Navigazione – Mappa del sito

HomeNumeri68Part III: crises of trust. Democr...Trust, discretion and arbitrarine...

Part III: crises of trust. Democratic and healthcare system

Trust, discretion and arbitrariness in democratic politics1

Patti Tamara Lenard
p. 83-104


Democratic institutions and practice depend on trust, in two ways. Citizens must trust each other to abide by shared rules and norms that together govern a political community; it is a feature of democratic states that they direct their resources not to enforcement of rule abidingness, but rather towards providing collective and public goods. Instead, states rely on the semi-voluntary compliance of citizens with these shared norms and laws. Citizens must also trust their political representatives, who via their election are granted discretion to make decisions that will have significant impact on the lives of those they represent. Yet, according to many scholars across many disciplines, this trust is at risk of being undermined, if it hasn’t yet been so.2 The possible reasons for weakened and broken trust are many, ranging from escalating wealth inequalities, to increasing cultural and ethnic diversity, to political incompetence of both voters and political representatives, to corruption among political representatives.3 In this article, I examine the trust that underpins the discretion accorded to elected representatives, to identify when the improper use of this discretion may reasonably be thought to be a source of distrust. In particular, I argue that where discretion is used arbitrarily, it may threaten to undermine the trust relations between citizens and their representatives, and therefore the efficient and effective running of democratic states.4

Torna su

Termini di indicizzazione

Torna su

Testo integrale

  • 1 I would like to thank two anonymous reviewers to this journal for their helpful comments, as well a (...)

1The first part of the article examines the notion of civic trust in democratic politics, in general terms. In the second and third parts of the article, I make the claim that at their heart, trust relations, both personally and politically, are about discretion. To say that we trust another person is to say that we believe that person will use his or her discretion well; to say that we do not trust another, or that another has betrayed our trust, is to say that we do not believe another will deploy, or has deployed, discretion, well. More specifically, the second part of the article offers an account of the relation between trust and discretion in some detail, as it is conceptualized in political theory; this examination suggests that, so far, what counts as ‘arbitrary’ is inadequately defined. The third part of the article therefore attempts to identify the features that define ‘arbitrariness.’ The fourth part of the article offers an account of non-arbitrary discretion, which is the form of discretion that sustains trust relations. Here, I identify four desiderata that must be fulfilled in order to describe discretion as non-arbitrary, and thereby able to sustain trust relations. It is worth observing that, lurking behind this way of thinking about trust is a confusion, which this article attempts to sort through, namely, the difference between a) claiming that someone has used her discretion poorly and b) claiming that someone has used her discretion arbitrarily. While both arbitrarily and poorly deployed discretion violate trust, in democratic states, the first is more dangerous than is the second.

1. Civic trust in democratic states

  • 5 Some have argued that trust violations produce feelings of betrayal, whereas misplaced reliance pro (...)
  • 6 Eric Uslaner understands trust as intimately connected to optimism. See Uslaner 2002. For an accoun (...)

2As it is ordinarily used, the word trust describes an attitude that some individuals have, either in general or in relation to particular others, which encourages or motivates them to believe that others will behave predictably and/or carry out the task with which they have been entrusted. Trust is thus best understood as a characteristic of relationships; individuals have trusting or distrusting (or somewhere in between) relations with particular or general others. This trusting attitude permits us to engage with others, by putting ourselves into a position of vulnerability with respect to their actions. When we trust others, we risk betrayal or disappointment, since those we trust can either fulfill or disappoint the trust we have placed in them.5 Crucially, to make the relation one of trust, rather than faith or reliance or optimism,6 we cannot be certain that the trustee will deliver on our expectations, nor can we simply be hopeful that they will do so; although disappointment or betrayal is unlikely, a certain degree of risk is an essential characteristic of trusting relations.

  • 7 Hardin 2002.
  • 8 For more discussions on the source of trust, see (Jones 1996; Adler 1994; Becker 1996).

3Using the language of risk in characterizing trust relations suggests that placing trust in others is a matter of probability assessment, as for example Russell Hardin suggests.7 On this view of trust, whether we should place or extend trust in others is an epistemic or cognitive question, where potential trusters use available evidence to ascertain whether their trust is likely to be rewarded. This way of understanding trust, however, does not capture the experience most individuals have of trusting; for most people, trusting others is implicit or unconscious, rather than explicit and calculated, and this is true whether the others are intimates or elected representatives. It will prove useful, below, to understand trust simultaneously as a matter of probability assessment and as extended (or not) more or less unconsciously; which aspect of trust (its epistemic or its implicit character) is central will depend on the specific circumstances.8

  • 9 Hardin 1999.
  • 10 Most importantly: Baier 1995.

4For some, trust is predominantly a relation among intimates: friends trust each other, family members trust each other, and perhaps co-workers can be said to trust each other. On this view, while trust is possible and welcome at the personal level – trust is partly what makes the personal relations we share valuable to us – the foundations of trust do not translate well to the larger scale. On the other hand, however it is we relate to elected representatives, or fellow citizens, it is not via trust relations; even if we refer to trust at the larger scale, moreover, it is not the same entity that is trust between intimates.9 In particular, explanations that require that we impute good will to others appear inadequate, according to some scholars,10 to explain the basis of extending trust relations to strangers.

5This conclusion is mistaken, however. The trust (or distrust) we have in intimates is based on our daily interactions with them – we collect information based on these interactions with them and extend trust or not. The depth of our knowledge is therefore great. At the community level, the relations we form with others are not based as clearly on intimate and deep knowledge, and so there is temptation to deny, as I described above, that the relations shared among co-members are ones of trust; in particular, it is impossible to know if other members bear good will towards us. When we say that a political community is pervaded by a climate of trust, we are clearly not saying that all members of that community bear good will towards each and every other member of that community. Rather, we are saying something more general, i.e., that we are mostly guided by a set of shared norms and values, which direct how we interact with each other in public space. These norms range from the simple, for example norms about whether to queue for the bus and who gives way when two people pass each other on a sidewalk, to the complex and controversial, for example, whether ‘we’ believe in socially provided health care or that taxes should on the whole be reduced. This sort of trust – perhaps it is valuable to refer to it as civic trust – motivates our cooperation in shared political and social institutions, and guides the informal and everyday interactions among members in shared space.

