three advantages in assisting us in arriving in a more adequate picture of collective responsibility.   Discussion on collective responsibility often begins with one question: how can a group of individuals transform into a collective agent that bears a normative status as an individual agent does? Adequate collective agency among individual members is one key factor. Collective agency can be strengthened when a group of individuals is jointly committed in a certain way and constitutes a plural subject. Each individual member of this collective can be considered to owe one another actions that promote the fulfillment of their shared commitment (Gilbert, 2002). When members of the group are sensitive to one another's interests or, on a larger scale, to the interest of the group as a whole, a high level of group solidarity is achieved, to the benefit of collective agency (Feinberg, 1974). A less demanding account of collective action argues that the agential capacity of a collective can be cultivated without forming any plural subject. As long as members of a group have related sub-plans, recognize the interdependence of their intentions, each other's intention into the content of their own intention and give weight to it in shared deliberation, the group can carry out collective action and functions as a collective agent (Bratman, 1992). At last, a minimalistic condition for an individual to bear collective responsibility is to contribute to collective action, intentionally or not, knowingly or not. By complicity, individuals form participatory relations with collective action, and thereby are justified objects of responsibility attribution. (Kutz, 2001).  Without committing myself to any specific answer here, I believe a general observation can be made. For a collective to bear normative status and to take 111  responsibility, it needs to be internally well-integrated, responsive to external rationality, and capable of performing collective actions. In order to turn a collective into a proper normative agent, the interaction of its individual members has to be properly structured in order to generate adequate collective agency. Although I cannot provide a complete list of objective standards for adequate collective agent to be a proper moral agent, I intend to claim that the following helps cultivating a collective into a competent responsibility taker.   Firstly, given that collective responsibility is contextual and relational, a collective is in need of a mechanism which reveals relevant information to its members. Group members need to engage with communication and exchange of perception and judgment concerning the collective action in question. They need to share their understanding and knowledge concerning the purpose of collective action, the context where it is performed, the relations among all the effected parties, as well as various reasonable expectation on this action from all parties. Group deliberation generates epistemically more reliable knowledge of collective action among participants, makes more transparent distribution of duties and responsibility, and thereby better facilitates collective decision-making and better collective responsibility taking.  Secondly, to cultivate competent collective responsibility taker, it requires individual participants to take up a perspective of collective action different from their individual intentions and interests. Individuals need to perceive their decision and action as a constituting part of overall collective picture. Group deliberation, the process of giving and asking for justifications in public, enables individual members to experience 112  self-transformation, become public-spirited, tolerant, knowledgeable and attentive to the interests of others. Through this self-transformation, participants genuinely take other participants' intention, belief and value into their reasoning and perception, form a meshing web of sub-plans, to use Bratman's terms, and perform collective action together. The picture of shared responsibility relations among participating members has its chance to be constructed through public deliberation, well-recognized by its constructing parties and thereby gains its binding force from relevant parties.    Thirdly, through group deliberation, participants gain the extra motivation to take collective responsibility independently of whether collective decisions accommodate their individual private interests or intentions. The group in deliberation form relationships based on shared beliefs, judgment and commitments, and thereby becomes a plural subject. To be responsible for a collective decision resulting from group deliberation is, in this sense, to express one's self-identity as a member of a deliberative group which come to its decision in a normatively justified fashion. Taking responsibility implies an individual willingness to grant legitimacy to the public will of a political collective and to recognize one's place and contribution to this collective.   In sum, when deliberative democracy constructs responsibility for political decision and distributes this responsibility among participants, it is less challenged by the disappearance of collective responsibility. In other words, group deliberation cultivates competent participants of collective action, structures the interaction among them properly, and promotes collective agency among a group of individuals. Without group deliberation, group members do not have authorship over collective decision, because 113  interaction among them properly, and promotes collective agency among a group of individuals. Without group deliberation, group members do not have authorship over collective decision, because the collective decision resulting from an aggregation of individual preferences is merely what happens to individual members, rather than a collective choice.  