NORMATIVITY FOR SOCIAL SCIENCES The expressivist’s recipe1

Normativity is a trending topic in philosophy. It could be claimed, in fact, that there is no sub-domain inside philosophy where normativity is left unmentioned. But things were different some time ago, of course. And things look different from other disciplines studying human thought and action. The merit of Stephen Turner’s rich and rewarding book is to bring into focus a different view on the explanatory importance of normativity for the study of some basic human capacities. For those who, like myself, are interested in normativity from the philosophical side there is a lot to learn from Explaining the Normative (ETN hereafter). There is a small lesson about what sociologists and social scientists think about the normative –about how they construct this special concept or about the sense in which they think that such a concept could be reinterpreted. And there is a big lesson, a framework for thinking about normativity. The basic aim of my comment is to supplement Tuner’s general framework about normativity with some important notes from expressivism about normativity. I will argue that if social scientists are looking for a philosophical theory of normativity, expressivism is the place to put their money.


the proBlem of normAtivity in etn
The problem of normativity, according to Turner, can be formulated by means of two different questions.The first one is about what is the explanation of normativity or normative facts.This is The Source Problem.The second one is about what normative facts explain -if anything.This is The Explanatory Problem.Turner summarizes this distinction in the following way: These are problems about the explanation of the normative.But there are questions on the other end, about the norms or normative facts that normativity consists in, and what explanatory work these facts do.(p. 3) This way of introducing the discussion about normativity raises some preliminary worries for the reader.There is a concern about the relative relevance of these formulations for both Turner's main thesis in ETN and for the general discussion in several domains around normativity.Which of the above formulations is at the core of ETN? Are they equally important when we look at the different places where the problem of normativity has been relevant in recent times?There is a popular line of response to the latter question that imposes by itself.It goes as follows.The Source Problem is the problem of normativity as it is formulated inside Philosophy.People like Robert Brandom, Christine Korsgaard, Ruth Millikan or Onora O'Neill would be concerned with this issue.The Explanatory Problem, on the contrary, encapsulates the basic concerns of those interested in normativity from the side of social sciences.
Although the above divide is initially appealing, things are not so neat when we attend to what philosophers and social scientist do.There are philosophers who are explicitly interested in explanatory issues -back in the seventies, Gilbert Harman and Nicholas Sturgeon started a discussion about explanation inside meta-ethics2 .And there is a long tradition of social thinkers interested in offering source-explanations of normativity -Hobbes and Smith are perfect candidates for this role.Thus a first question for Turner to answer is how the above questions are precisely connected in ETN.
My guess is that for Turner the relevant formulation of the problem of normativity is the one expressed by The Explanatory Problem.At this level, the problem of normativity amounts, in a nutshell, to the explanatory irreducibility of some normative facts that are at the core of some phenomena (meanings, beliefs, legal obligations, etc.).There are some things in need of an explanation, but when we offer a description of these things in causal or dispositional terms something significant is left out.For normativists, the significant thing left out is normativity (p.10).
But normativists are not only pointing to a gap in some explanations.They are also positing 'special and puzzling objects' to fill such a gap: ...self-authorising principles, rules, understood as tacit rules behind the explicit rules we might invent or propose; mathematical objects... objective values; dictates without dictators, the dictates of reason itself; performative utterances which have the property of being able to create a norm . . .(Turner,p. 15) These objects, Turner argues, are queer and mysterious.But if so, if they are things which are 'not realized in the physical world' (p.15), why do we speak as if these things were true or real?The answer, according to Turner, is that these objects are part of 'Good Bad Theories' (ETN,.
Turner introduces Good Bad Theories while highlighting the cultural variability of some commonsense concepts such as 'truth' or 'intention'.He writes: All of the diverse folk notions mentioned here ....have two relevant features: they are taken for granted, believed in, accepted, subscribed to, or used, by people, in particular, different, social settings.They work in those social settings to enable the participants to interact with each other, understand each other, and co-ordinate their conduct.None of them is "true" in the sense of being scientifically true.(Turner,p. 42 ) This is a well-known movement, at least in mainstream meta-ethics.We speak as if there were objective values, acts or situations that give us non-categorical reasons to do (or to refrain from doing) certain things.The problem with this discourse, however, is that when we check reality there are simply no such kind of entities.Objective moral facts, the argument goes, are queer, and not part of the usual furniture of the world3 .
