Moral Naturalism and Three Grades of
Normativity
by David Copp
This paper is a lengthy careful
examination of the question whether
moral naturalism conflicts with the
normative character of ethical inquiry.
Definition of Moral Naturalism:
Moral naturalism holds that in thinking of
things as morally right or wrong, god or
bad, we ascribe moral properties to
these things- properties such as moral
rightness and wrongness, goodness and
evil. It holds that there are such
properties, and it adds that these
properties are ordinary garden-variety
natural properties- properties that have
the same basic metaphysical and
epistemological status as the properties
a tree can have as being deciduous,
and the property a piece of paper can
have of being a 20 dollar bill.
The Problem
Moore’s Open-Question Argument
G.E. Moore argued that whatever
natural property we may consider, it
remains an open-question whether
anything that instantiates that property
ought, or ought not to be pursued.
Happiness, for example, is a natural
property. Most people think that
happiness is basically a good thing, but
Moore would claim that one can
meaningfully ask whether happiness
ought to be pursued. Even if we all
agree that the answer is yes, what he is
pointing out is that there is nothing in the
concept of happiness itself that implies
that it ought to be pursued.
Many people who, upon grasping
Moore’s point for the first time, find it
very compelling. It is, of course, a
version of the is-ought distinction, and it
depends fundamentally on the idea that
any statement of the form “x is good”
must be a necessary truth, knowable a
priori. So, to the extent that naturalism
rejects the a priori, it also clearly must
reject Moore’s open-question argument
as well.
But rejecting the a priori is not sufficient.
The naturalist owes a positive account
of normativity.
Moral Naturalism
For Copp, moral naturalism implies
moral realism. (Parallel:
epistemological naturalism implies
scientific realism.) This means:
1. There are moral properties and
relations.
2. Moral properties do not have a
special metaphysical status.
3. Moral predicates express such
properties.
4. Moral assertions express moral
beliefs with the same truth conditions
as any other belief.
5. Some moral assertions are true;
i.e. e.g., some actions really are right
or wrong.
In claiming that moral properties are
natural properties, he means that they
are empirical, and by this he means
roughly that empirical evidence is
relevant to evaluating the truth and
falsity of moral claims.
Three Grades of Normativity
Generic Normativity
“Generic normativity” is Copp’s term for
describing the “essential semantic
connection” of moral claims to action or
choice.
1. Cheating is wrong.
2. Cheating is common.
The first claim seems to entail that one
ought not to cheat, but the second has
no such entailment.
Motivational Normativity
“Motivational normativity” refers to the
assumption that the fact that a
practically rational person understands
something to be right or wrong gives
them a reason to pursue or avoid it. In
other words, all things being equal, if a
rational person says “Action X is good”
he will be motivated under the relevant
circumstances to do X.
Authoritative Normativity
It is possible to imagine someone who
recognizes both generic and
motivational normativity, but still finds
herself compelled to ask whether or not
they should do the right thing in a
particular circumstance. Many
philosophers would argue that this
question is actually incoherent given
both generic and motivational
normativity. But it is not incoherent.
There are different forms of normativity,
e.g.
Cheating is rational.
Cheating is widespread.
We can argue that rationality has its
own kind of generic normativity, and that
one can be motivated by considerations
of rationality in just the same way that
one can be motivated by morality. If this
is the case, then we would need a
reason to do what is right, rather than
what is in our practical self-interest.
(This is the point of Plato’s telling of the
myth of Gyges in the Republic.)
One way of addressing this is to say that
moral considerations are authoritative,
i.e., that they always trump other kinds
of normative considerations.
The Problem Revisited
A naturalist has two avenues with which
to express their naturalism in respect to
any of the three grades of normativity
identified.
Deny that it is essential to our
understanding of morality.
Accept that is essential, but assert
that it can be explained
naturalistically.
Given this, Copp understands the task
of the argument against naturalism with
respect to any of these three grades of
normativity is:
Show that it is essential to our
understanding of morality.