Group deliberation over responsibility in collective action helps destroy the myth that there is a single, definite, uniquely determined responsibility inherent in any given wrongdoing conducted by individuals in collective action. A successful account of responsibility represents the interpersonal relationships in a community and serves as an action-guiding principle that maintains and fosters moral, social and political rules, and the interests which those rules protect. Communication among group members reveals the nature of interpersonal relations among individual members. Group deliberation over collective action obliges participants to think through the significance of their relationship to the collective structure, to take possible responsibility at the collective level as necessary inputs to individual practical and ethical reasoning, and to make collective decisions on the basis of this understanding.  5.4  Case discussion  A morally acceptable idea of responsibility will depend on the ability of people to reason beyond their individual interests and to consider what can be justified to people who reasonably disagree with them. The deliberative hope is that citizens who share and communicate their reasons can recognize that a collective decision carries moral worth 114  even when they think it remains morally disputable. In the absence of public deliberation, citizens are not under any obligation of mutual respect toward a political decision with which they disagree. Group deliberation having been carried out, however citizens are obliged to accommodate the moral judgments and beliefs of their opponents to the greatest extent possible.   In working out a well-functioning idea of collective responsibility, attempts to seek a secure foundation for collective responsibility, at the level of the most fundamental and general philosophical principles, takes us in the wrong direction. While philosophers debate various principles and criteria to define collective responsibility, political action requires participants to reach some basis on which to justify collective decisions and responsibility in the present, in the absence of the knowledge that would act as a guide to the attainment of valid fundamental principles of moral responsibility.   Before concluding this chapter, I offer a hypothetical scenario exemplifying public deliberation over collective responsibility, namely, an apology by the Canadian government in year to former students of Indian residential schools. As explained in the statement of apology from the Canadian prime minister, in the 1870's, the federal government began to play a role in the development and administration of aboriginal residential schools. Two primary objectives of the residential schools system were to remove and isolate children from the influence of their homes, families, traditions and cultures, and to assimilate them into the dominant culture. These objectives were based on the assumption that aboriginal cultures and spiritual beliefs were inferior and needed to be destroyed. Indeed, some sought, as it was infamously said, "to kill the Indian in the 115  child." In 2008, the Canadian government recognized that this policy of assimilation was wrong, and that it resulted in great harm. This case offers an example of the attempt to correct historical injustice, a long standing topic in the discussion of collective responsibility. One group, the present Canadian government, is taking responsibility for the harm caused to aboriginal communities through the residential school system. Let us consider how group deliberation in advance of this decision to appologize might unfold., The parties directly affected are included in an assembly to decide on whether or not such an apology should be issued, and several premises have been identified as starting points for further deliberation. Suppose, for the sake of argument that the group is to make the decision on the basis of considering three separable reasons for collective responsibility: first, whether the apologizing group, in this case the Canadian government, is causally responsible for the harm the victim group, i.e. aboriginal community, suffered; second, whether the apology, once issued, helps to address the historical injustice inflicted upon the victim group; third, whether it is proper to demand an apology from the present-day Canadian government. Various considerations would come in to play, each resulting in different responses. Some might think that the Canadian government as a continuous political body is causally responsible for the mistreatment of aboriginal people, and that the current Canadian government is under just as much obligation as any past government to issue such an apology. Moreover, an apologetic stand adopted by the Canadian government would likely help improve future conditions of the aboriginal community. Let us call this view optimism. Meanwhile, some pessimistic citizens may hold the view that the present Canadian government is to be held responsible and ought to issue the apology if such an apology is gainful. However, such an apology would not only fail adequately to address historical injustice, but likely justify future ignorance of 116  the problem. It is also possible for other participants to hold a hesitant attitude, thinking that such an apology would help to restore historical justice and that it would be proper to demand such an apology from present Canadian government if it is causally responsible for the injustice towards aboriginal communities. However, given that the present members of the government and Canadian citizens were not alive during this particular chapter of Canadian history, it is improper to demand an apology from them.   