One of the main themes of my research is normativity. In this post, I want to provide something of a primer on normativity. Hopefully, this will go someway towards explaining why this is an issue we should take seriously and why normativity acts as a core concept for philosophers like myself. The immediate prompt to this post is the ongoing discussions between Levi and Pete, which keep running aground whenever normativity arises. Pete has set out the pertinent issues a number of times, but I thought it would be useful to approach them from my own perspective too, since a bit of triangulation might help us get a clearer sense of what is being said.
In short, normative issues concern correctness. Obviously, there’s many different senses of correctness and many different ways in which it arises as an issue. This helps to explain the tendency for normativity to to become a monolithic topic, which can seem to suck all philosophical light towards it like some supermassive body. One crucial distinction here though is between first-order normative inquiry and metanormative inquiry. I’ll explicate this in relation to some philosophically familiar topics.
In ethics, we are often asking what we ought to do — what would be good in the way of practical action. Should I give up philosophy and train to be a psychiatrist? Was it too callous to have decided not to meet your friend because you felt too drained to listen to their problems? Is a society without socialised medicine thereby unjust? These are first-order ethical questions, and they are normative because they are oriented by the question of how it is correct to act.
But we can also ask (as metaethicists do) about the metaphysics, epistemology and semantics involved. What does it mean to say I should do something? What is it for an action to be good? Is it possible to know whether I did the right thing? How? What, if anything, separates ethical demands from those of social etiquette? These are metanormative questions, also oriented by the notion of correctness in action, but which try to uncover what this notion of correctness amounts to. At its limit, it might even conclude that there is no sense in which actions are appropriate or inappropriate — they just are.
Similar distinctions between the first-order normative and the metanormative arise in other areas too. Another familiar example is deliberation about empirical facts and epistemology. Again, there are first-order matters: What is that animal? Would it be true to believe it is a chaffinich? Are physicists justified in believing that the Higgs boson exists? Do these problematic observations imply that the standard model of particle physics must be revised? These are normative matters, but only in the thin sense that they, arguably like all inquiry, are oriented by the question of what is good in the way of belief. The usual (and right) answer is that we should believe the truth, but we can also assess correctness along a number of other dimensions: justification, entitlement, inference, probabilifaction, consistency, coherence, accuracy, and so on.
One of the functions of epistemological inquiry is to examine the status of these first-order matters. What does it mean to say that we are not entitled to believe that P, even though P might turn out to be true? What is it for our perceptual beliefs to be justified? What is problemtic about inconsistency in belief? These questions have a metanormative dimension insofar as they abstract themselves from the immediate issue of what beliefs are good or bad along various dimensions in order to ask what any such assesments consist in.
The tentacles of normativity reach far and wide beyond these two examples too. We find them in the philosophy of language, where many have argued that meaning is normative insofar as terms have a correct and incorrect usage, and where the task is to flesh out this normative dimension to linguistic practice. The same goes for concepts and their conditions of application (alongside the murky notion of ‘mental content’ more generally) in the philosophy of mind. Aesthetics can be thought to be a discipline with a metanormative aspect too, especially when beauty and art are notions entangled up with endorsement, which a theory of aesthetic judgement may need confront. The theory of agency is another locus for normative issues, especially insofar many people (myself included) think that there is a distinctive logical form to action-explanation which needs to be articulated in relation to reasons in addition to brute causality. The list goes on… I certainly do not mean to endorse all these projects — I think many of them are misguided ventures — but merely to point to some of ways in which metanormative matters appear within contemporary philosophy.
Hopefully, it will now be clear that we can pitch normative inquiry at different levels. One worry that Levi has expressed is moralism: doesn’t an obsession with normativity lead us to fixate on judging people, weeding the unworthy from the worthy, and seeking to police people’s activity? With our distinction between first-order and metanormative inquiry in hand, we can respond by saying that focusing upon the normative does not necessarily betray a desire to be the judge of what is right and wrong. Instead, for philosophers interested in normativity, there is more often an attempt to understand the conditions under which assesments of correctness can succesfully be made at all and what the upshot of such assessments is. One way of articulating this is to say that we want to understand ‘the force of the better reason.’ I take this to be quite some distance from the right-wing obsession with ‘values’ that seems to be incensing Levi.
