Normative Equivalence

Sometimes moral philosophers engage in genuine debate. Sometimes, however, it turns out the apparent
debate is merely verbal; the philosophers do not disagree about the fundamental
normative facts. When this happens, let
us say that the apparently conflicting normative theories are normatively equivalent. It seems to me that it’s important to be able
to determine whether an apparent debate between ethicists is real or merely
apparent. In this post, I propose a test for determining when normative theories
are normatively equivalent, sketch an argument that this is a good test for
normative equivalency, and briefly discuss why it can still be difficult to use
this test.

Consider two cases, each involving allegedly competing

 Case 1:

 Philosopher A: I
think that a morally right action is one that maximizes utility; you ought to
do what is best. But I also believe in
an “all things considered ought”, and sometimes what one all things ought to do
is not what one morally ought to do. Sometimes one “all things considered”
ought to look out for number one, and take that fishing trip.

Philosopher B: Utilitarianism is a false theory; sometimes
it is morally permissible to not maximize utility. Maximizing utility is prima facie obligatory;
but so is looking out for number one. So
sometimes one ought not maximize utility, but instead take that fishing
trip. I also believe in an “all things
considered ought”, but I hold that one should always do what one morally ought
to do.

 Case 2:

 Philosopher A: The
only thing that directly contributes to welfare is pleasure. You want to live a good life? Then ensure that you are really happy. However, although I hold that the only thing
that improves your welfare is pleasure, I also hold that you can have rational
prudential interests in other things besides pleasure, such as accomplishing
important goals. And sometimes it is
prudent or rational even from a purely self-interested standpoint to sacrifice
your own welfare in order to accomplish those goals.

 Philosopher B: I am a pluralist about welfare: I believe
that pleasure and accomplishment both contribute to your welfare. However, I deny that you can have rational
prudential interests in things besides those that directly contribute to your
welfare. From a purely self-interested
standpoint, your own welfare is the only thing that matters.

 Both cases might seem to involve only verbal disagreement. But how could we check?

 I will make use of the notion of a reason when formulating a test.  A reason is a consideration in favor of doing
something or taking a stance towards some proposition. Reasons are things that
persons can have; the come in various strengths (my reasons for helping the
poor might be weaker than my reasons for helping my wife); they can be reasons
to perform actions; or they can be reasons to have a certain attitude (such as
fear, desire, or attitudinal pleasure) towards a certain proposition; finally,
one’s reasons can (presumably) differ from time to time. 

 Let us call any proposition of the form, “person p has a reason of strength n at time t for doing action A or standing in propositional attitude r to proposition o”, a reason-stating
An example of a
reason-stating proposition: Kris McDaniel has a reason of strength 10 right now
to feel somewhat guilty about not contributing to the Pea Soup Ethics Blog.” We will also allow propositions derivable by
replacing the free variables with variables bound by a quantifier to count as a
reason-stating proposition. Example: All persons have a reason of some strength
at some time for taking pleasure in the fact that Henry Heathwood is taking
innocent delight in discovering new things about the world. Finally, let’s count propositions consisting
of reason-stating propositions and Boolean operations on these propositions as
reason-stating propositions. [But we
will exclude tautological or contradictory propositions from counting as
reason-stating propositions.] Example:
Kris McDaniel has a reason of strength 10 right now to feel somewhat guilty
about not contributing to the Pea Soup Ethics Blog and Ben Bradley has a reason
of strength 4 right now to chastise Kris.

 The test is this:  Theory
A is normatively equivalent to theory B if and only if, for any reason-stating
proposition p, A entails p if and only if B entails p. 

In short, if two normative theories have exactly the same
implications about what reasons we have, then the two normative theories are
describing the same normative facts. (Note that the two theories may not be equivalent across the board,
since one theory may “take a stand” on something non-normative as well.)

Here is a quick, far too sketchy defense of the test:

(1) Any proposition attributing a normative feature (such
as being good, being unjust, or being vicious) to a thing or things is
necessarily equivalent to some reason-stating proposition. 