Now let us recall the deficiency with this aggregative model of collective action. In reference to this particular issue, one can see that any collective decision reached by the aggregation of votes would fail to reflect the rich and complicated normative considerations at play from various participating parties. Whether or not the affected parties taken as a group decide to issue an apology or not, the decision does not reflect a unified sense of why this decision is best in view of a specific set of circumstances n. When an individual member of this group is to explain or defend this particular decision and to provide normative reasons in defense of it, she is left with little to report other than this view being taken by the majority, albeit for reasons that will depend on the particular context.    Having considered the aggregative model, we can turn our attention towards the deliberative model. Rather than each party deciding beforehand from her own judgments and beliefs, and arriving at a private conclusion about what the group should do, members of the group should bring forward the relevant considerations with the goal of persuading others in the discussion, so the group as a whole can pool all kinds of opinions to work out a collective decision. The pooling of group considerations is 117  substantively different from reaching a unanimous agreement among group members. It gives rise to the commitment of the group as a whole to deliberating from those relevant considerations altogether, regardless of how individual members of the group might view them in private. In the case of group deliberation over Canadian government issuing an apology to aboriginal community, let us imagine that we have an individual participant holding a pessimistic view as described before. When group deliberation is conducted, we can reasonably expect the following scenario to happen to our pessimist participant. This pessimist would be ready to join the discussion with the private goal of convincing other members to veto this apology, because of its possible harmful consequences. Through group deliberation, our pessimist would express her opinions and concerns as well as listen to other people’s arguments for and against the decisions out of various considerations, some of which have occurred in her mind while some have not, some of which are agreeable yet some are not. Ideally, she will be able to convey her worry that an apology as such might have the unintended consequence of legitimizing future injustice against aboriginal people. At the same time, she might be convinced that this possible negative impact is not a strong reason against an apology. In contrary, an optimist might argue this apology serves as a good starting point for future acts to restore justice. The group as a whole might even go on to urge the government to commit itself to further policies and acts that will preclude this unintended negative impact. After the group arrives at a collective decision through discussions over the implications of these different aspects, our pessimist participant, despite having been opposed to the decision originally, can reasonably justify the group decision by saying that we, as a group, decided to issue the apology after considering its pros and cons. We, the group, are aware 118  of possible negative consequences of this proposed act and commit ourselves to taking future preventive steps.   When collective decision is reached in this fashion, it will not suffer from a lack of internal coherence and consistency since the group as a whole attends to all the rational relations of these considerations, and weighs them according to their relative importance and persuasiveness. Each individual member knows the exact reasons and processes that give rise to this outcome. She is able to report it, to recognize its reasonableness and significance, to justify it rightfully from the group perspective and to identify herself as an efficient contributor to this collective decision. Furthermore, our pessimist, along with the other participants, can be expected to be more ready to respond to negative consequences caused by this collective decision, and more capable of taking further collective action.   5.5  Replies to some possible criticism The main idea of this chapter is that group deliberation enhances collective responsibility, because it reveals relational and contextual understandings of group decisions, cultivates and sustains group identity through interaction, and enables further group decisions and action. However, it is not difficult to see that the success and efficiency of group deliberation can be challenged in various ways by solving the problem of collective responsibility. In this section, I will attempt to respond to some of these criticisms, and through these replies I hope to further clarify my position to some extent.  