Another worry of Levi’s may be more pressing though: turning to norms can seem to be a turning away from the world. We can think of this in terms of a kind of ‘cognitive ascent’ which proceeds in two steps. First, there is a distinction between talking about the world and talking about our orientation towards the world. On one view, there are plenty of philosophically interesting features of the world to talk about — the structure of objects, the nature of causality, the individuation of social actors — and we should just get on and do that. Talking about norms is not talking about the world in this way: either it is talking about how we should talk about the world, and not plainly talking about the world; or it is talking about how we should act in the world, which is both tediously anthropocentric and still not talking about the great outdoors. Furthermore, there is a second ascent here too, since there is not only first-order normative inquiry but metanormative inquiry. It is either talking about how we should talk about how we should talk about the world, or it is talking about how we should talk about we should act in the world. The reassuring cruch of reality is thus even further from being underfoot.
In response, there a few things to be said. Firstly, I do take discussions of normativity to be about the world directly — they are no mere escape from it. I will be brief here because in a way this does not meet Levi’s worries head on. Nevertheless, whilst it may be démodé in some quarters to be concerned with rational agency as a distinctive phenomenon, I think it is a pressing descriptive task for which marshalling the vocabulary of normativity is essential. (See Pippin’s excellent piece and the surrounding discussion for more on this.) Beyond this, normativity is itself part of the world and is threaded throughout countless human practices. When McDowell gives us an account of virtue alongside the capacities, social practices and formative processes needed to make sense of our responsiveness to it, or when Brandom excavates the structure of the practice of giving and asking for reasons, they are talking about real phenomena which need to be elucidated. It is perhaps worth stressing that there needs to be nothing a priori about these normative investigations, even if figures like Habermas (and Pete for that matter) sometimes try to derive minimal rational norms in a quasi-transcendental fashion.
It is the intersection of metaphysics and normativity that seems to be worrying Levi here though. I am less enamored of constructive metaphysics than Levi or Pete, but I think the latter does a brilliant job of demonstrating the methodological role that metanormative inquiry should have within any such metaphysics. I can’t do better than his own exploration of these issues in this essay.
Finally, I want to underscore the possibility of detatching normativity from deontology. The latter understands normativity in terms of necessity imposed through legislation. In Kant’s practical philosophy, this is bound up with a system of duties and rights which takes a rigorist form that many people find both implausible and distateful. Indeed, Lukács tries to show how the Kantian subject, delineated through reference to an abstract set of duties, is the one presupposed by capitalism, being a reactionary ideological symptom of modern forms of exchange.
My own understanding of normativity, both in the practical and theoretical realm, has a more Aristotelian character to it. I do not take the reasons we have to stem from legislation but rather from the concrete situations we face as agents. Nor do I typically articulate rational requirements in terms of necessities imposed on us by reason. Instead, I help myself to the less austere vocabulary of the good as well as the right, and try to extend the concern with what one can do to who one should be as well. In addition to my own, there are innumerable other ways of approaching normativity too. It is important that this point be made in order to correct the assumption that recourse to normative themes is meant to bolster to some quasi-Kantian project, when this is certainly not always the case.
I could continue but that’s more than enough for now.
Grundlegung 
Very well put, all of it. I really should address why I take a strictly deontological approach to normativity at some point more head on, insofar as there are definitely cases that are tricky for this kind of obligation-centred approach. The idea of gradations of value for instance (i.e., things being more or less good, rather than just good/bad). I think these can be incorporated within my approach and thus dissolve the philosophy of normativity in general into deontology more specifically, but it’s important to have the debate.
Anyway, bed for me.
This issue of the normative role in language use is critical from my perspective; all other normative enquiries depend upon it, especially in ethics. To my mind, a lot of the speculative metaphysical enquiries going on today have not been able to offer a compelling argument for why we should believe that what they are doing constitutes untangling reality and not untangling notions/words. The latter activity might even be the more useful one.
What you fall short of saying here is that an interest in the normative places the social dimension of existence centrally. Philosophy all too easily disappears up the drainpipe of introspection, but for ethics in particular this move is utterly fatal. Morality is constituted between people, and normativity is a crucuial part of that story.
I look forward to reading where your research takes you.
Best wishes!
Hi Chris,
Thanks for your comment and sorry about the delay in replying.
In a way, I think both of your comments raise a similar issue.
I agree that understanding normativity requires us to have a good grasp on the linguistic role of normative terms and the social context in which these terms are legitimately deployed. (It’s easy to say this, of course, but another to have it animate your thought!)