  • 11 Lenard 2007.

6I have detailed a more precise account of how this works in general, elsewhere, but to summarize my view: shared norms and values permit citizens to make assumptions, or to develop expectations, about how others will behave in certain situations; it gives them at least some information about those with whom they live, on the basis of which they can make predictions about how they will act in public environments.11 For example, if we all believe in the shared value of contributing our fair share to redistribution, then we will all comply with rules that require us to pay taxes to do so. Recall, that compliance with rules and regulations in democracies is required by law, but not easily and fully enforceable; the shared norms and values that underpin trust facilitate a more or less voluntary compliance with them.

7Thus, both civic trust and personal trust are attitudes that motivate the willingness to make ourselves vulnerable to others, who can then reward and reciprocate, or betray and disappoint, our trust. In the case of personal trust, we may be in a position to assess the motivation of those we may trust with considerable depth; in the case of civic or political trust, we rely on the existence of shared norms and values (for some scholars, which we impute to others and which thereby confirm their moral integrity) to extend trust.

  • 12 For an account of the various democratic institutions that have a representative function (and the (...)
  • 13 They are often intended to persuade us that their opponents are not similarly motivated or competen (...)

8How should we characterize the trust we have (or do not have) in our democratically elected political representatives?12 Do we trust them on the basis of our personal interactions with them? Or on the basis of the shared norms and values that we assume bind members of our political and social community? In fact, this trust relation is a hybrid, between personal and civic trust as described above. On the one hand, we do not trust political representatives on the basis of the same type of information we have about intimates. On the other hand, political representatives are trusted with more, and more of importance, than are those to which we extend simply civic trust; as I shall describe in more detail below, political representatives are trusted with considerable discretion to make decisions that often have as much impact on our lives as the decisions made by those closest to us. In the case of political representatives, we are provided with additional information, beyond their explicit commitment to shared norms and values, about their intentions, and motivations. In addition, we are provided information about their competence, i.e., about their abilities to carry out the job of political actor responsibly and reliably. This information, which aims to offer information on the basis of which potential electors can assess their relative trustworthiness, stems largely from what they provide us, or what we can learn, from a variety of public media. These media communications, where they stem directly from the representatives, are typically intended to persuade us that they share our concerns, and that they will act in our best interests, and that they are competent to do the latter.13 Third party actors – journalists, bloggers, etc. – can provide additional information from which we can assess the trustworthiness of potential representatives. Thus, the fundamental structure of the trust relation remains the same, but its basis is in part the shared norms and values that underpin civic trust and in part the extra information provided by public media, that allows for a kind of mimicking of the personal trust relation.

  • 14 For an account of the role of trust in the context of parenting, see Mullin 2005.

9All trust relations can be reciprocal – both parties in the relation trust the other – but they can also be one-sided, i.e., where one party trusts another, but the trust is not reciprocated. Perhaps the quintessential example of the latter is the child-parent relations, in which young children trust their parents, but the relation that parents have towards their children is not best described as trust.14 At first glance, the trust relation between constituent and political representative may appear to be one-sided; constituents place their trust in political representatives, and are therefore vulnerable to the decisions made by political representatives. Although standard mechanisms of transparency and accountability aim to protect citizens from the ways in which political representatives might, absent them, abuse their powers, it may appear that political representatives are not obligated to trust those they represent in turn. But the analysis above suggests that political representatives can make effective use of their discretion only if they trust that their constituents will accept, and abide by, the decisions they make. Where trust is high, moreover, representatives can count on this voluntary compliance even when decisions are contentious or controversial. In particular, we can say that in response to the responsible exercise of fiduciary duties by political representatives, citizens will comply with their decisions. Of course in democratic states, there are enforcement mechanisms that require citizens to behave in specific ways – i.e., that appear to enforce compliance with the dictates of political representatives. But, it is a mistake to ignore how few resources in a democratic state in fact go towards securing compliance; without the willing compliance with regulations and laws, democratic states would cease functioning, and representatives (and the political parties they are members of) rely on all citizens, even those who did not vote for them, to be more or less compliant with their decision-making.

  • 15 For analyses of trust that are attentive to these differences, see Williams 1998, Catala 2015.

10Having offered above a very general account of how trust operates in democratic states, it is worth noting that it is indeed a highly generalized account, which assumes a great deal of homogeneity among the values and norms that underpin trust relations among a population, and which assumes that the experience of complying with the rules and regulations that govern a population will (where trust is present) be experienced as more or less voluntary rather than coerced. But in any real democratic state, whether norms and values are truly shared, and correspondingly whether compliance in accordance with them is experienced as largely voluntary or coerced, is a matter of degree; any full evaluation of trust relations in a democratic state must be attentive to the ways in which trust itself is more accessible to some citizens than others (because the norms and values themselves only represent some groups and not others), and the ways in which governing laws are experienced as (mainly) voluntary or coercive map onto social and political cleavages.15

2. Discretion and trust

  • 16 Goodin 1986: 233.
  • 17 Ibidem: 234.
  • 18 Green 2009.

11What is discretion and what is its value? Robert Goodin offers two ways to think of discretion, one positive and one negative. The positive way to interpret discretion highlights the way in which someone, for example a political actor, is ‘empowered to pursue some social goal(s)... in such a way as he judges to be best calculated, in the circumstances, to promote those goals’.16 This understanding of discretion highlights that a political actor is permitted to take a range of possible actions, using her own judgement of how best to achieve an agreed objective. For example, if (to continue an example deployed above) citizens are known to be committed to redistributive taxation, then elected representatives have discretion to adopt a taxation policy that achieves this goal. The negative way to interpret discretion highlights instead the ‘area of conduct which is generally governed by rules but where the rules are indeterminate’.17 In other words, on this understanding, elected representatives are generally bound by a set of rules which govern their behaviour, and where the rules fail to specify what action is appropriate, they are permitted to use their discretion. The nature of politics is such that they will have ample opportunity to use their discretion in this way. The point is that elected representatives possess some freedom to act, i.e., to use their discretion, within the boundaries of agreed rules. To take a rather extreme example, elected representatives are often forced to respond quickly in the face of terrorist acts of significant magnitude, and in so doing, they are permitted, indeed they are required, to use their discretion to act in the ways that they believe are in the best interests of those they represent. Importantly, we would not ordinarily say that their use of discretion is arbitrary, even where we do not believe they made the best possible decision given the circumstances they faced: ‘they are guided by merit-based considerations, and they may also be guided by law even though not fully determined by it’.18 What counts as a ‘merit-based’ consideration is by no means transparent, and whether a political actor has adequately considered the merits of various options open to her before proceeding to make a decision is inevitably subject to scrutiny after the fact.