119   5.5.1 Failure to generate normativity When normatively contested ideas such as responsibility are left to be decided by group deliberation, how can one be sure that the outcome is morally tenable? The worry is that group deliberations are conducted by all kinds of participants in all kinds of ways. The deliberative outcome may stray from a normatively justifiable foundation, thereby failing for multiple reasons. For example, group deliberation would result in highly problematic decisions, when conducted by individuals who narrowly focus upon their own private interests, persistently refuse to commit to the value of rationality and impartiality, and try to manipulate others by eloquence and demagoguery in conversation, instead of equal, free and public reasoning. The critics f deliberative democracy have suggested numerous ways in which the public exchange of views ends up being manipulative, instead of deliberative. The parties to a discussion can suffer from asymmetrical sharing of relevant information, unjustified influence of strategic use of rhetoric, and other unfair and arbitrary distributions of deliberative power. Through public communication, people can be induced to hold inaccurate beliefs, and manipulated to have preferences that are not in line with their own interests79. As Susan Stokes pointed out, there is a deeper level of potential influence of deliberation: “to mold [people’s] very sense of who they are and what their capabilities are”80 . Though deliberative democrats embrace the idea that deliberation can transform people into better citizens, it is worrisome at the same time that deliberation might also “reduce the                                                  79 For detailed study, see Joshua “Cheap Talk” (1996), Diego Gambetta “Claro: An Essay on Discursive Machismo”(1998)  80Susan Stoke, “Pathologies of Deliberation”, Deliberative Democracy, Cambridge University Press, 1998, 124.  120  subject’s sense of his or her capabilities or to foist on them a sense of self at odds with his or her real needs and interests.”81 Pseudo preferences and identities can be created and fostered through group interaction and communication, and thereby displace real public interests with the special interests of certain parties in the deliberation, which are capable of manipulating and dominating public discourses.   The solution proposed by deliberative democrats to exclude manipulation of public opinion from legitimate group deliberation is often presented in two parts. The first step is advanced by Rawls as the idea of “percepts of reasonable discussion”82, which suggests the exclusion of unreasonable views or behaviors in group discussion. Those considerations that are driven by self-interests, blinded by prejudices and biased by ideology are not legitimately admissible to public discussion and debates. Only reasonable arguments are permitted in the domain of public deliberation83. The second step to prevent pathological deliberation is procedural. An early suggestion is made by Habermas with his proposal of an “Ideal Speech Situation”. According to this approach, when rules of public discourse are set properly, and people can talk with one another as free and rational equals, the force left at play in ideal speech situation would be the force of better argument. The ideal speech situation would be a space, where the voices of all concerned are uttered and listened to, no argument is arbitrarily excluded from consideration and only the force of the better argument prevails. If an ideal speech                                                  81Susan Stoke, “Pathologies of Deliberation”, Deliberative Democracy, Cambridge University Press, 1998, 124. 82Rawls, “Domain of the Political and Overlapping Consensus”, New York University Law Review, 1989, 233 83Note one major problem here is if emotion, instead of reasons, can gain its legitimacy in entering the public round of conversation.  121  situation is realized and group deliberation is successfully conducted, it will result in a consensus on the basis of reasons acceptable to all, and the resulting agreement will thus acquire its normativity and legitimacy. Communication, when conducted by parties during group deliberation to seek to persuade and convince one another, will, in a sense, make interlocutors commit themselves to the value of reasonableness when they are advancing relevant claims and positions. When interlocutors commit themselves to reasonableness in a group deliberation, either out of sincere respect for it or strategic consideration to convince other members, the deliberation will be less likely to result in morally objectionable conclusions. Even when a party participates in discussion with the goal of advancing her own private interests or biased proposition, she has to appeal to reasons and values that are shared by the group to justify her proposal. Over time, normativity and reasonableness will stop being mere lip service, and bind the participant with genuine accountability. This is what Jon Elster calls “the civilizing force of hypocrisy”84.   