It’s still March, but only just so we’ll keep the comments open well into April.
This next reading, suggested by Sean, is from the Special Issue on Virtue Ethics and Role Ethics mentioned in an earlier post.
Dare, Tim (2017) Robust Role-Obligation: How Do Roles Make a Moral Difference? in Journal of Value Inquiry, 50(4): 703-19.
Summary of the piece to follow soon in the comments section, but the topic in the first paragraph is instantly gripping:
How could it be right, Macaulay asked, ‘‘that a man should, with a wig on his head and a band around his neck, do for a guinea what, without these appendages, he would think it wicked and infamous to do for an empire?’’ How can a role make a moral difference? (p. 703)
ROLE ETHICS NETWORK 
Firstly to declare an interest: one reason, among others, for my picking and liking this paper is that for me it raises the normative question of how institutions should operate or be ‘designed’ with respect to the roles and role-obligations they determine. For me this is a tricky and fascinating question, encompassing as it does some considerations reflecting ‘general’ morality (social institutions should almost by definition be pro-social and promote some good(s)); but also the (proper?) function of an institution (social institutions are particular- not just any social institution should promote morality or the good in any old way: charities are not police forces and vice versa). So, were we to take the Rawlsian model outlined by in Tim Dare’s section 6, my basic question would be the criteria by which the reformer should externally criticize some institution or practice – how she would justify claims that its rules should change. A general moral consideration or principle such as maximisation of the good, preference satisfaction, Eudaimonia, or adherence to rights will get us so far, until we reach the specificity of say, ‘Army’ as distinct from ‘ University’. Whilst any of the foregoing principles might constrain or inform how each different institution is constituted or sustained, clearly this or that institution must do some very different things in order to be this or that kind of institution. At this point it would seem intuitive to appeal to function(s) and purpose(s) and in a way that avoids some of the problems of functional explanations in biology. (For it looks like a social institution is more like a social artefact and tool than it is like a living organism organ. More like a hammer than a heart.) But also at this point, it is not always clear what some kinds of institution is or should be doing, where in an instance of an institution, functional accounts of design, use, and good consequences can each be plausible but come into conflict (University being one example, I think).
On a different point, I remain sceptical of approaches that seeks to explain role-obligations in terms of more general moral obligations. The Luban quote, in Dare p.703 and p.713 is interesting here:
‘‘if a lawyer is permitted to puff, bluff or threaten on certain occasions this is … because in such circumstances anyone would be permitted to do these things’’.
It’s always seemed to me that this could be read as true but trivial; or as question-begging, such that it tells against one important way of pushing the approach just mentioned. On the first reading- the ‘circumstances’ that would permit ‘anyone’ to puff, bluff, etc. would include the circumstance of ‘being a lawyer’. This cannot be what Luban means, as it would amount to the unremarkable claim that anyone who were a lawyer would be permitted to do what a lawyer is morally permitted to do. The alternative reading is that there could be circumstances, relevantly similar to the lawyer’s, in which the non-lawyer would be permitted to puff and bluff etc. within the constraints of general morality. This version of the claim looks to beg the question in favour of the direct route or possibly the indirect route too, for it assumes that the lawyer’s obligation to puff and bluff etc. is not generated exclusively by their role. At the very least that assumption looks hard to sustain without implausibly robbing roles of the specificity which gives rise to the problem of role-ethics in the first place. That specificity is salient as long as an answer to the question ‘why do you puff, bluff etc.?’ yields the correct answer ‘well, because I am a lawyer’. For whilst there may be all sorts of circumstances in which ‘anyone’ may be morally permitted to dissimilate, only qua lawyer will a person dissimulate for reasons to do with e.g. serving their client’s interests, or Queens Counsel or Crown Prosecution (in the UK), or whatever. And those specific reasons for lawyers’ actions of puffing and bluffing etc. obtain by virtue of the legal institution(s) of which their roles are part. To the extent that they act for these reasons as lawyers, they are acting on something other than generalizable considerations of say, prudence or means/end reasoning, which could apply equally to non-lawyers in roles and circumstances outside the legal context.
Thinking back to the Macaulay example, it seems hard to think how the Judge’s duty to have someone imprisoned for 25 years could be explained by or assimilated to general morality with the ‘anyone in the circumstances’ approach. We may be able to justify in quite general moral terms the sentence, the institution of punishment, or the practice of incarceration. But this does not sufficiently describe the circumstances in which ‘anyone’ could be morally justified in passing the sentence. In order to be in the relevantly similar circumstances the ‘anyone’ would need to bear the very special authority and responsibility of a judge to prosecute the law (otherwise one would likely be doing the opposite:, i.e. committing the crime of false imprisonment).
Hope that makes some sense! Thanks everyone.
Sean, thanks for picking this, great paper. First, the summary I promised:
Tim Dare’s topic is the justification of role based obligation. After setting out what roles are (section 1) he argues THAT we have role obligations (section 2) before considering the tougher question of HOW we have them.
