DarlBundren wrote:Just to make sure I understand you correctly, you think that morality is not subjective because it is based on a system and, consequently, an assertion can be proved to be true or wrong withing said system.
In order to be useful, it must be. But, like any definition, some people may not agree with that notion, and this can be challenging to deal with because... people are stupid.
What is a chair? Is there anything that is objectively a chair, or is a chair dependent on use and intent? And if it's dependent on use, can I point to a ball of spun sugar and call it a chair? If a chair is a thing with the use of being sat upon and supporting the subject, and a ball of spun sugar is incapable of this, we can call it not a chair (or, at least a bad chair).
If somebody questions such a notion of morality, just start deconstructing all of the words that person uses and the concepts the person is appealing to. At a certain point, you should render your opponent speechless, or at least get the person to agree on certain ground rules that are needed to form and convey useful ideas.
DarlBundren wrote:(That is, if a person were to reply ' I have a different system' you could say that theirs is wrong).
More along the lines of how I would say the person is wrong when he or she refers to a chair as a hat and a hat as a chair.
Unless I can show that the system is internally inconsistent and useless, it may still be a valid system for something: but it is not proper to call it morality.
Which system is deserving of the label "morality" is a semantic issue.
Words in and of themselves have no meaning, so good doesn't have to be called "good". In an alternate reality, evil may be called good, and good may be called evil: this is no less valid in that reality. And in that reality, because good is called evil and evil called good, it would be wrong (right) to do evil (good).
To solve semantic issues, we have to look at the purpose and utility of language itself.
Why do we call a hat a "hat" and a chair a "chair" and not the other way around? Would it be acceptable to arbitrarily switch these around, and use personal definitions?
DarlBundren wrote:The system can be proved to be better ( and so more objective) than another.
Some can be.
A person, Bob, whose system is based around doing what Bob wants -- and in order to consider other people moral, they must also do what Bob wants -- has created a system that is not properly objective. It is inexplicably and unjustifiably Bob-centric and arbitrary in nature.
However, if we defined a system by which we should do unto others precisely what those others do not want done to them, that system would be objective.
Objective, in opposition to morality, and evil.
The reason we call that evil instead of good is only as complex as the reason we call a hat a hat and a chair a chair.
DarlBundren wrote:The system is based on interests, which are universal. Every person is entitled to their own interests, so it's not a matter of what you think is right, but what they want. Correct?
Basically (if I understand what you said). However, regard for interests can be framed negatively (as explained above) or positively.
One can also construct a Discordian system wherein whether somebody should consider another's interest negatively or positively in any given decision is based on a binary randomizer (like a coin toss).