In sum, critics are skeptical about the legitimacy and normativity of collective decisions made by group deliberation, because group deliberation can be conducted in problematic ways by selfish, biased and manipulative participants. The reply is that group deliberation can deal with these problems by 1) setting up a standard to exclude illegitimate considerations, such that private interest and biased judgments will not come into play; and 2) preserving an ideal communicative situation where arguments with reason are the sole force at work in public discourse. When it comes to deciding                                                  84 J. Elster, “Deliberation and Constitution Making”, Deliberative Democracy, Cambridge University Press, 1998, 97 122  collective responsibility and its in-group distribution, the hope is that any understanding of responsibility attribution and distribution established by successful group deliberation as such will be deemed acceptable to all relevant parties, and becomes a solid basis for further collective decisions and action.   5.5.2 Against value pluralism The second challenge to group deliberation is value pluralism. The concern can be summarized as follows. Suppose a group has a well-functioning deliberative mechanism, which protects public deliberation from manipulation and distortion, and agreements produced through this deliberative mechanism will be considered as legitimate. If deliberative democrats are right, when such a collective decisionmaking mechanism is installed and operates in public life, it will forge shared consensus in society, locate and define political and moral truth, and foster respect for those whose cultures and world views are different from our own. The deeper problem of deliberative democracy is whether the group can reach a consensus against the background of moral conflicts and diversity of values, even when illegitimate considerations have been excluded and ideal procedures have been set up.   Some opponents of deliberative democracy, for instance, Fred Schauer, argue that the role of public deliberation is over-celebrated. The skepticism can be presented as a dilemma: deliberation is either unnecessary or unhelpful against the pluralism of values. On one hand, when it comes to particular normative problems, if there is fundamentally conflicting and diverse values and judgments in public views, deliberation even with all the ideal and favorable conditions in place is not going to give rise to group consensus 123  and fail to suggest solution acceptable to all parties. On the other hand, if there were no such persisting value conflicts, with ideal interactive situations and effective procedures in operation, other collective decision-making mechanism can produce consensus as legitimate as that created through public deliberation and possibly with more efficiency. Talking in and of itself does not contribute much to solve fundamental normative problems. As Schauer puts it, “discourse is better seen as embodying rather than transcending the range of virtues and pathologies of our collective existence”85.   In foregoing chapters I have proposed that the determination of collective responsibility is contextual and positional, which means responsibility is a normatively contested concept and its truth heavily depends upon specific social political and cultural contexts. It does not take much to see how realistic and forceful the pluralistic challenge is. Imagine that a deliberative group consists of Strawsonians, Frankfurtians and Parfitians. The group would have a great deal of hard time to propose a criterion of moral responsibility that is acceptable to all deciding parties. Even within a highly homogenous community, say a group of analytic ethicists, where every interlocutor has more or less equal deliberative and argumentative power, and shares perhaps more respect for rationality and reasons than any other social groups, it is hard to achieve consensual opinions concerning the determination of responsibility. As Sommers points out, in real life political practice, discussions are often carried out among people from dramatically different physical and social environments. As honor cultures have a difficult time understanding and justifying collective responsibility, members from societies of shame culture find it intuitively natural to connect blameworthiness with social disapproval and                                                  85 Frederick Schauer, “Discourse and Its Discontents”, Notre Dame Law Review, Vol72, 1313 124  to feel responsible for public acts over which they have little or no control86. The disagreement over the concept of collective responsibility does not reflect simple misunderstanding or mere lack of information on the part of deliberating parties, but rather the irreconcilability of conflicting values that run deep in the structure of our normative cognition.   I do not intend to deny the fundamental split in people’s views of collective responsibility. It is very likely that through group deliberation, participants cannot arrive at any consensus or agreements concerning the determination and distribution of collective responsibility, even after the relevant facts are laid out, the legitimate considerations are discussed and the deliberative procedures are strictly followed. How, then, does group deliberation enhance our understanding of collective responsibility in face of pluralistic criticism?   The answer to address the foregoing question consists in two parts. The first part invites us to revisit the limits of deliberative democracy, i.e. what deliberative democracy can and cannot deliver. The second part is to consider the remaining normative significance of deliberative democracy regardless of its limits.   Admittedly deliberative democrats often suggest that deliberation transforms preexisting disagreements through the give and take of argument, negotiation and compromise. As James Bohman explains in his survey article, “the goal of deliberation                                                  86Sommers, Relative Justice, Prince University Press, 2012, 65. 