The main moves are made in sections 4 to 6, where he runs through three possible answers to the ‘how’ question, advocating for the third. They differ over how role obligations are tied to ‘ordinary morality’, i.e. morality that is not overtly tied to role occupancy. A recurrent example is a lawyer’s obligation to act in the client’s interests.
First, the direct account: acting on a role obligation is simply a direct application of ordinary morality. He illustrates this with the assumption of act (aka ‘direct’) utilitarianism serving as the ordinary morality: acting on a role obligation is required because it promotes utility. Good objections to this are on p. 711.
The second account is where the connection is indirect: e.g. our role obligations are comparable to the rules that would be endorsed by a rule (aka ‘indirect’) utilitarianism. His objections to this rehearse familiar dilemmas confronting rule utilitarians.
The third position is even more indirect. He calls it the ‘clean break’ model. Here, social and institutional roles are justified using criteria drawn from ordinary moralilty but, once in place, no reference to morality is needed to justify their bindingness. He draws a comparison with Rawls’s two levels of justification, OF institutions and WITHIN institutions.
And now some comments…
Sean, on your first observation, yes, any justification of roles in terms of institutions does seem to raise parallel questions at a higher level, viz., what is the distinctive role of the institution, and why so. And on your second, I was having related thoughts while puzzling over what ‘ordinary morality’ is if it excludes the very ordinary phenomenon of being in the circumstance of occupying a role. In my summary above I went for ‘morality that is not overtly tied to role occupancy’, which is closer, perhaps, to ‘background’ morality (an alternative label offered by Dare in his first paragraph, though it falls away later).
My two comments:
I enjoyed the fresh way in which Dare sets up the topic (direct vs indirect vs clean break). It differs from the more familiar contractual vs non-contractual contrast. I also appreciated the discussion of what a role is in section 2. That said, both my comments are on these aspects.
On the way Dare sets it up, I wander if utilitarianism is serving as less as a mere illustration of the distinction between direct and indirect strategies and more as a required assumption for the argument to work (pace p. 714). If the background ‘ordinary morality’ were, say, contract-based, the direct and possibly the indirect positions he rejects seem to have more going for them. (Which is not to say that the ‘levels’ distinction at work in his clean break account isn’t useful here too.) Also, in passing: Brad Hooker has an impressive go at getting around the problems with rule consequentialism surveyed here, in chapter 4 of Ideal Code, Real World (2002). Interestingly, though it is not about roles, Hooker’s solution in some respects prefigures Dare’s clean-break model.
On what a role is, Dare calls it ‘a distinctive set of normative statuses – rights, duties, powers, permissions, and the like – that attach to a role-occupant by virtue of her occupation of that role’. That may be an unhelpfully narrow range of normative statuses. That is, we might twin these norms, which apply to a person insofar as they are IN the role, with a further set of ‘normative statuses’ that deal with ENTRY into (or exit from) the role. So, legitimate entrance to the role of monarch is a matter of hereditary status, legitimate exit from the role of police officer is handing in your notice or being fired. I suspect this slightly richer normative ontology for roles would complement the ‘robust roles’ account Dare prefers, not undermine it. At a couple of places Dare considers the question of whether poor performance of obligations can exclude you from the role, e.g. of father. That is better framed as the question of whether failing to meet the obligations of a role is sometimes an exit condition. (It certainly isn’t always so – otherwise we could always just walk away from the role and hence the obligations by just not performing them.)
A recurring feature of discussions of role ethics is a misleading treatment of the concept of role. It is commonly treated in a way that makes little or no distinction between the concept of role and that of function or job description. Tim Dare’s article is one of the exceptions.
He explicitly distinguishes between the institutionally specified function of, for example, a lawyer and the social expectations of occupants of the role of lawyer. As he says, it is the social expectations that constitute the role, expectations about the normative commitments the occupant of the role will meet (p. 706). Note though that role occupants must often engage with the expectations of more than one constituency, and that different constituencies may have conflicting expectations. For example, shareholders in a company may have expectations of its directors that conflict with those of its employees. That is why in order to be able to assess the moral significance of an act, it is often necessary to know both in which role someone acted (as Dare points out, p. 710), and in whose interests. (Incidentally, there is much more to the resources of role theory in sociology that are applicable to role ethics than is represented in Dare’s article).
Like Alex, I am puzzled by the distinction between ‘ordinary morality’ and role-specific morality. If role-specific normative expectations are those which agents ordinarily take into account when interacting, what is the separate ‘ordinary morality’ that is to be invoked as a basis for evaluating roles and the conduct of their occupants? Perhaps ordinary morality is best understood as referring to normative expectations that apply to most social roles, such as the expectation that those in a constituency affected by a role (for example, a lecturer’s students) will all be treated fairly. It seems to be a failure to meet this kind of cross-role expectation (rather than the requirements of a normative theory) that Dare has in mind when he refers to the lawyer who relies on a technicality to save his guilty client from punishment (p.718). On this view, the answer to Sean’s question, ‘how should a change in an institution’s rules be justified’ is, ‘The present rules allow practices that are incompatible with this society’s norms of fairness, promise-keeping, truthfulness or whatever norm it is that the present rules permit to be flouted’.
