Comments

  • Meta-ethics and philosophy of language


    Moral realists (or 'objectivists') have nothing more to support their claim than "it is not ruled out as an option".

    That's certainly not true. Robust moral realists do make positive arguments in favor of realism. Cuneo's The Normative Web and Scanlon's Being Realistic About Reasons are two well-regarded books that do just that.

    You're just not likely to see the cream-of-the-crop of philosophical arguments if you're getting them from a discussion forum rather than from the works of professional philosophers.

    I think parsimony (a principle I find mostly useful), would suggest relativism, as realism needs some objective truth-maker and we don't seem to be able to reach any kind of idea of what that might be. All that happens is the can gets kicked further down the subjective road.

    If reasoning is subjective, any principle you suggest to support relativism equally supports robust realism.

    Anyhow, this notion that there are "no ideas" of what an objective truthmaker for realism could be is related to your misguided claim above that moral realists have no positive arguments. You will realize that this just isn't the case if you read more about metaethics and moral philosophy.

    "Moral realists of this sort allow that moral facts are not natural facts, and moral knowledge is not simply of a piece with scientific knowledge, even as they defend the idea that there are moral facts and (at least in principle) moral knowledge. They thus reject the idea that science is the measure and test of all things (Shafer-Landau 2003, Parfit 2011, Scanlon 2014).

    Impressed by the plausibility of naturalism, though, many moral realists have tried, in one way or another, to show that the moral facts they are committed to are either themselves natural facts or are at least appropriately compatible with such facts (Boyd 1988, Brink 1989, Railton 1986). If they are right, then naturalism poses no special threat to moral realism."
    (https://plato.stanford.edu/entries/moral-realism/#2)

    A moral claim is taken to be correct if it somehow 'accounts for' everyone's intuitions - how do we judge if it's 'accounted' for them? Turns out that's just a subjective 'feeling' that it has.

    The problem here is that you keep referring back to judging. Yes, whether or not you think something is true comes down to a "subjective feeling" of whether or not you think it's true. This "subjective feeling" has no bearing on whether or not it actually is true, but it is something we use to assess what seems true. This is necessary to all facts, because all judgment of what is true is done in people's brains. There could be moral facts such that we could never know what they are. A moral claim is not taken to be true in virtue of accounting for everyone's intuitions. Different theories have differing views on what the truthmakers of moral claims are. Accounting for basic intuitions is something that makes certain moral theories seem more plausible. For example, a theory that what is moral is what maximizes the amount of guitars in the world is likely not congruent with anyone's reflective intuitions about morality.
  • Meta-ethics and philosophy of language


    This is a matter of principle vs practice again. Anything that can have any causal effect on something else is in principle capable of communicating with it, even if in practice they have no conventional obvious communication ability. (There are some amazing hacks that can get information off of computers not connected to any network, or monitor speech in a room with a computer with no microphone, etc, by using overlooked tiny effects between hardware and software, for example). If two things are causally isolated such that in principle no information about one of them can reach anything that can reach the other one, then from each of their perspectives the other seems not to exist at all, so they’re basically in separate universes.

    You're stretching the ideas of "communication" and "information" too broadly to justify this. We could be dealing with conscious beings that have no bodily control and just float around non-autonomously. If there are no causal links between a being's thoughts and some capacity for action, it cannot communicate in principle. People in completely vegetative states are like this. You could imagine a being who is like a vegetative human, but still aware. Everyone on the planet could be this way. If two conscious, sedentary buoys bump into each other carried atop the tides, they have not just communicated. At least, no more than me being hit by a bus while the driver is asleep at the wheel is me communicating with the bus driver.


    We necessarily imagine from some perspective or another though. If we imagine a world where we don’t exist, we imagine a world that doesn’t contain us as we really are, from some disembodied viewpoint. When we’re imagining the three men and the elephant, we’re imagining it from some viewpoint where we can observe all four of them. But if there isn’t actually any such viewpoint possible, because they’re all so completely isolated from each other, then all we should be imagining is each of their separate viewpoints, from which none of the others can ever seem to exist, nor the whole elephant, so what would it mean to claim that those things do exist, in some way that will never make any noticeable difference to anyone?

    No, this is not necessarily the case. We can say "let's imagine a world where we don't exist, in any form, whatsoever. In such a world, my mother would have only three kids, the high school I attended would have one less student..." etc. The idea that we need to stipulate some kind of omniscient observer to talk about counterfactual situations is a unique proposition of your theory. I don't see any reason why it must be true.