125  was consensus, the agreement of all those affected by a decision”87. However, it also has been recognized by many deliberative democrats that actual political deliberation rarely produces a consensus88. Cass Sunstein, for example, observes that deliberation sometimes gives rise to group polarization, a situation in which “members of a deliberating group predictably move toward a more extreme point in the direction indicated by the members’ pre-deliberation tendencies.”89The reply to such an objection is to see the limited capacity of deliberative democracy in solving fundamental conflicts in moral beliefs and values. It can be the case that after genuine group discussion over the fundamental truth of morality, people come to realize they have less in common than they had at first supposed. However, the moral differences and conflicts in public life revealed through deliberation should not be viewed as mere obstacles in political practice. On the contrary, it is the persisting conflicts in values and beliefs that make democratic politics ongoing and necessary. I would even go so far as to say that it is neither realistic nor desirable for any political arrangement to guarantee consensual consequence at all times. Now, having made it clear that consensus outcome is not necessarily what deliberative democracy is able to provide against a pluralism of values we are ready to take up the second part of the answer to the question: what normative significance does group deliberation have as a collective decision-making mechanism if its outcome is not acceptable to all the participants?                                                    87James Bohman, (1998).Survey article: The coming of age of deliberative democracy” ,Journal of Political Philosophy, 6(4), 400 88James Bohman, (1997). Deliberative democracy: essays on reason and politics. MIT Press. 648 89Cass Sunstein, “The Law of Group Polarization”, Journal of Political Philosophy, 2002, 81 126  Let me break down my defense of group deliberation to two parts. First, I intend to argue that group deliberation has positive significance in normative discourse regardless of whether or not it results in consensus of views, and its positive roles are different given whether an agreement is reached. I will use a hypothetical group deliberation over collective responsibility to demonstrate my point. Suppose there is a shared ground among all the deliberating participants concerning the determination of collective responsibility for a certain action, the rules agreed upon for responsibility attribution and distribution are more legitimate when they are derived from mutual consideration of different moral perspectives and standpoints than when they are the result of strategic calculation over the strength of competing opinions and interests90. Deliberation among people who agree with one another reveals the substantive beliefs and judgments they share. Participants become aware of the exact shared reasons and know the precise communal values they commit to together as a group by agreeing to the determination of collective responsibility. Such enhanced acceptance and mutual understanding would be necessary and valuable even under conditions where participants always share a great deal in common before agreeing to a collective decision.   The absence of such shared ground among participants concerning the understanding of collective responsibility public discussion, as I observed earlier, may prevent reaching an agreement or consensus. The value and significance of deliberation does not lie in recognizing or celebrating nonexistent communal values or shared reasons. Through rational disagreement, the democratic hope is that people would learn about the exact opinions and reasons they fail to share, and people would agree to disagree, agree                                                  90See Amy Gutman, Deliberation and Disagreement 127  about what to disagree and how to disagree. Not only agreement needs reasons for justification, but disagreement needs reasons for identification and legitimacy. Politics starts at this precise moment, when people need to cope with incommensurable conflicts and difference at the most fundamental level by perhaps suppressing them, ameliorating them, tolerating them, resolving them, or transforming them.91 As Benjmin Barber eloquently puts it, “to choose and act politically is to choose and act responsibly, reasonably, and publicly yet without the guidance of independent consensual norms. Where there is certain knowledge, true science, or absolute right, there is no conflict that cannot be resolved by reference to the unity of truth, and thus there is no necessity for politics…Politics concerns itself only with those realms where truth is not yet-known.”92 In other words, group deliberation is necessary particularly when the moral truth of collective responsibility is not yet-known and needs to be determined by all the participants within a specific context who stand to gain or lose as a result of such deliberations.                                                    