John, I’m probably not alone in being interested to know more about role theory. Is there anywhere I/we non-sociologists should be looking (overviews, classic statements, etc.)?
Thanks for the point you make about the competing pressures on the specification of a role. I wonder if the institutionally specified function can sometimes BE a role, though, without mangling the word too much. There seems to be a single notion, the role of teacher, but with alternative sources we can look to if we want to understand what the norms are that attach to the role. We can be more specific as needs be (the role as given in the job spec, vs the public role of teacher, etc.), especially if we want to address the kinds of pressures that can pull occupants in different directions.
Nice suggestion for interpreting ‘ordinary morality’ but I don’t know that it quite does the trick. I’m thinking here of Hardimon’s point that we can take role morality seriously without assuming that all morality is role-based. So ordinary morality is not so much morality that applies to all roles, as morality that is not tied to roles in any way. E.g. with ‘Keep your promises’, there is no implicit ‘In all roles, …’. On top of which, there is an understanding that in some roles, not keeping promises is required (e.g. spies, while spying. Unless we say that promises made while spying are not real promises, but that wouldn’t work for the ordinary-morality norm: don’t spy on people!)
Alex, first a response to your request for references.
Not all sociologists subscribe to it, but I believe the dominant conception of role in sociology and the one most relevant to role ethics is the one developed in a branch of the discipline known as ‘symbolic interactionism’ (SI). The foundations for this branch were laid by the American philosopher George Herbert Mead, one of the founders of American pragmatism. For a useful account of the derivation of SI from Mead’s ideas, see the following article by Sheldon Stryker: Stryker, S. 2008. From Mead to structural symbolic interactionism and beyond, Annual Review of Sociology 2008 34: 15-31. ln addition, or as an alternative for those who prefer an impressive ethnographic treatment of the role concept that acknowledges the influence of Mead, I suggest Goffman’s famous book: Goffman, Erving. 1959 [1990]. The Presentation of Self in Everyday Life. London: Penguin.
Stryker himself has made important theoretical and empirical contributions to SI, and in particular to the specialism within it that he designated ‘Identity theory’, which makes central use of the concept of role. ‘Social behavior is specified by taking “role choice” – the opting by persons to meet expectations of one role rather than another – as that which the theory seeks to explain’ (p. 20). The article, which cites numerous empirical studies in support of the theory, also suggests a number of ways in which it is related to other theories within and beyond sociology.
A derivation of Stryker’s version of identity theory (there are now other versions) has been applied to the concept of ‘moral identity’. Jan Stets and Michael Carter review the background to this development and describe an application: Stets, Jan. E. and Michael J Carter. 2012. A theory of the self for the sociology of morality, American Sociological Review 77(1): 120-140. However, the article is not directly pertinent to the issue of role ethics. An article that is directly pertinent is one by Leavitt and colleagues: Leavitt, Keith et al. 2012. Different hats, different obligations: Plural occupational identities and situated moral judgements, Academy of Management Journal l55(6): 1316-1333. It is not explicitly based on symbolic interactionism or identity theory but, in my view, in some respects it is very illuminating about the significance of both for role ethics. The authors’ citations make clear the proximate provenance of their ideas. I recommend reading this article first.
Now to the points you raise about my earlier comments.
I do think it is important to distinguish function from role. The social role of teacher (I shall assume you mean school teacher) can only be understood in relation to partner or counterpart roles, in particular those of pupil, parent, and head. A description of the teacher’s role is a description of the expectations of the teacher’s conduct shared by teacher and role partners. A description of the teacher’s function is usually a specification of what an institution expects of occupants of the role, or perhaps what some other body (such as the DoE) believes ought to be expected of teachers. It usually assumes the existence of the role and describes performance criteria that occupants are required to meet.
The second issue you raise – the relation between ordinary morality and role morality – is complex. I agree it is not sufficient to say that ordinary morality refers to normative expectations that apply to most social roles, though I think there are some norms, such as truth-telling, that we believe ought to apply to most. As you note, Hardimon’s alternative is that there are duties that are independent of our roles, which he says ‘apply to people generally’. But would he not include promise-keeping in this set, and would he not allow the exceptions you mention as special cases? As he says, ‘conflicts between the obligations of our roles and our general duties are bound to arise’ (p. 341). (Incidentally, I have a problem understanding his aversion to the idea that signing on for a contractual role entails promising to carry out its duties, and his preference for ‘agreeing to carry out its tasks’ (p. 356). Is agreeing not equivalent to promising here?)
It seems to me that a more fundamental problem is that the category ‘general’ (or ‘ordinary’, or Leavitt et al’s alternative, ‘universal’) obligation is too vague to be a useful contrast with ‘role-specific’. It lumps together obligations that are usually benign in their consequences, such as promise-keeping, with those that also apply generally but often have seriously malign consequences, such as those that are often reflected in the rights defended by dominant political philosophies.