The Naturalistic Fallacy, introduced by G.E. Moore at the start of the 20th century, stands as an objection to any attempt to reduce moral terms such as "good" to natural terms such as 'happiness'.
Moore attempting to argue that 'good' refers to a basic, unnatural property. We cannot reduce this term any further. We cannot reduce it to 'pleasure', for example; or of 'happiness'. Nor can we reduce it to non-natural properties such as 'that which is pleasing to God'. There is nothing more basic than "good" that we can use to understand this term. In a sense, it would be more accurate to call this fallacy the reductionist fallacy.
The Open Question Test[]
This type of move is shown to be a mistake through the Open Question Argument. According to this test, we can show that a reduction from Term 1 to Term 2 fails by showing that competent users of the language can question the relationship.
For example, let us look at an attempt to reduce the term 'bachelor' to 'unmarried male.' We ask a competent speaker of English a question like, "John is a male who is not married, but is he a bachelor?" The competent speaker of English who understands the meanings of the terms would not consider this an open question. "Certainly he's a bachelor. That's what 'bachelor' means."
This marks a successful reduction.
When we attempt to reduce 'good' we get nothing like this type of success. If we try to reduce good to pleasure, a competent speaker who understands the terms can still sensibly ask, "Bullying that kid was pleasant, but was it good?" It even works on non-natural properties. "Releasing a plague that killed all first-born children was pleasing to God, but was it good?" In both cases, the answer might be "yes". Then again, it might not be. This is an open question. Consequently, these two attempted reductions fail Moore's Open Question test.
Desirism, one may argue, would also fail the Open Question test. "X stands in the relevant relationship to the relevant desires, but is it good?" This will remain an open question regardless of how we flesh out 'proper relationship' and 'relevant desires'.
Desirism's Response[]
Good Reduced to Reasons for Action[]
The claim that desirism reduces 'good' to relationships between states of affairs and desires is false. Desirism reduces 'good' to reasons for action that exist. The fact that desires are the only reasons for action that exist is not true by definition. If categorical imperatives, intrinsic values, divine commands, social contracts, and the like were real, they would provide reasons for action relevant to whether something is good. The fact that they are not real is not something that can be learned from a conceptual analysis of the word 'good'.
Ambiguity of Good[]
Another confounding fact that bears on Moore's Open Question test rests on the fact that 'good', even when referring to relationships between states of affairs and desires, is ambiguous. One can always ask the question, "Which desires?"
Consider the following:
San Francisco is closer to the north pole than Los Angeles, but is it north of Los Angeles?
This is not an open question. Competent speakers of English who understand the terms will answer, "Of course. That's what 'north' means."
However, if we change the wording slightly, we get:
San Francisco is closer to the north pole than Los Angeles, but is it north?
North of what? North of you? I don't know. Where are you?
The question becomes an open question. However, its openness is not due to any inability to reduce the term 'north'. It is due to the fact that 'north' is a relational term and the way the question is phrased makes the relationship that we are interested in unclear. The question likely refers to the relationship between San Francisco and Los Angeles, but it might not.
Relationships between states of affairs and desires suffer the same effects as relationships between objects. The fact that something stands in a particular relationship to one set of desires does not eliminate the possibility that it stands in a different relationship a different set of desires. Consequently, "This is the best tasting chocolate cake on the whole planet, but is it good?" remains an open question. Is it good for you? Absolutely not.
The Masked Man Fallacy
Another reason we can dismiss the Open Question argument is because it is an instance of another fallacy, the Masked Man Fallacy.
Imagine yourself in London in the late 1800s. There is a robber about, known only as “the Masked Man,” who has been stopping carriages on dark roads at night and robbing the passengers. One day, while you are at a party, an inspector from Scotland Yard comes in. He walks up to the host of the party and says, “We have arrested your brother. We believe that he is the Masked Man.”
Your host answers the Inspector, “That cannot be true. Yesterday, I was talking about the Masked Man with some dinner guests. At the time, I certainly knew the name of my brother James. However, I did not know the name of the Masked Man. In other words, the question, 'James is my brother, but he is the Masked Man?' was an open question. It still is, for that matter. Therefore, inspector, you have arrested the wrong man."
The inspector can dismiss this objection as bogus. Certainly his claim that the host's brother and the Masked Man are the same person does not depend on what the Host was aware of the day before.
Similarly, the fact that, “X objectively satisfies the relevant desires, but is it good?" is an open question would not prove that ‘good’ and ‘is such as to objectively satisfy the desires in question’ are not the same thing.
We look to see whether James is the Masked Man by looking to see if there is anything true of James that is not true of the Masked Man. If we learn, for example, that James was attending an important business meeting at 10:00 yesterday night – the same time that the Masked Man robbed a carriage on the corner of Wadsworth and Main, then we know that James could not be the Masked Man.
To be fair to Moore, it is not clear whether he was referring to this type of reduction. Moore's argument may be interpreted as being fixed on the issue of reducing one definition to another, not one phenomena to another. However, people often attempt to use Moore's open question argument in this way.
We see similar objections raised in other fields of study. A person may attempt to describe a toothache in terms of decomposition of a tooth resulting in the firings of a signal along nerve fibers to the brain that is processed in a particular way, causing the agent to put his hand to his face and say "Ow!". Against this, somebody may argue that he can be fully aware of all of these physical facts and still not know whether the agent is in pain. So, we cannot, in fact, reduce a toothache to a set of physical interactions.
However, using the agent's ability to question whether the person is experiencing actual pain as reason to deny the reduction follows the masked man fallacy. The host's ability to know all he did about the masked man and still be able to question whether the masked man was his brother is not a valid objection to the claim that the masked man is his brother. Similarly, the ability to question whether an entity experiencing all of the physical qualities of a person with a toothache without feeling pain does not disprove the thesis that toothaches can be reduced to that set of physical interactions.
Testing a Reduction[]
The test for whether desirism provides an accurate account of morality is not found in an open question test. It is found in the fact that the terms actually make sense of a great many of our moral practices.
For example, equating a wrong act with an act that people generally have many and strong reasons to condemn (in terms of promoting aversions that would aid in the fulfillment of many and strong desires) explains the moral practice of "excuse". An examination of excuses shows that everything that counts as an excuse is a claim that denies the proposition that people generally have many and strong reasons to condemn somebody who performs such an act.
It makes sense of the practice of incorporating praise and condemnation in our moral statements - since these social tools work on the reward centers of the brain to mold malleable desires.
It makes sense of the principle that "ought" implies "can" because using praise and condemnation on non-malleable desires or on actions that cannot be affected by changes in malleable desires is a waste of energy.
The Open Question Test is not a legitimate test of a moral theory. A test as to whether one phenomena can fill the same functional role is a legitimate test. This is a test that desirism can pass.