91Benjamin Barber, (2003). Strong democracy. University of California Press, 129 92Ibid    Bibliography  Arrow, K. J. (1970). Social choice and individual values, vol. 12, Yale University press. Barber, B. (2003). Strong Democracy, University of California Press. Barry, B. (1991). Essays in Political Theory: Democracy and power. Clarendon Press. Benhabib, S. (1996) Toward a deliberative model of democratic legitimacy. Democracy and difference: Contesting the boundaries of the political, 67–94. Besson, S. & Marti, J. L. (2006) Deliberative Democracy And Its Discontents. Ashgate Publishing, Ltd. Bianchi, G. (2008). Introducing Deliberative Democracy: A Goal, a Tool, or Just a Context? Human Affairs: A Postdisciplinary Journal for Humanities and Social Sciences, 18(1), 100–106. Bohman, J. (2006). Deliberative Democracy and the Epistemic Benefits of Diversity. Episteme: A Journal of Social Epistemology, 3(3), 175–191. Bohman, J. (2009). Epistemic Value and Deliberative Democracy. The Good Society, 18(2), 28–34.  Bohman, J. F., & Rehg, W. (1997). Deliberative democracy: essays on reason and politics. MIT Press. Bratman, M. (1999). Faces of Intention: Selected Essays on Intention and Agency. Cambridge University Press Bratman, M. (1993). Shared intention. Ethics, 104(1), 97–113.   Bratman, M. E. (2006). Dynamics of Sociality. Midwest Studies In Philosophy, 30(1), 1–15.  Buchanan, A. (2003). Justice, Legitimacy, and Self-Determination : Moral Foundations for International Law: Moral Foundations for International Law. Oxford University Press. Buchanan, A. (2004). Political Liberalism and Social Epistemology. Philosophy & Public Affairs, 32(2), 95–130.  Cerny, P. G. (1995). Globalization and the changing logic of collective action. International organization, 49(4), 595–625. Christiano, T. (1995) Voting and Democracy. Canadian Journal of Philosophy, 25(3), 395–414. Cohen, J. (1997). Procedure and substance in deliberative democracy. Deliberative Democracy: Essays on Reason and Politics, MIT Press, Cambridge, 407–437. Cohen, Joshua. (1989). Deliberation and democratic legitimacy. Debates in Contemporary Political Philosophy, 342. Copp, by D. (2011). Reasonable Acceptability and Democratic Legitimacy,. Ethics, 121(2), 239–269. Copp, D., Hampton, J., & Roemer, J. E. (1995). The Idea of Democracy. CUP Archive. Crocker, D. A. (2006). Ethics of Global Development: Agency, Capability, and Deliberative Democracy: An Introduction. Philosophy and Public Policy Quarterly, 26(1-2), 21–27. Dewey, J. (2012). The Public and Its Problems: An Essay in Political Inquiry. Penn State Press.   Dryzek, J & List, C. (2003) Social Choice Theory and Deliberative Democracy: A Reconciliation. British Journal of Political Science, 33(1), 1–28. Estlund, D. (2011). Reply to Copp, Gaus, Richardson, and Edmundson. Ethics, 121(2), 354–389. Estlund, D. (1990). Democracy without preference. The Philosophical Review, 99(3), 397–423 Farrelly, C. (2003). Contemporary Political Theory: A Reader. SAGE. Frankfurt, H. G. (2005). What we are morally responsible for. Free Will: Libertarianism, alternative possibilities, and moral responsibility, 3, 280. Freeman, S. (2005). Deliberative democracy: A sympathetic comment. Philosophy & public affairs, 29(4), 371–418. French, P. A. (1979). The corporation as a moral person. American Philosophical Quarterly, 207–215. French, P. A. (1984). Collective and corporate responsibility. Columbia University Press New York Fung, A. (2005). Deliberation before the Revolution: Toward an Ethics of Deliberative Democracy in an Unjust World. Political Theory: An International Journal of Political Philosophy, 33(3), 397–419. Gambetta, D. (1998). “ Claro!”: an essay on discursive machismo. Cambridge University Press. Gehrlein, W. V., & Lepelley, D. (2011). Voting Paradoxes and Group Coherence. Springer.   Gibbard, A. (1973). Manipulation of Voting Schemes: A General Result. Econometrica, 41(4), 587–601. Gilbert, M. (1990). Walking together: A paradigmatic social phenomenon. Midwest Studies in Philosophy, 15(1), 1–14. Gilbert, M. (2006a). Rationality in collective action. Philosophy of the social sciences, 36(1), 3–17. Gilbert, M. (2006b). A theory of political obligation: Membership, commitment, and the bonds of society. Clarendon Press Oxford.  Glover, J. (1975). It Makes no Difference Whether or Not I Do It. Proceedings of the Aristotelian Society, Supplementary Volumes, 49, 171–209.  Goldman, A. I. (1994). Argumentation and Social Epistemology. The Journal of Philosophy, 91(1), 27–49.  Gomperz, H. (1939). Individual, collective, and social responsibility. Ethics 49 (3),329-342   Green, J. M. (2006). Pluralism and deliberative democracy: A Pragmatist Approach. In A Companion to Pragmatism. Malden MA: Blackwell Publishing. Habermas, J. (1995). Reconciliation through the public use of reason: remarks on John Rawls’s political liberalism. The journal of philosophy, 109–131. Habermas, J. (2006). Three normative models of democracy. Constellations, 1(1), 1–10. Hall, C. (2007). Recognizing the Passion in Deliberation: Toward a More Democratic Theory of Deliberative Democracy. Hypatia: A Journal of Feminist Philosophy, 22(4), 81–95.   Hardin, R. (1993). Public choice versus democracy. The Idea of Democracy, Cambridge University Press 157–72. Hedahl, M. (2013). The Collective Fallacy: The Possibility of Irreducibly Collective Action without Corresponding Collective Moral Responsibility. Philosophy of the Social Sciences.  Jackson, F. (1987). Group morality. Metaphysics and morality: essays in honour of JJC Smart, 91–110. Blackwell Publishing. Knight, J., & Johnson, J. (1994). Aggregation and Deliberation: On the Possibility of Democratic Legitimacy. Political Theory, 22(2), 277–296.  