  • 19 Burke 1774.
  • 20 Brown 2006.

12This way of thinking of the role that elected representatives is delineated in Edmund Burke’s ‘Speech to the Electors of Bristol’, in which he offers the groundwork for the traditional distinction between ‘representatives as delegates’ and ‘representatives as trustees.’ Where representatives are understood to be mere delegates, they have been handed ‘authoritative instructions… which the Member is bound blindly and implicitly to obey, to vote, and to argue for, though contrary to the clearest conviction of his judgment and conscience’.19 On the delegate model of representation, representatives are asked to accomplish very specific tasks, which are fully delineated in advance of an election. Delegates are fully bound ‘to their constituents through elections and communication between elections’.20

  • 21 In general, see Eulau et al. 1959.

13No modern democracy can function effectively without relying on representatives, of course, and it is by now widely agreed that political decision-making is far too complex, and opportunities for engagement with citizens to gauge their preferences far too infrequent, to make the delegate model a plausible system of representation. For one thing, given the range of issues on which political parties campaign (and given the small number of political parties that participate in many elections), it is a rare case in which voters support all parts of a candidate’s or party’s platform. And even where citizens do support all aspects of a platform, acting on these issues is not the only role ascribed to political representatives: they must also make decisions on a daily basis, in situations that were not predicted at election time.21 It is not only impossible to predict what issues a given representative will be asked to consider in an election cycle, it will often be impossible for a representative to consult with citizens adequately before decisions must be made.

  • 22 For several important recent examples, see Mansbridge 1999; Rehfeld 2009; Urbinati, Warren 2008; Pe (...)

14Additionally, as contemporary scholars of representation have noted, neither does the trustee model capture the wide range of ways in which elected representatives (and non-elected political actors) operate in the political sphere. It is simply not the case that electors extend trust to their representatives to act as they see fit. Rather, upon analysis, it is clear that elected officers are called upon to make a range of decisions, and in some cases, it is appropriate to behave more like a delegate and in others, more like a trustee. And so, contemporary scholars have multiplied the varieties of representation available to elected officials, to better capture the ways in which they are permitted and expected to operate on behalf of those whom they represent.22 Across these more fully developed types of representation is a theme, however, which perhaps Burke would point out: across nearly all of them, the discretion that he believed was at the heart of the trusteeship model of representation plays an essential role. Thus, whatever conception of representation is in operation, it almost always functions because of the trust extended to representatives, which permits them to use their discretion in making political decisions. The discretion extended to the trusted representative does not mean that she has free reign to act as she sees fit; rather this individual ought to stay in regular contact with constituents (even if they cannot be directly consulted prior to her every decision), so that their interests and preferences continue to guide decisions that are ultimately made. But, the extension of discretion does permit representatives to make decisions on a daily basis, in a range of distinct situations, having ideally understood in advance the values, norms and interests that ought to underpin this decision-making.

  • 23 O’Neill 2002.

15In part because citizens are profoundly vulnerable to their representatives, there are significant constraints on the actions taken by them in political systems; for example, as mentioned above, accountability and transparency mechanisms in democratic states serve to constrain representatives. Elections (including recall elections, where they are permitted) are the ultimate accountability mechanism, though especially in democratic states a wide range of institutions and organizations participate in securing transparency and accountability, so as to protect constituents in a range of ways. Typically, they operate by requiring that political representatives account for, and report, the decisions they have made, so that constituents can at least in principle monitor the ways in which representatives deploy their discretion.23 In spite of these constraints, however, the discretion extended to representatives in modern democracies, to make decisions on behalf the constituents they represent, is considerable.

3. Non-arbitrariness

16Above, I argued that discretion is at the heart of representation (whatever its form) in modern democracies. Representatives are granted the discretion to make decisions, I argued, because their constituents trust them to use it well. But how do we know what it means to use discretion well? How can we be sure that the distinction between well-used and poorly-used discretion is not as Thomas Hobbes suggested, the distinction between whether we approve or disapprove of the way in which it is used? In particular, democratic decision-making appears full of situations where representatives may use their discretion well in some sense, but where we nevertheless believe they have made the wrong decision. In this Section, I suggest that well-used discretion is synonymous with non-arbitrarily used discretion, so that understanding whether discretion is used well or otherwise depends on understanding what it means to say that discretion is used arbitrarily.

  • 24 Pettit 1997: 55.

17One of the most thorough accounts of the dangers of arbitrariness in political theory is provided by republican political theorists, who argue that one’s freedom is compromised to the extent that one is subject to the arbitrary whims of others. For example, Philip Pettit says the following: ‘an act of interference is perpetrated on an arbitrary basis, we can say, if it is subject just to ... the decision or judgement, of the agent; the agent was in a position to choose it or not choose it, at their pleasure’.24 In the case of elected representatives, this way of understanding arbitrariness aims to distinguish between actions that are aimed at, and can broadly be understood or justified as, in the service of meeting the needs and interests of constituents, and in the domain in which a political actor is understood to be authorized to act, from those that are simply taken absent such considerations and/or outside of the appropriate domain.

  • 25 Constant 1988.

18More generally, when we say that democracies are characterized by the rule of law, what we are intending to communicate is that there is a publicly known set of rules and regulations, which apply to all citizens; the rule of law aims to protect people from the whim of individual lawmakers or authority holders.25 Thus, in principle, those charged with ensuring that we continue to live in and contribute to a cooperative political environment are entitled to exert their authority to protect the rule of law broadly understood, but they are not similarly permitted simply to exert their authority over us for whatever reason they choose.