    "so what would it mean to claim that those things do exist, in some way that will never make any noticeable difference to anyone?"

    It means precisely that. That they exist, but in a way that will never make any noticeable difference to anyone. Do you think that if no people existed to make empirical measurements of things, nothing would exist?


    Conditional imperatives make perfect sense. It helps to remember that material implication is equivalent to a kind of disjunction: “if P then Q” is exactly equivalent to “Q or not P”. I can easily command someone to do Q or not do P, which is the logical equivalent of ordering them to do Q if they do P, or “if you do P, do Q”, without any kind of embedding trouble. It might look like there should be in the “if-then” form, but there’s clearly none in the “or” form which is identical to it.

    It might also help to resolve the appearance of the problem if we factor the “be()” out to the whole conditional at once:

    be(the saints being praised only if the demons being not praised)

    or

    be(the demons being not praised or the saints being not praised)

    "If there is a beer, then get me one" makes sense, while "If get me a beer, then there is a beer" does not and might as well be gibberish. The disjunctive form of the latter is "There is a beer or not get me a beer." I wouldn't be so hasty to claim that conditional imperatives make "perfect sense."

    "If the saints ought to be praised, then the demons ought not to be praised" does not seem equivalent to "if you praise the saints, then don't praise the demons." "You praise the saints" could be true while "the saints ought to be praised" is false.

    If you still don't see how the Frege-Geach problem presents a challenge to the idea that moral statements are inherently imperative, I'll just leave you with an article that covers the problem and the solutions that have been attempted for it. Considering that you aren't super familiar with this problem, and it is oft considered the predominant challenge for views like yours, I'd suggest that you get more acquainted with it than just talking to me about it.
    https://plato.stanford.edu/entries/moral-cognitivism/#EmbPro


    If that were a problem, then every account of what people are doing with words would be subject to the same problem. If you take an ordinary indicative sentence to be reporting a fact, as you say, that’s still doing something, but in the antecedent of a conditional is it still doing that same fact-reporting? Whatever solution allows ordinary conditionals to work there, it should also work for whatever else other kinds of speech are doing, so long as there is a “truth-value” that can be assigned to that kind of speech, i.e. each such utterance is either a correct or incorrect utterance of that kind.

    Yes, embedded truth-apt propositions are still in the business of reporting fact: that is, being evaluable as true or false and having no other baggage necessarily attached to them. What you are working with here appears to be a sort of "hybrid theory" solution to the problem. You accept that moral propositions are truth-apt. The easiest way to avoid the Frege-Geach problem for you is just to drop the necessary imperativity of moral utterances. "The saints being praised" could be truth-apt in the way relevant to your view(satisfies all appetites or doesn't), while not being an imperative at all. You could explain moral utterances with a kind of dual-purpose system, where they do whatever is required for truth-aptness(necessary, primary) and also serve as imperatives(contingent, secondary). Where you differ from traditional cognitivists is that they would hold that necessary, primary characteristic which establishes truth-aptitude could only be descriptivity.
    This is far more plausible than the inverse solution, that is, claiming all descriptive sentences to also be some sort of imperative. Some philosophers have tried this, though.
  • Meta-ethics and philosophy of language


    I think most people would (quite rightly) judge him 'ill', not 'wrong'. That's the point I was making about intuitive feelings of morally apt behaviour. They're just not relevant to actual moral dilemmas about which there's disagreement. The kind these 'moral systems' are aimed at. People supporting moral realism always seem to cite the agreement that would be had over some over-the-top act of evil, but ask yourself why have you had to choose such an example, and when was the last time you faced such a moral dilemma ("should I bash the people's skulls in or not?")? The answer to both will come down to the fact that real moral dilemmas are not solvable by relation to the instincts that we all share about empathy, care, and cooperation.

    Moral agreement isn't so much an argument for moral realism as it is a counterargument to the argument from moral disagreement. You might claim that people disagreeing about morality gives us reason to think it's relative. A moral realist can point to the disagreement about matters of physical fact(climate change, age of the earth), and the agreement about basic moral fact(child torture, genocide).
    Moral realists do not cite these things as arguments in favor of moral realism, but as a reminder to anyone who thinks that mere disagreement entails relativism.

    Possibly, yes. If it works I cannot see any reason why not. If he gets lucky a lot how are we calculating that it is just luck. If a coin lands on heads most of the time we presume a biased coin, not a lucky one.

    I'm stipulating that it is just luck. This is someone who makes erratic, unconsidered decisions. Due to purely situational luck, he has an excellent track record. Is he a good decision maker?