[A defense of (1): if (1) isn’t
true, then the alleged normative feature isn’t really normative, since it is
the giving of reasons (of some sort or other) that makes a feature normative.]

           (2) If (1), then the test is an adequate test of normative

            [A defense of (2): Suppose (1) is true. Then any normative theory is equivalent to
some                 theory whose normative content can be expressed entirely in terms of
reasons. Then the             normative contents of
any two theories are identical provided they say the same things                 about what
reasons we have. But this seems to mean that the proposed test for normative
            equivalency is adequate.]

In order to apply the test to the two cases above, we need
to reformulate both sets of theories in terms of reasons. Only then are we in a position to determine
whether the alleged disagreement is real or merely apparent. 

8 Replies to “Normative Equivalence

  1. Hi Guys — Sorry this is so long. I guess I haven’t figured out how to make the bulk of the text appear under the fold…

  2. Before we get further, I think there is a problem in the way you put formulate the reason-stating propositions. Here’s the quote:
    ““person p has a reason of strength n at time t for doing action A or standing in propositional attitude r to proposition o”, a reason-stating proposition. An example of a reason-stating proposition: Kris McDaniel has a reason of strength 10 right now to feel somewhat guilty about not contributing to the Pea Soup Ethics Blog.”
    I guess my puzzle is, does such reason-stating propositions state what the reasons the persons have are. If we look at the example, it only states the attitude required by the reason (feel guilty about not contributing to the Pea Soup Ethics) and not what is the reason for this. The example would not then fit the schema if it is taken to include what the reason is for the attitude.
    There is a way of reformulating the example to make it express the reason. Then not contributing to PSEB would be the reason for KM to feel guilty. This however does not fit with the general schema either. First, the reason, not contributing to the blog, seems to be a fact, state of things in the world and not a proposition as hinted by the schema (*proposition that* KM has not contributed…). It is not clear if propositions (whatever they really are, I never got a clear account of this) can serve as reasons.
    So, before attempting to assess the test, I guess I would like to see what the reason-stating propositions and the schemas they can be put to more precisely are.

  3. Hi Jussi,
    Thanks for the questions.
    The statement of the example was sloppy. A more perspicuous (but equivalent) statement is:
    “Kris McDaniel has a reason of strength 10 right now to feel somewhat guilty that he has not yet contributed to the Pea Soup Ethics blog.”
    It’s clear here I hope that the attitude in question — feeling somewhat guilty — is a propositional attitude; and I hope it’s also now clear what the relevant proposition is.
    You note that reason-stating propositions state what reasons people have; they don’t state the sources of those reasons. It’s one thing to say that I have a reason to feel a certain way or do an action. (It’s another thing altogether to say *why* I have those reasons. Just like it’s one thing to say that a person has a certain obligation, and another thing to say why that person has that obligation.)
    Reason-stating propositions are silent on the question of why people have the reasons they do — as I see things, it’s up to a normative theory to do this.
    I don’t understand your last worry, “It is not clear if propositions (whatever they really are, I never got a clear account of this) can serve as reasons.” I didn’t claim that “propositions served as reasons.” I did assume that one can have a reason to have a particular attitude towards a proposition, but I would have thought that this is relatively uncontroversial. For example, someone can have a reason to be afraid that something bad will happen, or a reason to be happy that his nephew is a healthy baby, or a reason to believe that there is no God.

  4. Kris,
    I like your proposal. I am concerned about one thing though. You don’t say what it takes for someone to “have” a reason. Suppose Tiny and Tony have different views about the psychology of reason possession. Tiny thinks R is a reason for S at t only if S could become aware of R on reflection at t. Tony thinks R is a reason for S at t only if S is aware of R at t.
    What I am worried about is that perhaps Tiny and Tony hold the same moral view. However, they disagree about which reason-stating propositions are entailed by their respective views because they disagree about what it take for someone to have a reason.
    I hope that was relatively clear.