Show that its normative character
can not be explained naturalistically.
Expressivism and Realist-
Expressivism
It is important to note here that there is a
naturalistic account of morality with
which Copp’s definition of moral
naturalism is incompatible. This is the
“expressivist” view that moral terms
function to express a certain kind of
conative attitude, namely acceptance or
rejection. (“Emotivism” and “non
cognitivism” are two other terms whose
meanings are closely aligned with
expressivism.)
Copp’s answer to this is simply that we
probably do use moral terms to express
conative attitudes, but we also use them
descriptively. Copp rejects a stronger
form of expressivism, according to
which expressing conative attitudes is
the sole purpose of moral claims. Most
descriptivists are very impatient with this
idea, and Copp is no exception. He
thinks it is obvious that this is wrong,
since a statement like:
Jim Bob is an idiot.
expresses an attitude of contempt, but it
is not normative in the moral sense of
the term. (I need to point out here that
expressivism is actually quite a bit more
sophisticated than this, and not so easily
refuted, though most moral philosophers
present a straw man version of it.
Moreover, expressivism would rate as a
certain kind of naturalism, essentially an
eliminativist form.)
Generic and Motivational Normativity
Copp’s task, then, is to provide a
naturalistic treatment of all the different
forms of normativity; either a naturalistic
basis for denying that they exist, or a
naturalistic account of their basic
functioning.
Naturalized Generic Normativity
Copp believes that the semantic aspect
of normativity can be established
naturalistically as a relation between an
action or choice and an accepted
standard of behavior.
To take a non moral example, if we say
that it’s rude to stare, we mean that
staring is on the list of behaviors
prohibited by a social rule proscribing
unprovoked behavior calculated to make
people feel uncomfortable.
For Copp, morality has the same basic
structure. The difference is that moral
rules has a status that rules of etiquette
and rudeness do not. Specifically:
“moral wrongness has the property of
being ruled out by a system of
standards the currency of which in the
relevant society would contribute better
than that of any other such system to
enabling the society to meet its needs.”
Copp notes that this is hardly an
unproblematic account. Certainly there
are epistemological difficulties involved
determining whether a system of
standards meets this criterion, but this
difficulty is practical and empirical, not
conceptual. What’s important is that this
account connects the meaning of
normative claims to action in a
naturalistic way.
Naturalized Motivational Normativity
Copp says that he does not actually
think that moral claims have
motivational normativity. (In other words,
he thinks fully rational people can have
a complete grasp of a normative claim,
yet experience no motivation to act
according to it, e.g., someone who is
clinically depressed.)
However, he thinks a naturalist could
acknowledge motivational normativity
just as easily as a non naturalist. All that
is needed is a theory of rational agency
according to which moral claims can
function as reasons for action, and Copp
thinks there is no special explanatory
apparatus required for moral reasons as
opposed to any other kind of reason.
Naturalized Authoritative Normativity
The rest of this paper is dedicated to a
discussion of Authoritative Normativity.
Again, it will turn out that Copp does not
think morality has authoritative
normativity, but he also explores some
naturalistic strategies that accept it.
Copp identifies two different ways of
thinking about authoritative normativity.
Consider the question:
Why be moral?
One account of authoritative normativity
stipulates that there is an answer to this
question; i.e.,
Requirements of some kind K have
authoritative normativity just in case,
necessarily, there is an authoritative
reason to act as one is K-required to
act.
Alternatively, what might claim that the
question “Why be moral?” is “closed”.
What this means is that
No fully rational person who is
thinking clearly can take there to be
any serious practical question of
whether to act in accord with K
requirement.
Doubts about the significance of
motivational and authoritative
normativity.
In this section Copp argues against
authoritative normativity, and he extends
this argument to cover motivational
authority as well. The argument
basically goes like this:
Both formulations of the concept of
authoritative normativity (above) define
the concept by reference to rationality.
