Trust and Institutions

Last week I attended part of a fascinating conference on Trust, organized by the Blavatnik School of Government in Oxford. In her opening paper, Katherine Hawley raised many interesting questions, including those of whether trustworthiness is a virtue and whether it can be a virtue of institutions.

As an Aristotelian, I think we have to say it’s a virtue if it involves certain actions central to human life. And being reliable does seem basic to the better forms of human relationship. So a trustworthy person will be someone who’s reliable when she should be, in the right way, for the right reasons, for the right people, and so on. The not uncommon deficient vice would of course be untrustworthiness. But there will also be an excess: being reliable when one shouldn’t be.

What would be an example of that vice? Consider the Great Train Robbers, who reliably showed up at Sears Crossing, on the West Coast Main Line in the early morning of 7 August 1963. It would seem odd to say to Ronnie Biggs; ‘Well, pretty well everything you did that morning was wrong. But we’ll give you some moral brownie points, at least, for being reliable’. We might call this kind of reliability merely ‘descriptive’ reliability, rather than a virtue. Biggs was the kind of person his fellow conspirators could justifiably believe was likely to turn up on time.

What about institutions? Again, the natural answer is affirmative – they can have at least some virtues, including reliability. Should we understand such apparent virtues as, really, individualistic? So the reliability of, say, Marks and Spencer consists just in their employees’ being reliable? We could, but on the face of it there seems no obvious metaphysical reason to do so. The reliability of Marks and Spencer on a certain occasion might be, in some sense, constituted by the actions of certain individual employees. But the company itself is not to be identified with them, or even all of its employees, and we naturally speak of institutional agency.

Many are suspicious of institutions: governments, corporations, banks. If you believe Thomas Pogge, for example, many major pharmaceutical companies are perpetuating injustice. Peter Singer believes that the current distribution of property in the world is highly morally defective, which raises questions about the moral status of actions taking place within the context of our current institutions as a whole, nearly all of which rely heavily on the current distribution of property. Perhaps the reliability of such institutions, even when advertised as a virtue, is in fact merely descriptive – like that of the Great Train Robbers.

  • Facebook
  • Twitter
  • Reddit

13 Responses to Trust and Institutions

  • Andrews says:

    I’m no expert on Aristotle’s moral philosophy but I have two problems with ascribing virtues to institutions. Maybe you can help me solve them.

    Virtues are grounded in affective dispositions. Or so I believe. In short, it seems to me plausible that affective dispositions are required for having virtues and for explaining the “unity” there is between intentions and ways of acting — a unity I think is an essential part of the concept of a virtue. Example: what makes you generous is the fact that you bear an affective relation to other’s interests or needs and that this relation motivates you to help satisfy/alleviate them. (This explains en passant what generous actions have in common: they all are motivated by a desire the occurrence of which, and the content of which, presuppose a certain awareness to evaluative features that the virtuous person could not have (the awareness, that is) unless he can bear affective relations (i.e. emotions, feelings, etc.) to her environment.)

    So the object goes roughly like this: no virtues without affective dispositions; no affective dispositions in institutions, so no virtues in institutions.

    Institutional agency is derivative upon formally defined and publicly identifiable attributive proceduresSorry ’bout the jargon but the idea is very simple: when attributing actions to institutions the man on the street or the judge in the court relies on attributive procedures. Concretely, this means that we assume among other things that:
    (a) the institution is well-defined according to the laws of the country in which it is based given its juridical form (this concerns more private institutions than private ones, such as corporations);
    (b) the intern procedures that typically produce an action from a decision from a deliberation are well-defined according to the laws of the country and have been correctly implemented (i.e. all the people involved have carried out their role as defined in their contract).
    Usually, when (a) and (b) obtains, it’s justified to ascribe an action performed by folks working in the institution to the institution per se (i.e. we say that the action was done in the institution’s name and therefore is an action of the institution.

    Now the objection goes roughly like this: even if virtues did not necessitate affective dispositions, there would be still be no criterion for ascribing virtues to institutions since there would be no criterion for identifying the motives governing the deliberation, decision and resulting action in question, and no criterion for attributing those motives to the institution per se.