19So, the first thing to notice about what makes discretion non-arbitrary is that is must be exercised in a space governed by the rule of law; in a democratic political space, the purpose of identifying and making public a set of rules, which are the result of democratic political decision-making procedures, is precisely to make clear that while individuals are subject to interference, they are not subject to violations of their freedom that stem from the application of rules that are kept secret from them. Yet, the application of publicly known rules is not adequate to render them non-arbitrary; if so, it would be the case that as long as regulations apply to individuals on the basis of some pre-determined criteria, they are non-arbitrary. But that is not necessarily the case; the applied rules must also be of a certain type.

20Second, the laws themselves must be subject to justification. In order to be non-arbitrary, it is not only that they must apply to all individuals, they must be justified to those to whom they apply. For example, taxation schemes are justified with reference to the valuable public goods that can thereby be centrally provided, and progressive taxation schemes are justified with reference to the importance of redistribution of wealth in political communities (and perhaps also with reference to the ills caused by excessive gaps in equality in political communities). This example highlights that where a law applies unequally to citizens, principled justification is required to render the apparent inequality non-arbitrary.

  • 26 Blake 2001.

21In his important considerations of coercion in political states, Michael Blake observes that while all political states act coercively upon its citizens, this coercion violates autonomy and is therefore unjustified only where no, or inadequate, justification is offered.26 This insight translates well to the discussion of arbitrariness – what makes the coercion arbitrary is where no justification is offered for the application of a law or where the justification that is offered cannot be accepted by those subject to the law. In the case of a political actor aiming to justify decisions she is making, the justification ought to make reference to what that actor was elected to accomplish, or at least more generally, to the interests and needs of her constituents, in the appropriate domain. For example, if an individual were elected to sit on a school board, but chose to divert the budget to provide better quality elder care, the individual would appropriately be described as acting arbitrarily for failing to act within the domain in which she was authorized to act.

  • 27 Many political theorists suggest that one way, or the right way, to understand what equality demand (...)

22Third (and this criterion can be understood as a specification of criterion 2, above), the rules must not only be determined in advance and publicly known, they must apply to all individuals equally, i.e., they are arbitrary where they apply only to some individuals, and not others, or where they differentiate between citizens on the basis of morally arbitrary characteristics, to use John Rawls’ well-known language. Historically, it was publicly known that the rights to vote and to own property were restricted to men (and often to white men), but these laws were nevertheless arbitrary because of the ways in which they differentiated between citizens on the basis of characteristics that ought to have been treated as irrelevant. A reader might object here, to point out that there are certainly cases where certain characteristics justify differential treatment; perhaps it makes sense to extend longer leaves to (birth) mothers (compared to fathers) because, in addition to having to care for a new baby, their bodies must physically recover from carrying and delivering a baby. Or, democratic states largely agree that individuals with mobility challenges are entitled to state provision of devices to enable their mobility and this violation in the distribution of resources stems from a commitment to protecting equality of all citizens in an important domain. In these cases, a commitment to equal respect for all citizens justifies the differential treatment.27

23Finally, in many cases, acceptable justifications will be those that defend a law with reference to the provision of public or collectively valued goods; the prioritization of some goods over others in a political community will reflect that community’s norms and values. The goods at stake are those that, sometimes, can only (or best) be provided to all citizens in common, from which it is difficult if not impossible to exclude free-riders, and include goods such as clean air and national defense. Second, they are often goods that are valued by citizens as a collective. Without getting bogged down in a complicated debate, for a variety of contingent, often historical, reasons, communities often come to value goods in common, to which they collectively believe it is worth contributing to protect over time. It is often said of Canadians that, although it is mere historical accident that they adopted a universal health care scheme in the 1970s while the US impeached the president who was pressing similarly forward, it is now a good that Canadians collectively value and are thus willing to contribute to sustaining over time. Other goods of this type are the preservation of green space or historical space, for example, or the sustaining of public libraries.

24We can therefore summarize non-arbitrary discretion in political communities as follows: 1) it is conducted in ways that are consistent with a set of publicly agreed rules, rather than the whim of an individual; 2) for which principled justification can be offered (in the case of elected representatives, based on the interests and needs of constituents); 3) it applies equally to all individuals, unless there are relevant characteristics that justify differential treatment that is consistent with a commitment to equal respect; and, 4) it is often aimed at furthering public or collectively valued goods (in the case of elected representatives, the provision of which falls within the domains in which they have been authorized to act), based upon the norms and values that define a given political community.

4. Arbitrary versus non-arbitrary discretion

25The violation of any of the above conditions renders discretion arbitrary, and thus undermines trust, though the mechanism through which the trust is undermined when each of the conditions is violated is distinct. There are two goals of this Section of the article, first, to articulate in some more depth what sort of discretionary actions taken by elected representatives violate each of these conditions and are thus arbitrary, and second, to articulate the specific way in which trust is undermined as a result of the violation of each of these conditions. This Section of the article is explicitly political, i.e., it focuses on situations where the discretion deployed is by political representatives, and thus trust between citizens and their representatives is undermined, and therefore also where the voluntary compliance on which democratic political practice relies more generally is undermined.

  • 28 For example Payton 2011.

26First, discretion is rendered arbitrary where it is deployed in ways that are inconsistent with publicly agreed rules. This is the historical account of arbitrary power, as I noted above, and its most common articulation. How does this account of arbitrariness apply to elected representatives and the discretion they are granted? Elected representatives are elected according to a set of rules which structure democratic politics in a particular state, and they are expected to respect not only these rules, but also the spirit in which they were constructed. As I indicated above in very general terms, the willingness to abide by shared rules and regulations is semi-voluntary in democratic states, and this applies to elected representatives as well. Electors expect their representatives to shape the law, certainly, but also to be willing to play by the agreed rules; the legitimacy of the laws they promote, defend and adopt depends in part on this willingness to, themselves, abide by the laws that govern the act of governing. Correspondingly, in cases where elected representatives violate these rules, they have violated the trust that underpins the discretion they have been granted; they have deployed their discretion arbitrarily by behaving as though the laws do not apply to them. As an example, consider the Canadian Progressive Conservative Party’s violations of electoral rules in 2011 (for example, by overspending on election campaigns)28 – for many, these apparently deliberate violations undermined trust in that party’s ability to govern (the party was governing at the time that its members were fined for over-spending) in accordance with shared laws.

  • 29 Lest there be any confusion, my claim is that arbitrarily-used discretion harms the trust relation (...)

