Ethical Theory

An ethical theory is the syntactic counterpart of an ethical philosophy: a set of sentences that expresses the philosophical ideas.

Discussion

In everyday human life, there is little need to distinguish ethical theory and philosophy; however, in SUMO and formal logic, we find it helpful to distinguish between syntactic languages and abstract semantics. Therefore, while the philosophy is what we wish to be discussing, most of the formal ethics ontology will live on the syntactic level of existence.

I chose to keep things simple: a theory will be a set of sentences. The different paradigms, types of theories, will be defined in terms of how they constrain the nature of the ethical sentences. In order to incorporate meta-ethical theories such as moral nihilism (and consequentialism), I also created a superclass of ethical theories (to avoid self-referentiality). I found it useful to introduce the idea of a justified theory: one whose sentences are supported by arguments.

Initially, I strictly defined theories, such as a deontological theory as those containing only deontological sentences, but when working on examples, such as some top-level lemmas in Gewirth’s Theory of Generic Consistency, it became clear that background theory should be included, too. The current idea is that the background theory has the purpose of being used in arguments for the deontological sentences in the theory.

SUMO

(documentation Theory EnglishLanguage "A set of sentences.")
(subclass Theory Set)

(<=>
  (instance ?T Theory)
  (forall (?S)
    (=>
      (element ?S ?T)
      (instance ?S Sentence))))

A theory is a set of sentences.

(documentation EthicalSentence EnglishLanguage "A sentence of an ethical theory.")
(subclass EthicalSentence Sentence)

(documentation EthicalTheory EnglishLanguage "A set of sentences in an ethical theory")
(subclass EthicalTheory Theory)

(<=>
  (instance ?SENTENCE EthicalSentence)
  (exists (?THEORY)
    (and
      (instance ?THEORY EthicalTheory)
      (element ?SENTENCE ?THEORY))))

(=>
  (instance ?S EthicalSentence)
  (exists (?J)
    (and
      (instance ?J Judging)
      (result ?J ?S))))

An ethical sentence is defined as a sentence of an ethical theory. Every ethical sentence should be the result of some judgment.

(documentation MetaEthicalTheory EnglishLanguage "A theory about ethical theories that is itself not technically a ethical theory.")
(subclass MetaEthicalTheory Theory)

(documentation MetaEthics Philosophy EnglishLanguage "Meta-ethics is a domain 
of philosophy concerned with the nature of Ethics.  Some of these theories 
may make claims about Ethics as a whole, so this class exists to side-step 
self-referentiality.")
(subclass MetaEthics Philosophy)
(relatedInternalConcept MetaEthics Ethics)

(theoryFieldPairSubclass MetaEthics MetaEthicalTheory)

(=>
  (instance ?ME MetaEthics)
  (refers ?ME Ethics))

Meta-ethics is defined very loosely to incorporate any philosophical theory that refers to ethics.

A formula A has the purpose of playing a role in an argument for a formula B if and only if A has the purpose of there existing an argument such that (the propositional content of) A is a premise and (the propositional content of) B is the conclusion. This predicate is used to incorporate background theory into the definitions of ethical paradigms and their associated theories.

(documentation hasPurposeInArgumentFor EnglishLanguage "This predicate (hasPurposeInArgumentFor ?A ?B) denotes that formula ?A has as its purpose the use as a premise in an argument for formula ?B.")
(domain hasPurposeInArgumentFor 1 Formula)
(domain hasPurposeInArgumentFor 2 Formula)
(instance hasPurposeInArgumentFor AsymmetricRelation)
(instance hasPurposeInArgumentFor BinaryPredicate)
(subrelation hasPurposeInArgumentFor hasPurpose)

(<=>
  (and
    (hasPurposeInArgumentFor ?A ?B)
    (containsInformation ?A ?PA)
    (containsInformation ?B ?PB))
  (hasPurpose ?A
    (exists (?ARG)
        (and
          (instance ?ARG Argument)
          (premise ?ARG ?PA)
          (conclusion ?ARG ?PB)))))

For example, in a deontological imperative theory, every sentence is either an imperative sentence or has the purpose of being used in an argument for an imperative sentence.

(documentation DeontologicalImperativeTheory EnglishLanguage "A set of sentences containing deontic attributes.")
(subclass DeontologicalImperativeTheory DeontologicalTheory)

(<=>
  (instance ?DIT DeontologicalImperativeTheory)
  (forall (?S)
    (=>
      (element ?S ?DIT)
      (or
        (instance ?S ImperativeSentence)
        (exists (?IS)
          (and
            (instance ?IS ImperativeSentence)
            (hasPurposeInArgumentFor ?S ?IS)))))))

In formalizing target-centered virtue ethics, this approach may be too restrictive: the point of virtue ethics could be seen as assiging virtues or vices to agents, yet in the target-centered approach, the point is to specify the field, basis, mode, and target of virtues to determine which acts are virtuous. Thus a target-centered virtue ethics theory could not be a subclass of virtue ethics theories; even if the theories were syntactically identical, one would need to choose which sentences are primary. One approach is to weaken what it means to be a theory: the sentences should at least referentially related to the core sentences of the ethical paradigm. I decided to make a weaker superclass of virtue ethics to encompass them both.

(<=>
  (instance ?V TargetCenteredVirtueEthicsTheory)
  (forall (?S)
    (=>
      (element ?S ?V)
      (or
        (instance ?S TargetCenteredVirtueEthicsSentence)
        (exists (?TCVES)
          (and
            (instance ?TCVES TargetCenteredVirtueEthicsSentence)
            (or
              (refers ?S ?TCVES)
              (refers ?TCVES ?S))))))))