Normativity and Rationality: Framing the Problem

© Springer International Publishing Switzerland 2015
Michał Araszkiewicz, Paweł Banaś, Tomasz Gizbert-Studnicki and Krzysztof Płeszka (eds.)Problems of Normativity, Rules and Rule-FollowingLaw and Philosophy Library11110.1007/978-3-319-09375-8_8

8. Normativity and Rationality: Framing the Problem

Joanna Klimczyk1, 2  

Department of Philosophy and Sociology, Polish Academy of Sciences, Nowy Świat 72, Warszawa, 00-330, Poland

Department of Philosophy, University of Szczecin, Nowy Świat 72, Warszawa, 00-330, Poland



Joanna Klimczyk


The paper is divided into four sections. The first one examines the hypothesis about whether the sharp distinction between two kinds of the normative requirements of rationality: the substantive and the non-substantive is a plausible view, given we can show that at least one particular non-substantive requirement of local attitudinal coherence is inscribed into the very idea of genuine normative requirement of whatever source. The second considers a particular version of a popular argument in favor of the substantive construal of the normativity of rationality that builds on the putative analogy between the normativity of rationality and the normativity of morality. The conclusion is that the argument remains unsuccesful because the analogy shows to be apparent. Section 8.3 explores the First-Personal Authority Account as an argument for the non-substantive normativity of rationality, and rejects it on the ground of its irrelevance. It is argued that the main problem with the the First-Personal Authority Account is that instead of establishing that attitudinal coherence is a normative claim of rationality, it provides support for the psychological interpretation of the normativity of rationality. Finally, granted that the arguments in the above sections are roughly correct, and the idea of the intrinsic normativity of rationality remains a muddle, a radical solution is advocated for. Instead of working hard on vindicating the normativity of rationality, we should rather rest content with the view that the only normativity of rationality for which we have support has external source in what we care about.

Local attitudinal coherenceNormativityRationalityReasonsFirst-Personal Authority Account

8.1 Introduction

Most participants in the debate over the normativity of rationality subscribe to three claims. The first says that you ought to reason correctly, where ‘correctly’ is a cover-term that includes various requirements of rational reasoning, such as belief-consistency: the requirement not to believe that p if you believe that not-p; belief-closure: the requirement to believe q if you believe that p and believe that if p then q; instrumental rationality: the requirement to intend to B if you intend to A and believe that your Bing requires that you A, just to mention the most noncontroversial from the list.

The second claim is that rationality requires you to respond correctly to reasons. ‘Correctly’ is a term of art here and there are hot debates about how to interpret correct responding to reasons. Interpretational query aside, the more important question concerns the issue whether rationality is to be appropriately thought of as consisting in responding to reasons at all (Broome 2007).

Finally, according to the third claim, what rationality requires of an agent is that she has to be in the relevant states of mind. If she fails to be in a certain state of mind at given circumstances, then she is not entirely as she ought to be. Behind that claim lurks the elusive thought that there is some normative truth about what states of mind an agent ought to be in, where that very truth is determined either by facts about how things are in the world, or by the contents of the agent’s attitudes. Many philosophers who believe in the PLATITUDE also believe that the truth of the above claims can only be explained in terms of the normativity of rationality. To put the idea crudely, if we reject the view that rationality is normative, not only does the source of the PLATITUDE remain obscure, but also those who argue for the PLATITUDE are guilty of defective thinking. The accusation can take the following form: how can you truly believe in something if you lack satisfactory support for your claim? Or, how can you believe in something if you cannot vindicate the thesis you find to be true?

In the paper I shall try to make a preliminary contribution to the project of ­demystyfying the view that rationality is the source of normative claims on subjects. The view that I criticize consists of either one, or two of the following theses. The first says that the requirements of rationality form a distinct kind of requirements conferred on subjects, analogously to the requirements stemming from such sources as morality, or prudence. According to the second, conforming to the requirements of rationality, however one pleases to construe them, is a genuine normative issue. By the expression “genuine normative”, I do not mean the issue of whether any individual requirements of rationality set out standards for correct reasoning either of theoretical, or of practical kind. Rather what I have in mind is something more substantive, regarding whether the requirements of rationality belong to the normative province of reasons. Normative reasons, I contend, are the only and appropriate extension of the term “normativity” that we are really interested in.

