3 ms·

Thank you for prompting me to reevaluate my position. Initially I claimed, “There are no valid arguments for normative claims.”, I would like to amend my claim

Thank you for prompting me to reevaluate my position. Initially I claimed, “There are no valid arguments for normative claims.”, I would like to amend my claim. Def: An inferential system is a tuple of axioms and inferential rules. Def: An argument is a non-empty ordered sequence of statements. The last statement is called its conclusion. Claim: If an inferential system produces an argument for a normative claim, the system is either inconsistent or permits inference from positive statements to normative statements or the argument proceeds from normative axioms. Proof. Suppose S is a consistent inferential system and that A is an argument in S from S’s axioms to a normative conclusion. If S has only positive axioms then A makes inference from positive to normative statements. QED Corollary: There are no arguments for normative claims that don’t proceed from normative axioms in consistent systems that don’t permit inference from normative to positive statements.

skissane 5y ago
You are saying that, a valid deductive argument with a normative conclusion must either have a normative premise, or else rely on a logic whose inference rules permit normative statements to be inferred from non-normative ones. I agree with you there. But you haven't explained how this (obviously true) statement is relevant to the original discussion, about the morality of providing life-saving medical treatment to perpetrators of serious crimes.
- burrows 5y ago
  > But you haven't explained how this (obviously true) statement is relevant to the original discussion, about the morality of providing life-saving medical treatment to perpetrators of serious crimes. My original point was in response to the claim that “the key principle in medicine is to treat anyone who is sick, regardless of who they are”. I took this to be an ought and wanted to point out that its truth isn’t established anymore than the ought indicated by Ed’s family, and that in fact there would be serious difficulties with establishing either one for the reason given by your last post: > a valid deductive argument with a normative conclusion must either have a normative premise, or else rely on a logic whose inference rules permit normative statements to be inferred from non-normative ones. But also I think that this poster makes a good point that the doctors are actually making a positive claim. https://news.ycombinator.com/item?id=30081152 https://news.ycombinator.com/item?id=30081152 > To call an argument "valid" is to make a normative claim; it is to claim that a rational agent ought to accept it. I think I had in mind a textbook definition I had read. An argument is called valid if it is impossible for the premises to be true and the conclusion false. And then we say that an argument is sound if it is both valid and has true premises. Which seems like a useful pair of definitions, and in that context I was confused about the terms. I actually haven’t ever thought about or heard of this idea of ‘when is a rational actor committed to accepting claims’. But yeah, seems interesting to me.
  - skissane 5y ago
    That textbook definition of “valid” only applies to purely deductive arguments. Outside of mathematics, most arguments humans make have a significant non-deductive component (inductive and/or abductive). So, unless we assume that moral reasoning is closer to mathematical reasoning than to the kinds of reasoning we normally use in the sciences and humanities-a very dubious assumption-the formal concept of “validity” you are invoking has rather limited relevance to morality. That’s why I read your statement as intending “valid” informally rather than formally, since in its informal sense the term is not restricted to purely deductive arguments; and informally it is very much a normative concept (formally, I think it has multiple formal senses, some of which contain more normativity than others) I think your overall point is, you are expressing scepticism about claims to know the truth of some particular moral proposition, grounded in a more general scepticism of the possibility of any moral knowledge at all. I think that kind of general moral scepticism is (usually) a form of normative scepticism. The big problem I see with normative scepticism, is that attempts to argue for it logically and rationally turn out to be self-refuting, because logic and rationality are themselves normative concepts. Furthermore, if one understands knowledge as having a normative component, normative scepticism collapses into radical scepticism (there is no such thing as knowledge), which obviously ought to be rejected.
    - burrows 5y ago
      > That textbook definition of “valid” only applies to purely deductive arguments. Outside of mathematics, most arguments humans make have a significant non-deductive component (inductive and/or abductive). So, unless we assume that moral reasoning is closer to mathematical reasoning than to the kinds of reasoning we normally use in the sciences and humanities-a very dubious assumption-the formal concept of “validity” you are invoking has rather limited relevance to morality. I don’t have any sense for the various properties of inductive arguments. So to the extent that moral arguments are inductive, I have nothing to say. Thanks for pointing out that I’m assuming moral arguments must be deductive. Any recommendations for an introductory text on inductive arguments? > That’s why I read your statement as intending “valid” informally rather than formally, since in its informal sense the term is not restricted to purely deductive arguments; and informally it is very much a normative concept (formally, I think it has multiple formal senses, some of which contain more normativity than others) Understood. > I think your overall point is, you are expressing scepticism about claims to know the truth of some particular moral proposition, grounded in a more general scepticism of the possibility of any moral knowledge at all. Yes, that is my point. > I think that kind of general moral scepticism is (usually) a form of normative scepticism. The big problem I see with normative scepticism, is that attempts to argue for it logically and rationally turn out to be self-refuting, because logic and rationality are themselves normative concepts. This is the first time that I’ve heard of normative epistemology, which I gather is what you are talking about. One thing I will say though, thinking about this seems smoother than thinking about normative morality because (I think) I don’t have as many childhood memories of people telling me what I ought to think compared to those of people telling me what I ought to do.