This means that whether or not
something is judged to have
authoritative normativity by a rational
person, ultimately depends on whether
the norms of rationality dictate this
judgment.
But this means that questions of the
form “Why be moral?” ultimately depend
on the answer to the question “Why be
rational?”
One can plausibly assert that a moral
person does not sincerely pose the
question “Why be moral?,” and that a
rational person does not pose the
question “Why be rational?” but the
point stands that in the absence of an
authoritative answer to the latter- i.e., a
demonstration that rationality has
authoritative normative significance,
then neither morality nor rationality may
be judged to have authoritative
normative significance.
Naturalism and Authoritative Normativity
On Copp’s view, then, the task of the
naturalist is just to produce a theory of
rational agency that has clear
implications for the question of
authoritative normativity.
According to his own fully naturalistic
view, rationality consists in “efficiently
servings one’s values,” and he proposes
a standard of rationality that consists of
a kind of categorical imperative to do
just that.
On this view, a rational person may
actually have no moral values at all, so
Copp’s brand of naturalism does not
acknowledge any authoritative
normativity for morality.
However, Copp is not especially
concerned here to peddle his particular
theory of rationality. He is only
concerned to show that fully naturalistic
answers can be given.
Self-Conception Strategies
Self-conception strategies are
essentially Kantian approaches, which
make some form of the claim that
rationality and morality are two sides of
the same coin.
Christine Korsgaard, for example,
claims that what it means for an action
to be morally required is that it is
required by our “identity” as rationally
reflective agents.
Put differently, to conceive of and value
yourself as a rationally reflective agent,
is to accept the moral requirement to
behave in a manner befitting a rationally
reflective agent. So, according to
Koorsgaard’s theory, the 2nd type of
authoritative normativity obtains; the
question “Why be moral?” really can not
arise for a rationally reflective agent.
(This is exactly the sort of thing that
Kant would say as well.)
Copp disagrees with this conclusion, but
his primary point here is that a naturalist
can accept Korsgaard’s analysis if she
can give a naturalist reading of her
requirements of moral agency, and
Copp things this can be done, basically
as follows:
On Korsgaard’s analysis, an action is
morally required iff
K1: the agent’s failure to perform it
would conflict with the agent valuing
herself as a rational agent.
K2: the agent must value rational
reflective agency if she is to value
anything and to see herself as an
agent.
The question, then, is whether these
properties have a naturalistic
interpretation. Copp thinks its clearly
plausible that they do. K2 appears to be
a conceptual truth, which Copp says can
be “set aside as analytic.” You may find
this to be a disturbing remark, since
naturalists are notoriously unfriendly to
analyticity.
But all Copp means here is that K2
simply expresses a deductive
relationship, i.e. “an agent has values”
deductively implies that “the agent is
rationally reflective.” So it does not
constitute any special challenge to
naturalism beyond the requirement that
deductive relationships themselves have
a naturalistic interpretation.
A naturalistic interpretation of K1
depends simply on a naturalistic
interpretation of a values conflict. For
example, if the failure to pay a debt
conflicts with a persons own values, the
naturalist must be able to say, in fully
naturalistic terms, what this means.
Again, Copp sees no obvious
conceptual hurdles to accomplishing
this.
Conclusion
So, Copp’s argument really comes down
to this.
The plausibility of the claim that
naturalism can’t capture normativity
depends on lumping the different kinds
of normativity together. When you
distinguish them, then it becomes
apparent that there are several avenues
by which the naturalist may capture
normativity.
(Of course, Copp’s own way makes you
into a flaming moral relativist, which
many non or anti naturalists would treat
as a reductio.)
Of the three alleged kinds of normativity
involved in moral claims- generic,
motivational and authoritative- only
generic is real, and the naturalist can
give a plausible account of generic
normativity.
However, even if Copp is wrong about
the above, there are still plausible
naturalized forms of motivational and
authoritative normativity.