27In these cases, political representatives appear to be deploying their discretion to explicitly reject the assumptions that underpin the basis on which trust is extended to them: 29 the view that everyone is playing by the same, publicly known, set of rules and regulations. Failure by elected representatives to play by the rules once elected (or to attempt to circumvent the electoral rules themselves, as in the example above) can damage trust severely. The harm stems from the explicit undermining of the beliefs that grounded the extension of trust, i.e., the belief that elected representatives were motivated by a commitment to the laws, and the spirit that underpins them, which characterize a political community. Where elected representatives take actions that are in explicit violation of how it is expected that someone who shares these norms and values should behave, trust in them in general is undermined. It can lead constituents to question whether these actors in fact share the commitment to the norms and values on which they extended their trust in the first place, or rather whether the actions in a particular domain are examples of a general disregard for shared norms and values that is likely to shape future decisions in which they had been authorized to act.

  • 30 One might imagine applying G.A. Cohen’s ‘interpersonal test’ here to determine whether laws have be (...)

28The second condition specifies that discretion is rendered arbitrary because its use is inadequately justified. Take for example the resurgence in proposed and adopted laws that permit the revocation of citizenship from some citizens – in particular dual citizens – who are found guilty of (or in some cases believed to have be planning) terrorism-related offenses. These laws are justified by the representatives defending them with reference to the important role they can play in fighting terror; citizens who aim to commit grievous acts of harm against their country of citizenship may well be deterred from doing so, if the punishment is loss of citizenship. In defending these laws, representatives provide little evidence that they will serve an important role in fighting terror, and they do even less to indicate why the already existing tools at their disposal are not sufficient to fight terror effectively. Absent this evidence, the vulnerability to which dual citizens are subject as a result cannot be justified to them.30 The failure of representatives who are proposing and defending citizenship revocation laws to offer plausible reasons to believe they will be effective at fighting terror, and indeed an improvement on existing laws that do so, undermines the trust that (at least some) citizens have in their government. Of course, in most political situations, there will not be unanimity in support of a given policy and the justifications offered for it; on the contrary, many policies and laws which can be plausibly justified with reference to the shared interests or needs of constituents, and implemented in the domains in which elected representatives are authorized to act, will nevertheless be rejected by some of those who will be bound by these decisions. Where no or inadequate justification is offered for a given, burdensome, decision, or where the offered justification cannot at least in principle be accepted by those who will feel its weight, it is arbitrary and therefore harmful to the trust that motivated extending discretion. An expectation of unanimity in support of a given policy is too high a burden to meet, however. The justificatory standard that elected representatives must meet here does not require that all citizens are always equally burdened by adopted policies, nor that they equally support these policies; the standard is that the justification must be one that offers a plausible connection between the goal that a policy aims at, and which should be widely shared, and the policy itself. Where a plausible connection of this type can be offered, the additional burden that some citizens may suffer can be acceptable to them, even where it is not desired, since the policy goal is one that they share. Absent this connection, however, the discretion that is deployed to pursue such a policy is arbitrary and therefore trust-undermining.

29In this case, where the harm to trust may be restricted to those who are burdened by a particular inadequately justified policy, and where these trust violations are limited to small numbers of citizens, a political community can persist without suffering significant ill effects. Over time, the failure of fellow citizens to stand up for others who are burdened by policies that are not plausibly connected to a shared policy goal can generate severe distrust among citizens that can prove difficult to reconstruct, however.

30The third condition, which can be understood as a specification of the second, highlights that discretion can be rendered arbitrary where it is deployed in ways that violate the principle that the equality that is meant to underpin the rule of law, in democratic states especially. In cases where elected representatives knowingly choose or support laws that burden some citizens more than others, or where they appear immune to worries that the application of laws are applied in more burdensome ways on some than on others, without adequate justification, or where they ignore evidence that highlights this unequal burden, discretion is rendered arbitrary. In the US state of Arizona, for example, citing the importance of apprehending irregular migrants, lawmakers granted police officers the authority to demand identity papers from citizens they stop for driving offenses. In this case, the law appears to meet the standards imposed by the first condition, i.e., that shared norms and values guide law-making in that state, but the application of the law permits the felt experience of harassment by American citizens of Hispanic origin.

  • 31 Pettit 1997.
  • 32 Having said that, of course there is considerable room given to judicial discretion, and a fuller d (...)

31The harm to trust is two-fold in this case: the trust that underpins the discretion permitted to elected representatives is violated and the civic trust between citizens is also at risk. The first harm stems from the willingness, sometimes consciously and often unconsciously, of elected representatives to differently burden citizens, who are ostensibly subject to the same laws; in the case of identity requirements in Arizona, the burdensome impact on Hispanic citizens was predictable, and the justification offered to them, that this burden was a legitimate side-effect of a policy deemed essential to apprehend irregular migrants, was inadequate from their perspective. A second harm stems from the awareness that some citizens are deemed less equal than others, in spite of purported claims to the contrary. In particular, the harms stems from the unwillingness of those who are less burdened by certain laws (in this case, White Americans) to work on behalf of those who are more burdened by them (in this case, Hispanic Americans) to attempt to equalize, as it were, the burden of penalties felt by all parties who violate the laws in question. As Philip Pettit observed, the civic trust that binds citizens is at risk where they are not willing to fight to protect each other from rights violations; this trust depends essentially on citizens being willing to stand up for those who are made vulnerable by laws that permit the violation of their rights in the name of supposedly collective goals.31 Two harms are in evidence here, one direct and one indirect: the harm to trust stems first and directly from arbitrarily deployed discretion, arbitrary since citizens cannot necessarily be sure when and to what extent they will be burdened by laws that, in principle, should be predictably applied to all citizens, equally, and second and indirectly, because those at risk may not be sure that their fellow citizens will stand up in their defense.32

32The fourth condition is the most controversial. This condition specifies that representatives should make decisions in ways that are consistent with the values and norms that govern a community. This condition specifies that whatever decisions are made, they are in keeping with the collective self-understanding of a political community. In these cases, discretion is arbitrarily deployed, when it is used in contravention of shared laws and/or norms; citizens may come to believe that their governors are deliberately contravening, or failing to understand, the ways in which citizens self-define as a political community. When US President Donald Trump announced his ‘travel ban’, which affected predominantly Muslim-majority countries, one repeated objection was that ‘America is a land of immigrants, and this policy is un-American!’. The point was to assert a shared value that defines the United States and to use it as a way to criticize a specific policy. Of course, not everyone agreed that such a ban was un-American, but it was a refrain that made sense in the US context.