It is an open question whether Turner would accept the line of argument presented above.In his reconstruction, it seems as if cultural diversity is what drives the introduction of Good Bad Theories to make sense of our commonsense normative talk.Here I will assume that, regardless of whether you are an error-theorist (Mackie, Joyce) or someone who is well aware of the pragmatic and contextually dependent function associated to some basic terms (intention, belief, or truth), a common objection can be raised against any strategy aiming to accommodate normative talk by appealing to Good Bad Theories.
The common and well-known objection is that maybe normative talk is not in the business of matching or representing an external reality, as those who propose an explanation in terms of Good Bad Theories seem to presuppose4 .It is only when we assume such a thing -along with a commitment with a certain variety of naturalism-that we are tempted to claim that our normative talk is somehow deficient.At several places Turner writes as if he were committed to this chain of assumptions (p. 190-191).Thus another question for Turner to answer is whether these assumptions (a representational view on normative discourse plus a variety of narrow naturalism) are in his bag.

expressivism. A metA-normAtive frAmework for the sociAl sciences?
Attributing massive error to our normative talk goes against the surface grammar of normative discourse.Otherwise, affirming that our normative talk is systematically false is in tension with those cases where we confidently say things like 'It is true that torturing babies is wrong'.I suspect that Turner is somehow aware of this tension between what we think we mean and the reconstruction offered by some philosophers.At the end of ETN, for instance, he writes: Normativism is defined by this thought: that we must accept an anomaly, something beyond normal explanation, that corresponds to normative language ...Is there a way out of this reasoning that is consistent with the use of normative language, and that allows for the possibility of philosophy as a discipline concerned with the normative?Can we have normative talk without normativity?(Turner,p. 194,emphasis added).
Turner is asking here if we could find a meta-normative framework that would allow us to talk as we do about what we ought to do or to believe without presupposing normative facts.As I noted above, his solution could be reformulated as a variety of error theory, in the sense favored by John Mackie.My point, on the contrary, is that Turner should be asking if we can find a meta-normative framework that would allow us to talk as we do about normative issues without assuming normative facts but also without presupposing that at the very end our normative talk is massively mistaken.Is there any such framework?
The answer is yes.Expressivism about normativity could do the trick at this level.In its current form5 , expressivism is fully committed with three general claims.First, it assumes that naturalism, broadly understood6 , is the right metaphysical commitment to locate the study of our normative commitments.Second, it accepts that any second-order approach to normativity should accommodate the surface grammar where normative language is located.It must explain, in sum, why we say things like 'It is true that torturing babies is wrong' and why we say such things in full confidence.In sum, and here goes the third commitment, it must avoid any error-based explanation of our normative talk and normative thought7 .
The above commitment has proven quite fruitful to understand the status of our normative opinions.As a result of it, expressivism is now seen as one of the most influential meta-ethical approaches 8 .Turner, however, does not pay any explicit attention to expres-sivism.There is no single reference to Blackburn or Gibbard in ETN, and we cannot even find the usual and dismissive attack to non-cognitivism; an earlier variety of expressivism defended by Alfred Ayer at the beginning of the past century9 .Despite this fact, there are some places where Turner writes as if a rapprochement between his favored framework and expressivism's own account were possible.In the remainder of this comment I will summarize some of these places.My intention here is constructive.I would like to know what Turner thinks of locating his empathy-based account of normativity inside the larger picture offered by expressivism.
To begin with, Turner is in full agreement with expressivism on the proper level and focus of any explanatory framework about normativity.For a long time, a standard division between first and second-order questions about normativity has been assumed in Philosophy.According to this divide, foundational issues about normativity are at the second-order level.This means that you can bracket any substantive or first-order normative commitment while answering foundational questions, not lacking any relevant information by such bracketing.