skissane 5y ago

You are saying that, a valid deductive argument with a normative conclusion must either have a normative premise, or else rely on a logic whose inference rules permit normative statements to be inferred from non-normative ones. I agree with you there. But you haven't explained how this (obviously true) statement is relevant to the original discussion, about the morality of providing life-saving medical treatment to perpetrators of serious crimes.

burrows 5y ago
> But you haven't explained how this (obviously true) statement is relevant to the original discussion, about the morality of providing life-saving medical treatment to perpetrators of serious crimes. My original point was in response to the claim that “the key principle in medicine is to treat anyone who is sick, regardless of who they are”. I took this to be an ought and wanted to point out that its truth isn’t established anymore than the ought indicated by Ed’s family, and that in fact there would be serious difficulties with establishing either one for the reason given by your last post: > a valid deductive argument with a normative conclusion must either have a normative premise, or else rely on a logic whose inference rules permit normative statements to be inferred from non-normative ones. But also I think that this poster makes a good point that the doctors are actually making a positive claim. https://news.ycombinator.com/item?id=30081152 https://news.ycombinator.com/item?id=30081152 > To call an argument "valid" is to make a normative claim; it is to claim that a rational agent ought to accept it. I think I had in mind a textbook definition I had read. An argument is called valid if it is impossible for the premises to be true and the conclusion false. And then we say that an argument is sound if it is both valid and has true premises. Which seems like a useful pair of definitions, and in that context I was confused about the terms. I actually haven’t ever thought about or heard of this idea of ‘when is a rational actor committed to accepting claims’. But yeah, seems interesting to me.
- skissane 5y ago
  That textbook definition of “valid” only applies to purely deductive arguments. Outside of mathematics, most arguments humans make have a significant non-deductive component (inductive and/or abductive). So, unless we assume that moral reasoning is closer to mathematical reasoning than to the kinds of reasoning we normally use in the sciences and humanities-a very dubious assumption-the formal concept of “validity” you are invoking has rather limited relevance to morality. That’s why I read your statement as intending “valid” informally rather than formally, since in its informal sense the term is not restricted to purely deductive arguments; and informally it is very much a normative concept (formally, I think it has multiple formal senses, some of which contain more normativity than others) I think your overall point is, you are expressing scepticism about claims to know the truth of some particular moral proposition, grounded in a more general scepticism of the possibility of any moral knowledge at all. I think that kind of general moral scepticism is (usually) a form of normative scepticism. The big problem I see with normative scepticism, is that attempts to argue for it logically and rationally turn out to be self-refuting, because logic and rationality are themselves normative concepts. Furthermore, if one understands knowledge as having a normative component, normative scepticism collapses into radical scepticism (there is no such thing as knowledge), which obviously ought to be rejected.
  - burrows 5y ago
    > That textbook definition of “valid” only applies to purely deductive arguments. Outside of mathematics, most arguments humans make have a significant non-deductive component (inductive and/or abductive). So, unless we assume that moral reasoning is closer to mathematical reasoning than to the kinds of reasoning we normally use in the sciences and humanities-a very dubious assumption-the formal concept of “validity” you are invoking has rather limited relevance to morality. I don’t have any sense for the various properties of inductive arguments. So to the extent that moral arguments are inductive, I have nothing to say. Thanks for pointing out that I’m assuming moral arguments must be deductive. Any recommendations for an introductory text on inductive arguments? > That’s why I read your statement as intending “valid” informally rather than formally, since in its informal sense the term is not restricted to purely deductive arguments; and informally it is very much a normative concept (formally, I think it has multiple formal senses, some of which contain more normativity than others) Understood. > I think your overall point is, you are expressing scepticism about claims to know the truth of some particular moral proposition, grounded in a more general scepticism of the possibility of any moral knowledge at all. Yes, that is my point. > I think that kind of general moral scepticism is (usually) a form of normative scepticism. The big problem I see with normative scepticism, is that attempts to argue for it logically and rationally turn out to be self-refuting, because logic and rationality are themselves normative concepts. This is the first time that I’ve heard of normative epistemology, which I gather is what you are talking about. One thing I will say though, thinking about this seems smoother than thinking about normative morality because (I think) I don’t have as many childhood memories of people telling me what I ought to think compared to those of people telling me what I ought to do.