33One reason to pay particular attention to the ways in which elected representatives understand the shared norms and values that govern a state, even where these are contested, is that in many cases, elected representatives are deploying their discretion in the service of providing or protecting public and collectively valuable goods. Discretion can therefore be rendered arbitrary, where the offered justification for a policy does not appear to support a public or collective good that is widely perceived to be valuable. In every political community, what counts as a public, or collectively valued good, and thus deserving of having public policy oriented towards its protection, is controversial.

34So, the third equality-of-respect condition constraints the justifications available (morally) to elected representatives, and the fourth condition specifies one significant type of justification that can be offered. Whether a good is collectively valued is often controversial in a political community and as a result subject to disagreement among voters and the parties they support. Consider proposals to make income tax more progressive, i.e., to require wealthier individuals to pay a larger portion of their income in taxes, to support those who are less well-off. Arguments for and against (more) progressive taxation will reference the public good, supposedly also collectively valued, that will thereby be produced, either redistributive income or the protection of individual property rights. Thus, for some, such plans will appear to be a fair way to distribute wealth in a political community and for others it will appear as unfair penalty on those who, through hard work or inheritance, are entitled to keep their income. Is discretion arbitrary if one or the other policies is pursued? Is trust thereby undermined?

  • 33 For example, see Festenstein 2009; Lenard 2007; O’Kelly 2006.
  • 34 Lenard 2012.

35Before I answer these questions, let me acknowledge two dangers here. First, it is of course always dangerous to claim that shared norms govern how ‘we’ engage in politics – as I indicated earlier, shared norms are never shared by all members of a political community, and they often reflect the preferences, values and interests of a dominant or majority group.33 Elsewhere, I have articulated the features of the best kind of shared public culture, which focus on ensuring that the space in which norms are deliberate are genuinely open.34 It is admittedly difficult to ensure that the public culture is genuinely public, and yet, where such shared norms do, genuinely, exist among a cross-Section of the population, they are taken to be nearly as robust as are laws in how a community understands the limits and constraints on decision-making by political representatives.

  • 35 I thank a reviewer for this journal of pressing me to address this particular challenge to my accou (...)

36A second and related danger is that there will be instances where the shared norms and values that define a political community are morally problematic in some way.35 Typically, these emerge in cases where the domains in which these norms and values are defined and deliberated are not fully open; as a result, exclusionary norms result. To take just one example: it may be the case that Hungarian public culture is defined by xenophobia (which is being encouraged by its present leader), but to the extent that it is so defined, it does not meet the definition of public that defines a genuinely shared public culture. One way to ensure that norms that appear to promote inequality among a population are not sustained over time is precisely to focus on opening the space in which norms and values are deliberated to those who have been, traditionally, excluded. The reason to make this challenge explicit here is that there may well be cases where representatives use their discretion in ways that contradict the norms and values that are said to be shared, to pursue more ‘progressive’ policies, for example to support marriage equality in an environment shaped by homophobia. In these cases, it is possible to both applaud the representative for making a choice that is consistent with equal respect for all citizens (and so meets the third criterion of non-arbitrary use of discretion, detailed above), but recognize that she may be understood by some among those she represents as having violated the trust on which discretion was extended, i.e., to have used her discretion arbitrarily.

  • 36 What counts as a ‘significant’ number is, of course, difficult to determine in general. I leave tha (...)

37How to adjudicate questions of public goods, and others like it, are difficult to evaluate outside of particular political contexts, since what matters is whether elected representatives adopt policies citing goods that can plausibly be interpreted to be valued by a significant proportion of citizens. In particular, the question at stake is whether citizens believe that the decision to go forward in one direction or the other can be justified with reference to some allegedly collectively valued good, even if it is not one that they themselves support. In evaluating this sort of case, the difference between arbitrarily and poorly used discretion comes to the fore. Discretion is arbitrary, and therefore trust is undermined, where citizens believe that the justification for a policy does not appear adequately connected to a good that can plausibly be justified as valuable to a significant number of citizens.36 Thus harm to trust can stem from the belief that elected representatives are not interested in what is genuinely in the best collective interest of a community; i.e., from arbitrarily used discretion.

  • 37 Anderson et al. 2005.

38Articulating the exact ways in which elected representatives can prove, to the satisfaction of constituents, that they are truly motivated by an interest in protecting a public or collectively valuable good are beyond the scope of this article, beyond nothing that these ways will be context dependent. Where policies are adopted that some citizens find problematic, trust may also be harmed, but less significantly when the harm stems from the belief that elected representatives are indeed interested in what is in the best collective interest of the community, but that there is a legitimate disagreement about how to characterize the collective interest. It is a necessary feature of democratic politics that citizens will disagree, and therefore that some will always be represented by individuals who disagree with them about the best path to pursue. But, democracies rely on the consent of ‘losers’ in these disagreements, and trust (and consent) is sustained between winners and losers where losers remain convinced that winners are abiding by the shared norms and values that form the foundation of trust in political communities in the first place.37 Among the foundational shared norms and values that must be shared, even in the face of disagreement on specific policy issues, are those that are required to support the democratic process in general, including support for the laws that govern the elections of representatives in the first place.

39Above I have offered a 4-pronged account of what makes discretion non-arbitrary. By way of concluding this Section, allow me to use these resources to distinguish between arbitrarily used discretion and poorly used discretion, both of which promise to undermine trust, by spelling out what constitutes the poor use of discretion.

  • 38 Weinstock 1999.