However, I think we should resist at least some moves that superficially look similar to such a recognition but which go further in thinking that the only legitimate notion of normativity is one that merely describes an expressive function (e.g. articulating psychological approval or disapproval of some actions) or a system of exerting social pressure and impsoing behavioural regularities.
In other words, I think we should separate out a methodological from a constitutive claim. Thus, I accept the following:
(1) A necessary condition of understanding normativity is understanding linguistic attributions of normative properties. If we want to understand what it means to have a reason to do something, then we need to understand what it is to say that someone has such a reason — to ascribe this reason.
This is a good old pragmatist maxim: one way to get to grips with a puzzling phenomena is to understand the linguistic conditions for using the term which is meant to pick it out. If you want to know what measurement is then start by stepping back and looking at what we call measuring, say.
I also accept this claim:
(2) A necessary condition of understanding normativity is understanding the social context in which appeals to normativity are made.
If we were not communicative mutually-dependent finite creatures with often competiting interests, then much of the point of normative discourse would evaporate. Sociality is part of the life-blood of normativity, especially when it comes to articulating thick concepts like honesty, loyalty and shame, whose use and conditions of application are guided by inextricably social factors, such as interpersonal attitudes and social identities.
Even though I accept these two main claims, I am wary of making the constitutive leap, which would identify normativity with the role it plays in language and social life. In a Brandomian idiom, I think we are dealing with cases of sense-dependence but not reference-dependence, at least in many cases. That is, there is a semantic but not fully ontological dependence — it is explanatory rather than constitutive.
This stems from my realism about normativity, which tries to show how human reason can be receptive to reasons which the world can provide rather than simply reasons that we fashion ourselves through our passions, interests and goals. Naturally, that’s a long, thorny and controversial topic, so I won’t jump into it here…
Tom: many thanks for your reply, and also for flagging it at my blog (I might well have missed it otherwise since I’ve been away!) I’ve subscribed to follow-up comments by email now to avoid any future issue.
I take your points here, and am very open to the kind of distinctions you are trying to make (I particularly recognise the distinction between the unavoidable explanatory dependence and the not-inevitable constitutive dependence!)
Reading between the lines, you might be suspecting that I am pushing some kind of anti-realist agenda – I get this a lot, because of the way I phrase some of my challenges, but I am closer in spirit to non-foundationalism than an anti-realism.
I’m not denying objective reality, but I am wary of claims to get at it absolutely, if you see what I mean. Like so many people, I’m sailing between the Scylla of premature certainty and the Charybdis of empty relativism. I believe not only that there is a viable path here, but that it might in fact be far wider and more easily traversed than has often been suspected. 🙂
Incidentally, have you read Julius Kovesi’s “Moral Notions” (1967)? It might be of interest to you.
I’m particularly interested in your work because, unlike a lot of what passes for modern realism, you have practical goals – what you are exploring could have direct value in mediating the ethical disputes of our modern plural societies. I look forward to reading more of your work.
Best wishes!
Chris.
Pingback: Wolfendale on Onticology « Larval Subjects .
Tom,
I think it’s worthwhile drawing a distinction between regulative and constitutive rules. When people like Brandom and McDowell talk about norms, particularly in relation to semantic and epistemological issues, I read them as concerned with constitutive rules (although they don’t call them that). This kind of normativity is different in important ways from the kind of norms you allude to in your discussion of first-order normative inquiry.
In my opinion, the so-called “anti-normativists” like Hattiangadi and Glüer have convincingly shown that semantic and epistemological norms cannot be treated as regulative or prescriptive. But at the same time, I also think they are wrong to read Brandom, for instance, as invoking a regulative notion of normativity (as opposed to a constitutive one). I suppose part of the problem here is that the discussion of constitutive rules hasn’t evolved much beyond Searle’s account, which suffers from important limitations. But Brandom’s talk of “norms implicit in practice” seems to me very much on the constitutive rather than regulative side of the distinction.
Also, you sneak in a distinction between rational and practical norms without making it clear what the significance of the distinction is. I believe this is important with respect to extending the idea of the normative beyond practical matters. Habermas, for instance, criticizes Brandom precisely for failing to adequately draw this distinction. And when you criticize Habermas for attempting a “quasi-transcendental” derivation of rational norms, Habermas would respond by emphasizing that this is possible only in relation to certain *rational* norms, and not practical norms (see his disagreement with Apel on this question). Once you take into account this restriction in scope, his project appears somewhat less controversial.
Pingback: Thinking, if not yet doing, normativity