    Really, there are two sub-issues here. One is that motives can be hard to pinpoint even from the insiders, so even more difficult to pinpoint for the outsiders. Another is that even if that was not the case, both the man on the street or the judge in the court would need a principle for ascribing motives to the institution per se. But where should one look at? In the head of the councellors? of the decision-makers? of the guys that implement the decision? No matter how you slice it, it’s hard to know in principle how motives are born, how they travel and what effects they have on folks that work for the institution, and thus it is hard to integrate and aggregate them into one final, coherent, collective motive.

    There is no such problem for action; when actions are concerned, we look at the institution’s intern procedures and assuming that they are well-defined and have been correctly implemented, we check whether the intentional structure stretches back far enough, i.e. to the decision-makers. We do not need to integrate or aggregate individual, potentially variable, motives into one or into one bunch of coherent, collective motive.

    And there is no such problem either for intentions since we have to look no further than at the intentions of the decision-makes.

  • Andrews says:

    Reading my post again I notice that trustworthiness does not qualify as a virtue under a conception of virtues as essentially related to affective dispositions. Of course maybe my concept of a virtue is severaly flawed and I should replace it with a different one. Yet, should it be adequate, one might wonder what trustworthiness is after all.

    I’d say it’s a way of taking a specific set of responsibilities seriously, namely responsibilities you are bound to by means of a contract, a convention or simply other people’s expectations provided they are of the type that produce duties. Thus, as a first pass, I’d say x is trustsworthy just in case (a) x is aware of others duty-binding expectations and (b) x tries her best to fulfill those duties.

    Now of course if this is what trushworthiness is, then it is entirely compatible with being blatantly blind to any form of morality, i.e. it is compatible with not seeing the moral relevance of certain duty-binding expectations and of fulfilling them (i.e. I refrain from cheating my wife because it’s part of an agreement between us and I should not violate agreements I’ve made vs. I am aware that cheating on her would negatively impinge on her morally relevant interests given her expectations about our relationship).

    • Roger Crisp says:

      Thanks, Andrews. Yes, this is a challenge, and you probably noticed that I left out feelings in my allegedly Aristotelian account! There are clearly two ways to go. One would be attempting to ascribe affective states to institutions. That is something people do — some organizations, for example, might be described as uncaring. But perhaps like you I’m inclined to see such claims as stand-ins for claims about actions or dispositions to act, or reducible to claims about individuals. So that leaves the other route — claiming that a capacity to act is enough for certain virtues. That might be more promising when considers virtues of individuals such as justice, certain forms of which depend heavily on acting in certain ways, bringing about certain results, or carrying out certain procedures.

      • Andrews says:

        I am not sure I completely understand your suggestion, Sir Crisp. Are you interested in knowing if things such as institutions can be literally virtuous, or virtuous in some derivative sense; or if they can be trustworthy in a literal or derivative sense whether or not trustworthiness is a virtue in some sense? Or more generally, if they can applied concepts of dispositions to act in a good way, be those dispositions virtues or something else?

        If, as I feel, what you are interested in is the last question, then we have to be clear on what counts as a disposition to act, and what it means for such a disposition to be at least partly individuated in terms of the goodness of the action that manifests it.

        If, as I’ve tried to argue, we cannot trace back this form of goodness to the goodness of a motive (for not criterion for doing this is forthcoming, and might not even be coherent with a correct conception of institutions), it’s not clear at all that we indeed can individuate a putative disposition to act in a good way in terms of, among other things, the goodness of the action that manifests it.

        Any such analysis might leave us with nothing else than the goodness of the action according to a strictly consequentialist or deontological evaluation, which of course falls short of individuating any disposition manifested by or in the action. If this was the case, I would not be surprised that the special agentive structure of institutions is the main issue — i.e. because that agentive structure depends in part on conventions and not only on intentional features of its constituents (the attributive procedures mentioned in a previous post are supposed to explain this).

        So, I would like to know simply why you feel so inclined to ascribe dispositions to act in a good way to institutions. Why isn’t it enough to ascribe to them intentions, actions and possibly behaviours? I suspect that the concept of behaviour would allow you to say all you want to say about institutions without conflicting with some of the observations I’ve made. The concept is rich enough for saying, for instance, that institutions do good or bad things on a regular basis without implying that the reason for this is ultimately a disposition to act in a good way.