Kuper, A. (2005). Global responsibilities: who must deliver on human rights? Routledge New York.  Kutz, C. (2002). The collective work of citizenship. Legal Theory, 8(04), 471–494. Kutz, C. (2007). Complicity: Ethics and Law for a Collective Age. Cambridge University Press. Landemore, H. (2012). Democratic Reason: Politics, Collective Intelligence, and the Rule of the Many. Princeton University Press. Latane, B., & Darley, J. M. (1969). “Bystander Apathy.” American Scientist, 57(2), 244–268. Lewis, H. D. (1948). Collective Responsibility. Philosophy, 23(84), 3–18.  Mackenzie, C. (2006). Imagining Other Lives. Philosophical Papers, 35(3), 293–325. Mansbridge, J. J. (Ed.). (1990). Beyond Self-Interest. Chicago: University of Chicago Press.   Markovits, E. (2006). The Trouble with Being Earnest: Deliberative Democracy and the Sincerity Norm. Journal of Political Philosophy, 14(3), 249–269. Matravers, D., & Pike, J. (2002). Debates in Contemporary Political Philosophy: An Anthology. Routledge. May, L., & Hoffman, S. (1991a). Collective Responsibility: Five Decades of Debate in Theoretical and Applied Ethics. Rowman & Littlefield. McAfee, N. (2009). On Democracy’s Epistemic Value. The Good Society, 18(2), 41–47.  McMahon, C. (2001). Collective Rationality and Collective Reasoning. Cambridge University Press. McMahon, C. (2009). Reasonable Disagreement: A Theory of Political Morality (1st ed.). Cambridge University Press. Michelman, F. (1997). How can the people ever make the laws? A critique of deliberative democracy. Deliberative democracy: Essays on reason and politics, 145–71. Miller, D. (2001). Distributing responsibilities. Journal of political philosophy, 9(4), 453–471. Miller, David. (1992). Deliberative Democracy and Social Choice. Political Studies, 40, 54–67.  Miller, S. (2010). The Moral Foundations of Social Institutions: A Philosophical Study. Cambridge University Press. Misak, C. (2002). Truth, Politics, Morality: Pragmatism and Deliberation. Routledge Press. Mouffe, C. (1999). Deliberative democracy or agonistic pluralism? Social research, 745–758. 134 Munro, D. (2007). Norms, Motives and Radical Democracy: Habermas and the Problem of Motivation. Journal of Political Philosophy, 15(4), 447–472. Olson, M. (1965). The logic of collective action: public goods and the theory of groups. (Vol. 124). Harvard University Press.  Pasternak, A. (2011). The Collective Responsibility of Democratic Publics. Canadian Journal of Philosophy, 41(1), 99–123 Peter, F. (2007). Democratic Legitimacy and Proceduralist Social Epistemology. Politics, Philosophy and Economics, 6(3), 329–353. Pettit, P. (2007). Responsibility Incorporated*. Ethics, 117(2), 171–201. Rai, S. M. (2007). Deliberative Democracy and the Politics of Redistribution: The Case of the Indian “Panchayats.” Hypatia: A Journal of Feminist Philosophy, 22(4), 64–80. Rawls, J. (1989). Domain of the Political and Overlapping Consensus, The. New York University Law Review, 64, 233. Riker, W. H. (1982). Liberalism Against Populism: A Confrontation Between the Theory of Democracy and the Theory of Social Choice. Waveland Press. Rosenberg, S. W. (2007). Rethinking Democratic Deliberation: The Limits and Potential of Citizen Participation. Polity, 39(3), 335–360 Schauer, F. (1996). Discourse and its Discontents. Notre Dame L. Rev., 72, 1309. Scheffler, S. (2001). Boundaries and allegiances: Problems of justice and responsibility in liberal thought. Oxford University Press Oxford.   135  Searing, D. D., Solt, F., Conover, P. J., & Crewe, I. (2007). Public Discussion in the Deliberative System: Does It Make Better Citizens? British Journal of Political Science, 37(4), 587–618.  Searle, J. R. (1993). The Problem of Consciousness. Social Research, 60(1), 3–16.  Sen, A. K. (1999). Democracy as a Universal Value. Journal of Democracy, 10(3), 3–17.  Sher, G. (2009). Who knew?: responsibility without awareness. New York: Oxford University Press. Sommers, T. (2012). Relative justice. Princeton: Princeton University Press. Teorell, J. (2006). Political participation and three theories of democracy: A research inventory and agenda. European Journal of Political Research, 45(5), 787–810. Vanderveen, Z. (2007). Pragmatism and Democratic Legitimacy: Beyond Minimalist Accounts of Deliberation. Journal of Speculative Philosophy, 21(4), 243–258. Velasquez, M. G. (1983). Why Corporations Are Not Morally Responsible for Anything They Do. Business & Professional Ethics Journal, 2(3), 1–18. Vitale, D. (2006). Between Deliberative and Participatory Democracy: A Contribution on Habermas. Philosophy and Social Criticism, 32(6), 739–766. Warren, M. E. (1992a). Democratic Theory and Self-Transformation. The American Political Science Review, 86(1), 8–23.  Warren, M. E. (1993). Can Participatory Democracy Produce Better Selves? Psychological Dimensions of Habermas’s Discursive Model of Democracy. Political Psychology, 14(2), 209–234.   136  Warren, M. E. (1996a). Deliberative Democracy and Authority. The American Political Science Review, 90(1), 46–60.  Warren, M. E. (1996b). What Should We Expect from More Democracy?: Radically Democratic Responses to Politics. Political Theory, 24(2), 241–270 Warren, M. E. (2006). What Should and Should Not Be Said: Deliberating Sensitive Issues. Journal of Social Philosophy, 37(2), 163–181. Weithman, P. (2005). Deliberative Character. Journal of Political Philosophy, 13(3), 263–283. Young, I. M. (2002). Inclusion und Democracy. Oxford University Press.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items