The paper takes as its point of departure the bundle of claims about what rationality requires that fall under the PLATITUDE, and then proceeds as follows:. I examine the first two claims of the PLATITUDE, and raise doubts whether it makes sense to think of two, independent, general requirements of rationality, which respectively I call the ‘substantive’ and the ‘non-substantive’ ones. Next, I put into question the suggestion that the PLATITUDE’s credibility turns on the claim that rationality is normative. Finally, I sketch an alternative way of thinking of the requirements of rationality in terms of the requirements of comprehensibility. The main advantage of the comprehensibility approach, as far as I see it, is that this approach allows us to save what philosophers from the opposite sides want to save, that is, the idea of the first-personal authority of the requirements of rationality, yet dismiss the task with which we have fared so badly, namely the struggle to demonstrate that rationality itself, alike morality, is the source of intrinsically normative claims.

My suggestion, if plausible, obviously shares the fate common to bold philosophical theses: it has its own price. The price to be paid in this case is to come to terms with the possibility that normativity of rationality, if there is any, comes from somewhere else. This ‘somewhere else’, I have in mind, is what I call ‘the desire of comprehensibility’, which is a sort of a primitive desire of imperfect yet capable to rationality creatures such as we are. In short, I shall end with the conclusion that the apparent intrinsic normativity of rationality is confused with the straightforwardly non-normative requirements of comprehensibility.

8.2 Are There Really Two Kinds of the Requirements of Rationality?

Being in the grip of the requirements of one sort or another is something natural to social creatures of our kind. You can consider these various requirements that apply to human beings in terms of their source (Broome 2005, p. 324; 2008, pp. 96–97; Southwood 2008, pp. 17–18): moral, prudential, professional etc., or object. The requirements understood in terms of their object are the requirements that apply to the same object, yet stem from various sources. One example of requirements construed in terms of object are the requirements towards the poor. As a moral agent you have a certain set of obligations towards those who live in the poverty, and as a social worker you have another set of obligations that derive from your professional code. An alternative, simple way of thinking about requirements is to think of them in terms of substance and coherence. Requirements of substance are source-requirements, provided by morality, prudence or professional codes, whereas requirements of coherence seem to have no source in the sense that morality, or prudence are sources of various demands. The latter are usually construed of in terms of the general non-substantive requirement of rationality. What this general non-substantive requirement of rationality requires of you is not so much to conform to some particular norm of some particular source, or sources, but rather to think with certain pattern of correct thinking. Which pattern is the one you should think through depends on what sort of reasoning you are engaged in. If you intend to catch the 9 a.m. train to Cracow, and you know that this task will prove unfeasible unless you perform certain initial actions, including getting up at least two hours before the departure of the train on the day you plan to go to Cracow, then the norm or pattern of correct reasoning for your case is instrumental rationality. In this case the fact that you have intention to go to Cracow determines which other attitudes you ought to have, or ought not to have. For example, if it is true of you that you do intend to go to Cracow by train at 9 a.m. on the particular Monday, it must also be true of you that you do not have the intention not to go to Cracow by train at 9 a.m. on that very day. Or, it must be true that you do not have the intention to go there by plane and so on.

However, the sharp distinction between the requirements of substance and the requirements of coherence, as I label them, cannot be entirely correct. Recall what is the stipulated difference between the two kinds of requirements in question. Requirements of substance are the requirements whose normative force come from the relevant source, where what makes some source the province of normativity has to do with importance ascribed to the source in question. For example, morality is standardly thought of as a domain of value in one sense of the term “value” and prudence is thought of as a domain of value in another sense. On the other hand, the requirements of coherence amount to the general requirement of rationality that can be roughly characterized as the requirement to have attitudes that fit together in the coherent way. Call the source-normative requirements the substantive normative requirements, and the general requirement of rationality the non-substantive normative requirement.

Now, some philosophers are inclined to think that these two general requirements make for the requirements of rationality, though in a somewhat different sense. The rationale behind that view, which I label the Double Binding View, as I understand it, is that, if it is true that morality, prudence or any other source of substantive normative claims require of you something, conforming to that particular source-requirement is what is rational for you to do in one sense of the term “rational”, in which rationality is construed of in terms of appropriate responding to reasons. According to that line of thinking on rationality, if it is true that you ought to comply with a particular moral norm, or a particular norm of prudence, or a particular norm of ettiquette, you “gain”, so to speak, the relevant normative reasons to conform to them, and because it is true that you ought, or have normative reasons to conform to the substantive requirement in question, complying with this requirement is also the correct way of responding to the normative reasons you have. Consequently, if responding to correct normative reasons is one of the hallmarks of rationality, then responding to the source-normative requirements is one way of manifesting one’s rationality.