40Poorly used discretion is of two broad types. In one sort of case, poorly used discretion is a matter of perception, i.e., there is a considerable subjective element to it; where political disagreement divides constituents and their preferred representatives, citizens interpret their opponents as using their discretion poorly. They believe their opponents are genuinely committed to the political community, but reject the way in which they interpret the needs of that community, and thus the policies that are in its best interest. In a second sort of case, discretion is poorly used, because it is based on faulty reasoning, or evidence that is clearly inadequate, and identifiably so at the time it is relied upon; elected representatives are acting incompetently in these cases of cases, and thus trust in them was misplaced. For example, consider Colin Powell’s assertions that Iraq possessed the ability to create and stockpile weapons of mass destruction, and thus that his best interpretation of the available evidence justified the case for the American-led war in Iraq. Those who believe that he acted on the best evidence available do not believe he acted arbitrarily; those who believe that the evidence on which Powell made his decisions and assertions was recognizably false at the time he relied on it, believe he acted arbitrarily; and those who believe that inadequate action was taken once the evidence was proven to be faulty likewise believe that those responsible acted arbitrarily. Trust is warranted or justified only where the evidence on which it is based is robust over time; and although trust is widely known to be resistant to evidence to a considerable degree,38 it is ultimately vulnerable to evidence that it is misplaced.

  • 39 This is a bit more complicated, of course, since whether judges possess and deploy discretion in th (...)

41In these cases, the conditions for the non-arbitrary use of discretion are met, but inadequately so in some sense. Both kinds of poorly used discretion can generate harms to trust, and while these harms have the potential to be long-standing, they need not be, since in both cases policies can be altered in response to a new election, or evidence that constituents’ preferences have changed, or in response to better evidence. When the United States’ Supreme Court upheld marriage equality, one significant reason it did so (and why many elected representatives hoped for marriage equality and thus publicly supported the decision) was because the mores of American citizens had changed sufficiently to warrant a change in policy. It is conceivable that a different decision would have threatened to undermine trust between political representatives and constituents (though of course Supreme Court justices are not elected officials, they do nevertheless have significant discretion, and their selection is among the most important jobs of elected officials in the United States).39

42To summarize: distinguishing between arbitrarily and poorly used discretion in political environments, both of which are the result of the extension of trust from constituents to political representatives, highlights two forms of harm to trust. The former, arbitrarily used discretion, harms trust by forcing constituents to reconsider whether those they have trusted continue to operate following the same set of norms and values on which the trust was extended in the first place. The trust violations in these cases are profound, and threaten to generate wide distrust in the political-decision making capabilities of representatives in general. The latter, poorly used discretion, harms trust by suggesting that representatives are not able or willing to use the best available evidence to make policy decisions in the interests and preferences of their constituents, or that they are unwilling to recognize when they have done so mistakenly. The worry is that representatives are well-meaning, but incompetent, and thus that the trust placed in them is unwarranted. This latter form of harm is less egregious however, and can be remedied either by compensating actions by the representative or her political party (altering of the policy, for example).

5. Conclusion

43In a political environment in which commentators are decrying the absence of trust in democratic politics, and in which they worry about the dangers to falling levels of civic trust, it is important to know which trust violations by elected representatives are likely to damage democratic political practice. The willingness to extend trust to political representatives is essential to democratic politics, for the ways in which it permits and indeed authorizes their use of discretion. In light of this, the goal of this article has been to offer a typology that permits us to focus attention on the dangers to trust of extending discretion to elected representatives. The article began by examining the nature of trust and its role in the discretion extended to elected representatives in particular. It then considered the harms of arbitrariness, as depicted in republican political theory, and suggested that discretion is sometimes used arbitrarily in some political situations and poorly in others. I offered a four-pronged account of arbitrary discretion and used this account to distinguish it from merely poorly used discretion. The former, I argued, is more dangerous to trust relations in democratic states, and the latter though also trust-damaging, can be more easily remedied.

Torna su


Adler, J.E., 1994, Testimony, trust, knowing, “Journal of Philosophy”, 91, 5: 264-275.

Anderson, C.J., Blais, A., Bowler, S., 2005, Losers” Consent: Elections and Democratic Legitimacy, Oxford, Oxford University Press.

Baier, A., 1995, Moral Prejudices: Essays on Ethics, Cambridge, Harvard University Press.

Becker, L.C., 1996, Trust as noncognitive security about motives, “Ethics”, 107: 43-61.

Blake, M., 2001, Distributive justice, state coercion, and autonomy, “Philosophy & Public Affairs”, 30, 3: 257-296.

Brown, M.B., 2006, Survey article: Citizen panels and the concept of representation, “Journal of Political Philosophy”, 14, 2.

Burke, E., 1776, Speech to the electors of Bristol, “Online Library of Liberty” (

Catala, A., 2015, Democracy, trust, and epistemic justice, “The Monist”, 98, 4.

Cohen, G.A., 2009, Rescuing justice and equality, Cambridge (MA), Harvard University Press.

Constant, B., 1988 [1816], The liberty of ancients compared with that of moderns, in B. Fontana (ed.), Benjamin Constant: Political Writings, Cambridge, Cambridge University Press.

Dogan, M. (ed.), 2005, Political Mistrust and the Discrediting of Politicians, Leiden, Brill.

Dworkin, R., 1963, Judicial discretion, “The Journal of Philosophy”, 60, 1.

Eulau, H. et al., 1959, The role of the representative: Some empirical observations on the theory of Edmund Burke, “American Political Science Review”, 53, 3: 742-756.

Festenstein, M., 2009, National identity, political trust and the public realm, “Critical Review of International Social and Political Philosophy”, 12, 2: 279-296.

Goodin, R.E., 1986, Welfare, rights and discretion, “Oxford Journal of Legal Studies”, 6, 2.

Green, L., 2009, Legal Positivism, in E.N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (

Hardin, E.N., 1999, Do we want trust in government?, in M.A. Warren (ed.), Democracy and trust, Cambridge, Cambridge University Press.

Hardin, E.N., 2002, Trust and Trustworthiness, New York, Russell Sage Foundation.

Hertzberg, L., 1988, On the attitude of trust, “Inquiry”, 31, 3.

Holton, R., 1994, Deciding to trust, coming to believe, “Australasian Journal of Philosophy”, 72, 1.

Jones, R., 1996, Trust as an affective attitude, “Ethics”, 107, 1: 4-25.

Lenard, P.T., 2007, Shared public culture: A reliable source of trust, “Contemporary Political Theory”, 6, 4: 385-404.

Lenard, P.T., 2012, Trust, Democracy and Multicultural Challenges, University Park, Pennsylvania University State Press.