Keeping this divide in mind, what normativists are doing is simply muddling two levels of description in their explanations of normativity.They claim, in sum, that any account of normative phenomena and normative talk must include, at the very end, a reference to the first-order normative judgments of those who participate in the normative practice10 .In explaining the force of a legal system, for instance, an appeal to what we consider a valid legal system is somehow ineliminable.Turner notes at some places that normative talk is not ineliminable from a correct explanation of normative phenomena (p.187).As he says "judging is unnecessary for the explanation-giver".But this general commitment, again, puts Turner on the boat of those that, like expressivists, are interested in a secondorder account of our normative practices; an account that leaves aside the specific content of normative convictions.
Now, once we agree on the level of analysis, what is the proper focus of a secondlevel explanation?For Turner, and so for contemporary expressivists, we must start with the agent's psychology (p.186).Instead of classical metaphysical or epistemological questions about normativity, expressivists begin by asking about the kind of mental states expressed by means of normative utterances.Allan Gibbard (2003) writes: The expressivist' strategy is to change the question.Don't ask directly how to define good, for no correct definition can break out of a normative circle . . .Instead of a straight definition, expressivists propose, seek a characterization of a different form.Ask what states of mind ethical statements express.(p. 6) Shifting the focus from big philosophical issues (semantic, ontological or epistemological) to psychological questions has proven to be fruitful, enabling some expressivists to make connections between their second-order claims and some scientific findings11 .The kind of benefits I have in mind are methodological.If we endorse a second-order view of morality where non-cognitive states are central while the better science about our capacity for normative guidance also stresses the importance of emotions, then maybe the desired unification inside social sciences could move a step further, minimizing criticisms based on our commonsense commitments to the status of our normative talk.Expressivism allows us to claim that this shared view is in perfect order, i.e. that normative claims can be true, objective, and so on.But expressivism, and it has proven to be crucial, does so while assuming that at the very end our normative talk is in the business of expressing emotions whose function is the guidance of action.This duality inside contemporary expressivism, i.e. the kind of empirically informed story it offers to us about morality plus its commitment with the normative appearances, should be valued by those who are interested in achieving a unified account of normativity12 .
But there are some other similarities.Expressivists are animated by a non-representational semantic.In their view, the proper semantic category to analyze moral or normative discourse is not representation but rather expression (of some psychological state still to be defined).For Turner, representation is also a suspicious category.The reason is that such a category is the hidden component that animates substantive accounts of normativity, i.e. accounts where normative talk is about some realm of facts that are 'out there in the world' (a problem with this criticism of Turner, by the way, is that sometimes he himself uses the category of representation to describe the commitments of all normativists and it is of course an open question whether all normativists are committed to such a strong thesis about the semantics of normative discourse13 ).
Another important feature of expressivism, at least in its current form, has to do with the surface grammar through which our moral commitments are expressed.What is nowadays known as quasi-realism claims that our normative talk is in perfect shape, with no reform of our discursive uses in moral contexts being necessary.There is no heavyweight metaphysical problem in saying that 'It is true that torturing babies is wrong', or even that 'It is an objective fact that torturing babies is wrong'.The only thing we need to be aware of is that these uses stand for different and more complex forms of expressing a negative attitude toward the particular act of torturing babies.
There are places where Turner endorses quite a similar take on normative talk.To my mind, clear cases of commitment on his side with something similar to quasi-realism can be found, for instance, in the pages where he appeals to the metaphor of 'wearing a normative lens' for characterizing the kind of account he wants to favor (p. 8, 12 and 194).On these pages, and while commenting on a quote of Michael Friedman, Turner claims that speaking of normative facts is perfectly fine as far as we do not try to locate them in the natural world as facts.Turner writes: The door Friedman's argument opens is to a way of making sense of normative language as normative without invoking a shadow world of normativities (p. 8-9) This door has been open inside meta-ethics for a long time.Behind this door is, as I have been suggesting, If you are looking for a philosophical account of normativity able to integrate findings coming from several scientific domains (a philosophical account starting with basic emotions and their entanglement with other, more representational, mental states), then expressivism is your best bet.A final question for Turner is thus if there is some hope that expressivism can accommodate his empathybased account of normativity.