    Yes, I suppose you could claim that, but it gets very difficult when it comes to the more advanced areas of maths, logic, and science. You could well argue that assuming the stone follows some physical law when it drops to the ground is just a post hoc rationalisation for my gut feeling that it would, but I don't see how you could argue that the energy level predicted of the Higgs Boson being found where the theory expected it to be was just a post hoc rationalisation of our gut feeling that it would be there.

    This is a good point. We don't tend to have intuitions about these high-level concepts beyond "it makes sense" or "it doesn't make sense."

    I don't have any disagreement with all this. It's not far off the way I imagine our judgements to be made. None of this defeats relativism (there's nothing about something 'seeming' to one person to be the case, even on reflection, that makes it more likely to actually be the case).

    No, it does not defeat relativism. But neither does
    "(there's nothing about something 'seeming' to one person to be the case, even on reflection, that makes it more likely to actually be the case)"
    imply relativism. Even if my seemings are completely divorced from what is the case, that doesn't entail relativism. It entails that I'm wrong.

    Looking at the totality of our decision-making through the lens of reflective equilibrium is helpful to discard the notion that there is anything inherently special about "science experiments" for finding truth. Whether we are judging scientific data or a mathematical proof, its plausibility to us is ultimately filtered through our intuitions. Some people find what they take to be intuitions about moral facts(slavery is wrong!) to be among their most stable seemings. Many people share these same reflective intuitions, about equality, fairness, and justice. Moral theories are seen as more plausible when they explain our foundational moral seemings effectively. While relativism is not "proven" wrong, many realists find that they have just as much justification to believe that slavery is wrong as they do anything else they believe.
  • Meta-ethics and philosophy of language


    Happy birthday!

    Intentions, as I mean them, are "second-order desires", in the same way that beliefs are "second-order perceptions", though neither in quite so straightforward a way, hence the quotes here. "Thoughts" in general (beliefs and intentions) are, on my account, what happens when we turn our awareness and control inward, look at our "feelings" (desires and perceptions) and then judge whether they are correct or not. To think something is good and to intend it are thus synonymous: the thing that you think is good, that you intend, is the thing that you judge it would be correct for you to desire.

    I think this is a plausible enough account. I don't find it strikingly more plausible than moral opinions just being beliefs, but it works well on its own terms.
    Just as beliefs could be described as perceptions about our perceptions, intentions could be described as desires about our desires. Is something like this what you are implying?


    Bearing in mind that if there is no way in principle of them communicating with each other or anything they have mutual access to, then on what grounds could you say they even exist together in the same world?

    It's not clear that an ability to communicate is a necessary feature of conscious beings. I don't find it problematic to imagine a single world that contains three beings who cannot communicate with each other. A conscious being could have no ability to manipulate its own body, for instance.

    When we imagine this, we're imagining that you and I have some kind of privileged access where we're aware of all three of them and of the elephant, but they're all absolutely blocked off from awareness of each other or of any part of the elephant besides their tiny little bit. But if we can interact with them (to observe them), and with the whole of the elephant (to observe it), then in principle there is a communication channel, through us, by which they could observe each other and the whole of the elephant.

    I certainly wasn't imagining this. I was imagining three men and an elephant, not myself watching three men and an elephant. When we stipulate a hypothetical, just "what if X," it's not required that we assume we are there watching X. I can cogently say "let's assume a hypothetical world where neither of us exist." We couldn't possibly be in such a scenario to observe ourselves. We can still talk about what might be the case if it were true.

    I don't think you have sufficient grounds to claim that three beings who cannot communicate must therefore be in different worlds.


    E.g. if the dead bugs say "I don't like your hat", but there isn't actually anybody who doesn't like anyone's hat who wrote that, the dead bugs just look like that sentence, what meaning should a reader take away from it? Who should they feel insulted by? Nobody, because nobody actually wrote that.

    Yes, I agree with this. Of course, if they interpret the sentence as expressive, impressive, descriptive, imperative, whatever, this will change if and how they will truth-evaluate the sentence.

    Sure, if they take Moore's sentence to mean "I believe X but I don't believe X" or "X is true but X is false" or something like that, then they can take it to be contradictory. My account of impression/expression is an account of why it seems like it shouldn't seem contradictory, but nevertheless it does seem contradictory, i.e. why this is a "paradox" and not just an obvious either contradiction or non-contradiction. Someone who took Moore's sentence to mean one of the things above wouldn't see anything paradoxical about it, but people generally do, so some explanation of the differences and relations in meaning between "I believe X" and "X is true" is needed to account for why they do.