  5. Hi Christian,
    Nice to hear from you again. Your worry is interesting. Just to make something clear: reasons, in the sense that I am using that term, are simply considerations in favor of doing something or having some attitude. The reasons themselves are not (necessarily) psychological states.
    When I formulated the test, I made use of the notion a person “having” a reason. I now think this phrase is unfortunate, since it suggests that I am talking about psychological states. So let me reformulate the notion of a reason-stating proposition. Let us instead say that a reason-stating proposition has this form:
    “At t, there is a reason of strength n for person p to do action A or have the propositional attitude r to proposition o”
    Now I’ve eliminated talk of a person “having” a reason here, and so have eliminated any implicit suggestion that this might have something to do with the person’s psychology.
    I think something like your worry could still arise — I’m not sure, and want to think about this more before responding– but maybe it’s easier to solve with this more careful formulation.

  6. Hi Kris,
    big thank you for the clarifications. Unfortunately, I do still have my worries. I’ll start with this:
    ‘Reason-stating propositions are silent on the question of why people have the reasons they do’.
    Now, this seems to imply that the the reason-stating propositions do not express what the reasons the agents have but only that they have reasons. This creates a problem for the test. It may for instance be that rule-consequentialism and contractualism are in fact co-extensive in their prescriptions. In all the same situations, they would say the agents has a reason of the same strength to do the same thing. However, they do locate the sources of reasons to different places – contractualism to rationality of others and rule-consequentialism to general well-being. Your test, as it is unable to express this difference, would make them ‘normatively equivalent’. I am uneasy about that especially if it is taken to mean that there is no clear disagreement but merely a ‘verbal disbute. Surely there is a disagreement about what is the source of fundamental moral reasons or what the reasons are, and that doesn’t look like a mere fight over words.
    And, I still have worries your conception of attitudes and propositions. It is true that emotions like guilt and fear are propositional attitudes. They do have propositional content – I am guilty about such-and-such and afraid of this-and-that (compare to being just angry). This however does not mean that object of the attitude is a proposition or that the attitude is directed at a proposition. I take it that this is what you think, because in the reason-stating proposition it says attitude *to* proposition p. Instead of propositions, our propositional attitudes are directed towards the worlds and things in there. Fear is the best example. I fear that a snake will bite me. This has propositional content, but I am not afraid of any proposition (again, whatever they are) but the actual snake biting me. Similarly, I think that you would feel more guilty about the *fact* that you haven’t contributed to PEA SOUP and not about the proposition. Who really cares about propositions…

  7. Kris,
    Don’t reasons come in all the various flavors that obligations do? I can have prudential reasons, moral reason, perhaps all-things-considered reasons, etc.
    Your test seems to rely not just on translating talk of obligation into talk of reasons, but on translating talk of various sorts of obligation into talk of one sort of reason.
    Similarly, your premise (1) needs to be understood as talking about one sort of reason, presumably all-things-considered reasons. So understood, I don’t think the premise is obvious. Consider some examples:
    A) Eating a cookie right now is in my own best interest.
    (A) is equivalent to a reason-stating proposition: I have a prudential reason to eat a cookie.
    This may entail some all-things-considered reason statement, but I can’t think of what such statement would be necessarily equivalent to (A).
    B) I am legally obligated to pay a quarterly income tax estimate on June 15th.
    So, I have some legal reason to pay my tax estimate.
    Again, this might entail some sort of all-things-considered reason — though in this case you’d need all sorts of auxilary facts about the consequences of violating the law, etc — but I can’t imagine it’s equivalent to an a-t-c reason.
    My point in summary is this: To the extent that your test works, it’s not because of the move from obligations to reasons, but because of the move from a plurality of normative concepts to a unique one. Why not have your (allegedly) competing theories tranlate into talk of all-things-considered obligations? Would that work as well?

  8. Hey Kris,
    I don’t see any way to formulate my worry without taking a stand on what kinds of things reasons are. And I don’t know what they are exactly. I like your new formulation.
    And I was bummed that you decided to stay in Syracuse, I was hoping to do some philosophy with you and Chris. Good luck though!

Comments are closed.