Lenard, P.T., 2018, Democratic institutions and representative trust, in F. Boucher, D. Karmis (eds), Trust, Distrust and Mistrust in Multinational Democracies, Montreal, McGill-Queen”s University Press: 68-85.

Mansbridge, J., 1999, Should Blacks represent Blacks and women represent women? A contingent “Yes”, “The Journal of Politics”, 61, 3: 628-657.

Mullin, A., 2005, Trust, social norms, and motherhood, “Journal of Social Philosophy”, 36, 3: 316-330.

O’Kelly, C., 2006, Public institutions, overlapping consensus and trust, “Critical Review of International Social and Political Philosophy”, 9, 4: 559-572.

O’Neill, O., 2002, A Question of Trust: The BBC Reith Lectures 2002, Cambridge, Cambridge University Press.

Payton, L., 2011, Conservative Party Fined for Breaking Election Laws, CBC News online, 10 November 2011 (

Pettit, P., 1997, Republicanism: A Theory of Freedom and Government, Oxford, Oxford University Press.

Pettit, P., 2010, Representation, responsive and indicative, “Constellations”, 17, 3: 426-434.

Rehfeld, A., 2009, Representation rethought: On trustees, delegates, and gyroscopes in the study of political representation and democracy, “American Political Science Review”, 103, 2: 214-230.

Urbinati, N., Warren, M.A., 2008, The concept of representation in contemporary democratic theory, “Annual Review of Political Science”, 11: 387-412.

Uslaner, E., 2002, The Moral Foundations of Trust, Cambridge, Cambridge University Press.

Uslaner, E., 2004, Divided citizens: How inequality undermines trust in America, “Demos: A Network for Ideas and Action” (

Warren, M.A., 2004a, Democracy and deceit: Regulating appearances of corruption, “American Journal of Political Science”, 50, 1: 160-174.

Warren, M.A., 2004a, 2004b, What does corruption mean in a democracy?, “American Journal of Political Science”, 48, 2: 328-343.

Weinstock, D., 1999, Building trust in divided societies, “Journal of Political Philosophy”, 7, 3.

Williams, M.S., 1998, Voice, Trust and Memory: Marginalized Groups and the Failings of Liberal Representation, Princeton, Princeton University Press.

Torna su


1 I would like to thank two anonymous reviewers to this journal for their helpful comments, as well as Carolyn McLeod and Amy Mullin for written comments on an earlier version of this article.

2 A note about language: in this article, I use the expressions ‘harm to’ trust and ‘undermining of trust’ relations to be equivalent in meaning. It is not (necessarily) a moral claim (i.e., there may well be conditions where undermining trust relations or harming them is a good to be pursued).

3 For example, Warren 2006, 2004; Uslaner 2004; Dogan 2005.

4 This article is a normative exploration of the ways in which one possible trust violation – the violation of the discretion extended to political representatives – can undermine trust overall. It should therefore be understood as predictive rather than empirical.

5 Some have argued that trust violations produce feelings of betrayal, whereas misplaced reliance produces disappointment. See for example Holton 1994. See, also Mullin 2005.

6 Eric Uslaner understands trust as intimately connected to optimism. See Uslaner 2002. For an account of the distinction between trust and reliance, see Hertzberg 1988.

7 Hardin 2002.

8 For more discussions on the source of trust, see (Jones 1996; Adler 1994; Becker 1996).

9 Hardin 1999.

10 Most importantly: Baier 1995.

11 Lenard 2007.

12 For an account of the various democratic institutions that have a representative function (and the ways in which trust is extended to those who operate these institutions), see Lenard 2018: 68-85.

13 They are often intended to persuade us that their opponents are not similarly motivated or competent, as well.

14 For an account of the role of trust in the context of parenting, see Mullin 2005.

15 For analyses of trust that are attentive to these differences, see Williams 1998, Catala 2015.

16 Goodin 1986: 233.

17 Ibidem: 234.

18 Green 2009.

19 Burke 1774.

20 Brown 2006.

21 In general, see Eulau et al. 1959.

22 For several important recent examples, see Mansbridge 1999; Rehfeld 2009; Urbinati, Warren 2008; Pettit 2010.

23 O’Neill 2002.

24 Pettit 1997: 55.

25 Constant 1988.

26 Blake 2001.

27 Many political theorists suggest that one way, or the right way, to understand what equality demands in democratic states is as equal respect or recognition.

28 For example Payton 2011.

29 Lest there be any confusion, my claim is that arbitrarily-used discretion harms the trust relation between representative and constituent, not that it is the only way in which it can be harmed. I do happen to believe it is a central reason for the decline in trust in politics overall, but I cannot make that argument here.

30 One might imagine applying G.A. Cohen’s ‘interpersonal test’ here to determine whether laws have been adequately justified. See Cohen 2009.

31 Pettit 1997.

32 Having said that, of course there is considerable room given to judicial discretion, and a fuller discussion here warrants considering where and when this discretion is properly used.

33 For example, see Festenstein 2009; Lenard 2007; O’Kelly 2006.

34 Lenard 2012.

35 I thank a reviewer for this journal of pressing me to address this particular challenge to my account.

36 What counts as a ‘significant’ number is, of course, difficult to determine in general. I leave that question aside here.

37 Anderson et al. 2005.

38 Weinstock 1999.

39 This is a bit more complicated, of course, since whether judges possess and deploy discretion in their reasoning is a matter of debate among legal theorists. Famously, Ronald Dworkin rejects the claim that judges use their discretion to make decisions. See Dworkin 1963.

Torna su

Per citare questo articolo

Notizia bibliografica

Patti Tamara Lenard, «Trust, discretion and arbitrariness in democratic politics»Rivista di estetica, 68 | 2018, 83-104.

Notizia bibliografica digitale

Patti Tamara Lenard, «Trust, discretion and arbitrariness in democratic politics»Rivista di estetica [Online], 68 | 2018, online dal 01 mars 2019, consultato il 19 juin 2024. URL:; DOI:

Torna su

Diritti d’autore


Solamente il testo è utilizzabile con licenza CC BY-NC-ND 4.0. Salvo diversa indicazione, per tutti agli altri elementi (illustrazioni, allegati importati) la copia non è autorizzata ("Tutti i diritti riservati").

Torna su
Cerca su OpenEdition Search

Sarai reindirizzato su OpenEdition Search