In what follows, I shall not discuss the issue whether the substantive connection between rationality and reasons is a tenable view, and specifically whether rationality consists in responding correctly to reasons, though I am sceptical about that. Rather, what I am interested in, and what will be the focal point of this section, is the problem that, as it seems to me, has not been given due attention in the ongoing debate over the normativity of rationality. The problem I have in mind concerns whether assuming the existence of two distinct kinds of the normative requirements of rationality along the lines of the PLATITUDE: the substantive one and the non-substantive one is not a bit far-fetched. The hypothesis I find worth testing is whether at least certain requirement of rationality in the non-substantive sense that is certain specific requirement of coherence among one’s attitudes is not already included, or entailed by the substantive normative requirement itself. To put the same idea in less complicated words: what I want to inquire about is, whether, if it is true that you ought, or have normative reason to conform to a particular ­requirement in the source sense of normative requirement, then it is also true that the content of that very requirement forces upon your attitudes certain requirement of attitudinal consistency of the local character.

Let us, provisionally, take the tentative thesis to say:

Inter-normative Transmission Rule of Rationality (ITRR)

Necessarily, if it is true that you ought to X or have normative reason to X, where X is what is required of you in the substantive, source sense of normativity, then it is also true that there is a certain pattern of local attitudinal coherence, such that you are required to satisfy if you are to satisfy a substantive requirement of normativity.

Spelt out more carefully, the Inter-normative Transmission Rule (ITRR) says that being under some normative requirement of substantive kind (in the source-sense of the normative requirement) entails being also under the particular non-substantive requirement of rationality of not having the combination of attitudes that in principle are irreconciable with the content of the particular requirement in the substantive sense of rationality that applies to you.

To see how the IRRR works, consider moral requirements, for example. Moral requirements require you to do, or not to do certain things, but they also require you to have coherent attitudes as determined by the content of some particular substantive moral requirement that applies to you at some particular time. Thus, if it is true that morality requires you not to cheat on other people, thus requires you not to behave in a certain way, it also requires you to believe that cheating on others is morally wrong, or generally morally reprehensible. Moreover, most people find what I here coin the requirement of content-fitting attitudes, a necessary condition on being moral at all. By ‘content-fitting attitudes’ I understand the attitudes that the subject is required to have when s/he remains under some particular source-normative requirement(s). Morality is an obvious case of such source-normative requirements that entail the relevant requirements of content-fitting attitudes. That this is so is what Kant has taught us when he claimed that acting on a moral principle is not yet the evidence of moral action. Such acting can merely be the evidence of acting in accordance with a moral principle. Moral action, if it is to deserve the name ‘moral’, must also spring from moral motivation, and moral motivation is nothing other than having the very attitudes that morality in the source sense of normativity requires you to have. What about prudence? Prudence is also unanimously taken to consist of a set of normative requirements in the source sense of normativity. Prudence requires you to do this or that. Suppose prudence requires you not to jump from the fourth floor of your apartment in order to attract the attention of by-passers on the street in which you live. Does the requirement of prudence, in the source sense of normativity, imply any particular requirement of content-fitting attitudes, for example one which says that if it is against the requirement of prudence to jump from the fourth floor of your apartment for no good reason, it is also against the requirement of prudence to believe that jumping from the fourth floor is something imprudent or stupid? It seems to me that it does. Generally and necessarily, normative requirements in the source sense of normativity go together with the relevant source-derivative requirements of content-fitting attitudes, being the part of the same package of what amounts to the requirements of rationality.

The tentative view defended here says that genuinely normative requirements, in the source sense of normativity, entail the requirements of local coherence among your attitudes, where the requirements in question are determined by some particular requirement you should in some particular context satisfy. If that is so, as I think it is, then the idea of introducing two distinct kinds of normative requirements: the substantive and those that concern coherence among one’s attitudes is, in an important sense, mistaken. Instead of bringing us closer to understanding the nature of the requirements of rationality, the idea in question makes us plunge into more confusion. That is so because we are told that there is a normative reason to have coherent attitudes or that we ought to have coherent attitudes, as if the required coherence among one’s attitudes formed a separate requirement that would be added to the one provided by some specific source-normative requirement. As if it were truly possible to satisfy the second (the substantive one) without satisfying the first (the coherential one). As if, for example, the moral requirement that forbids you cheating on other people could be genuinely satisfied without your having the relevant belief (if you are a moral cognitivist), or having some non-cognitivist attitude, say the attitude of being committed to the norm in question (if you are a non-cognitivist) that is determined by the very moral norm in question.