        Instead of saying, very akwardly, that a significant change in the institution’s behaviour entails the loss or the acquisition of a disposition of the relevant kind, you can simply say that, given that behaviours supervene on “internal” interactions, there must be some change in such interactions (maybe good people working for the institution have turned into bad guys, maybe the organisation of the institution has changed in bad way, i.e. creating loopholes in the internal procedures that make it difficult to prevent negligences, etc.).


        • Roger Crisp says:

          Thanks, Andrews. I’m interested in all those questions! I suspect part of your worry is that without a motivational background, normal moral praise and blame aren’t appropriate. But I guess I’m working with the common-sense notion of such praise and blame, and here people do seem quite happy to praise and blame institutions for what they do. And if that is the case, then we can just read back from those actions to a disposition to perform them (one which needn’t be elucidated merely in terms of subjunctive conditionals, but could involve some kind of categorical basis in e.g. the institutions capacities for making good decisions, etc).

          • Andrews says:

            I suspect part of your worry is that without a motivational background, normal moral praise and blame aren’t appropriate.

            Nope. It’s fine by me to praise or blame institutions for their actions. Doing this is merely saying that they acted wrongly — by abuse or negligence. Actions can be ascribed if there exists some appropriate procedure given the intentional structure from the decision to the implementation, and abuses and negligences can be ascribed given then intentional structure of the action and given a set of moral responsibilities.

            What’s not fine by me is to apply virtue concepts to institutions. For the reasons I’ve tried to explain, it either relies on a very thin notion of virtues (i.e. virtues without affective dispositions) or on a mistaken notion of disposition (i.e. there is no categorical basis to ground the non-affective disposition to act rightly if such dispositions depend intrapersonal processes and states) or on an inapplication notion of disposition (i.e. if we cannot identify its categorical basis since there cannot be a procedure mapping personal dispositions to act rightly onto institutional dispositions to act rightly although there exists such a procedure for actions and intentions).

  • Dave Frame says:

    Hmm… you might want to disaggregate “institutions” a bit. In some policy circles people distinguish between “organisations”, which might include governments, corporations, charities, etc, and “institutions” which are the “rules of the game” by which they play (the distinction dates to Douglass North, at least). If you separate things out in this way, I’d have thought that for institutions things like reliability, trustworthiness and so on are fairly clearly a good idea*, since generally you might expect these things to increase social capital or lower transactions costs or however else you want to put it – basically increase the scope for gains from trade by enabling people to work together. Organisations on the other hand, might be more prone to the sorts of critique outlined in the original post. i.e. there’s a difference (I think) between thinking about virtues (trust, reliability etc) as they map onto institutions (eg the rules of football) vs organisations (eg FIFA) that’s relevant to Roger’s post.

    *Though obviously if the rules are rotten then these virtues are a bit irrelevant. (But then if the rules are rotten then I don’t see why people would want to play the game, so the whole thing seems a bit moot.)

  • Roger Crisp says:

    Thanks, Dave. This is very helpful. But I guess the difference might be one of degree: some institutions (in your sense) might be unjust, even if in general it’s a good thing if institutions are trustworthy.

    • Dave Frame says:

      Hi Roger. Sure – some institutions might be unjust – imagine if the rules of golf stated that you have one shot added to your score for every year spent in Devon, or if the rules of rugby ignored tries scored by people with surnames beginning “Mc” or Mac”*, or if a country’s rules around employment stipulated different salaries for men and women, etc. But just because rules tend to be general and organisational decisions specific, I’d expect injustices in the former to be more obvious than in the latter. Company X might just happen to employ statistically unusual numbers of red-heads because of a random draw. Put very simplistically, in a probability sense you might expect to observe 5% of firms as having workforces that deviate from the norm by 2 standard deviations, just by luck. Without some expensive inquiry (that itself might raise justice-flavoured eyebrows) you’d find it hard to know whether any given firm with observed biases really was biased, or was just a statistical oddity. So I think it’s more a point that the texture of “injustice” might be quite different in the case of institutions (where general arguments about initial conditions, rules, and outcomes ought to be fairly useful) than it is in the case of organisations (where arguments about motivations and other messiness) might be more important.

      *Certainly consistent with Scotland’s recent 6 Nations campaigns…

  • Anthony Drinkwater says:

    Thanks for this interesting post, Roger
    On your first question : is trust a virtue, you write : “And being reliable does seem basic to the better forms of human relationship.”
    Isn’t being reliable also basic to other forms of human relationship : the criminal or terrorist organisation needs trust as much as (or even more than) “the better forms”. To call this merely “descriptive trust” implies (or so it seems to me) that the virtuous trust is not descriptive, which seems to me odd : there is surely a descriptive element necessary to understand what constitutes trust to prevent the term being used to mean anything whatsoever ?
    To argue that “a trustworthy person will be someone who’s reliable when she should be, in the right way, for the right reasons, for the right people, and so on” seems to me to expand the notion of trust and to bind it in with necessarily other virtues. What’s wrong with accepting that a person might have some virtues that are misused ? if you don’t accept that, aren’t you setting an impossibly high (and very difficult to define) standard for virtues ?
    We’re all imperfect, but the person devoid of any virtue at all is very rare. And isn’t there something odd in the idea that Ronnie Biggs would have been more virtuous if he had simple failed to turn up that day, ie was not trustworthy ?

  • Roger Crisp says:

    Thx, Anthony. It would indeed be odd to say that virtue-terms (when used to ascribe virtues) aren’t descriptive. So my thought was that when we say that some virtuous person is trustworthy, we’re describing them *and* evaluating them. Whereas when we say that Biggs was reliable, we’re not evaluating him. If we did want to evaluate him, on the Aristotelian view, we have to say, rather oddly, that he was ‘too reliable’ — he shouldn’t have shown up. (This is why Biggs would have been, in one respect, better if he hadn’t turned up: he would have failed to do something excessively reliable.) On the relation to other virtues. Aristotle discusses this and is rather strict about it: if you have one virtue, then you must have the lot. His reason is a good one: if you have a vice, it may interfere with the operation of what appeared to be a virtue. Imagine I appear to be generous, but I am also cowardly. Often, I give money in the right way, and so on. But on some occasion when courage is called for at the same time as generosity (for example, I’m afraid of being laughed at for giving to some charity), I fail. For Aristotle, virtue requires success, not just good intentions. But Aristotle was also rather a sensible person. So I’m sure he would have allowed that what he is describing is an ideal, and that most, if not all, of us are at some distance from that ideal. What we have to aim for is getting as close to it as possible.

    • Anthony Drinkwater says:

      Thanks for your reply, Roger.
      I bow to your knowledge of Aristotle, but if for him “virtue requires success, not just good intentions”, I would reply that this excludes a very large number of virtuous acts : to cite just one example, courageous resistance against the rise of fascism in 1930’s Germany. Agreed, those acts might be considered even more virtuous had they been successful, but as I wrote above, this seems to set the bar impossibly high. Even worse, it could be argued that it discourages virtue : “I’m not certain of succeeding, better accept things as they are”….

      To come back to Biggs, the description of his reliability as being excessive still seems to me perverse. Agreed, the cause to which he was so trustworthy was far from virtuous but in the context of his group he exemplified a virtue.
      Perhaps, however, this thought should lead us to exclude the language of morality from such groups – or, to follow up on your second question, from institutions. Within any institution, there is a code of conduct, and quasi-moral judgements can be made, or certain “virtues” encouraged. But I wonder whether this isn’t some form of category mistake – as (if I understand him correctly) Comte-Sponville argues in criticising those who seek the “moralisation” of capitalism.

  • Roger Crisp says:

    Good points, Anthony. I spoke too loosely. The criterion for an action’s being virtuous is that a virtuous person would do it in the same circumstances, so the resisters actions were virtuous. But, unfortunately for them, what they achieved wasn’t as noble as it would have been had they defeated the Nazis. This brings in moral luck, of course, but Aristotle seems less concerned about that than, say, Kant. That criterion also helps to explain why Biggs’s reliability is excessive: a virtuous person wouldn’t have shown up (or, rather, they would have advised not showing up, since of course they’d never have got into the position Biggs was in that morning). I think this is a counter-intuitive conception of virtue, but it’s got enough going for it not to be perverse, I’d suggest.


Subscribe Via Email