    I see. We're in agreement here.

    Remember that "be(x being F)" is a shorter form of something like "it-ought-to-be-the-case-that-there-is(x being F)". Likewise "is(x being F)" is a shorter form of something like "it-is-the-case-that-there-is(x being F)". These are meant to be equivalent to "x ought to be F" and "x is F"; we're just pulling the "is"-ness and "ought"-ness out into functions that we apply to the same object, the same state of affairs, "x being F".

    So if "if x ought to be F, then..." is no problem, then "if be(x being F), then ..." should be no problem either, because the latter is just an encoding of the former in a formal language meant to elucidate the relations between "is" and "ought" statements about the same state of affairs.

    By your account of what "if x ought to be F, then..." means, it is just as problematic as "if be(x being F)." Like you say, they are equivalent in meaning.

    The problem is arising due to you interpreting all moral statements as prescriptions.

    I take "oughts" to be a kind of generalization of imperatives: "you ought to F" and "you, F!" are equivalent on my account, but you can say things of the form "x ought to be G" that can't be put into normal imperative form. "Oughts" are more like exhortations than imperatives: "Saints be praised!" isn't an order to the saints to go get praised, but it is basically the same as a general imperative to everyone (but nobody in particular) to "praise the saints!" and also basically the same as "the saints ought to be praised", which likewise implies that everybody (but nobody in particular) ought to "praise the saints!"

    This is still running into the Frege-Geach problem.

    P1. The saints ought to be praised.
    P2. If the saints ought to be praised, then the demons ought not to be praised.
    C. The demons ought not to be praised.

    An exhortation is "an address or communication emphatically urging someone to do something."
    An imperative is "an authoritative command."
    The important thing about both of these is that the semantic content of them is, necessarily, an urging. The fact that you class "the saints ought to be praised" as an exhortation means that, by speaking it in P1, I am necessarily urging that the saints be praised. However in P2, I say the same words as in P1, but I don't urge that the saints be praised or not praised. So, it seems that when I say "the saints ought to be praised," the content of my sentence cannot necessarily be any kind of imperative if we want moral modus ponens to work.

    This is a problem that typical cognitivists, who would classify "the saints ought to be praised" as a claim purporting to report a fact, do not encounter.

    Perhaps we could more neutrally distinguish them as "cognitive truth" and "descriptive truth", since the most important feature of my moral semantics is rejecting descriptivism without rejecting cognitivism. On my account moral claims are "truth-apt" in the sense that matters for cognitivism, but not "not truth-apt" in the sense that matters for descriptivism. They're not telling you something about the way the world is, but they are nevertheless fit to be assigned yes/no, correct/incorrect, 1/0, "truth" values.

    Sure, that's better. I like the new terminology more.

    I would state the parallel principles instead as:

    "X is descriptively true if and only if X satisfies all sensations / observations"

    and

    "X is prescriptively good if and only if X satisfies all appetites".

    Claims that something is descriptively true or that something prescriptively good can both be "cognitively true" / correct in the same way, they can both carry boolean values that can be processed through logical functions.

    I'll remember this. My main discomfort with it is the conflation of ontology with epistemology, "what we can know" and "what is" becoming the same concepts. But, we're already discussing that as we discuss verificationism.


    The meanings, on my account, of ordinary non-moral claims, and moral claims, respectively, are to impress upon the audience either a "belief", an opinion that something is (descriptively) true, that it is, that reality is some way; or an "intention", an opinion that something is (prescriptively) good, that it ought to be, that morality is some way.

    If the meaning of regular descriptive claims has to to impress a belief, this seems like it will subject non-moral modus ponens to the Frege-Geach problem as well.

    It's technically a different question as to what kinds of states of affairs can be real or can be moral, and then a further question still as to how we sort out which of such states of affairs actually is real or moral to the best of our limited abilities. Those are topics I intend to have later threads about: ontology and epistemology, and two halves of what's usually reckoned as normative ethics which I term "teleology" and "deontology".

    I'll see you there, then.

    The difference is that I'm not claiming "X is wrong" describes some kind of abstract moral property of wrongness of object X, and that on account of that property, we ought not to do X. I claim that "X is wrong" just means "X ought not happen", which in turn is a more general, universal form of sentences like "(everybody) don't do X" or "let X not happen!"

    The is-ought distinction isn't an inherent problem with deriving moral claims from properties. For example, the following deduction respects the is-ought gap:

    P1. X is painful.
    P2. If X is painful, then we ought to avoid X.
    C. We ought to avoid X.

    A view espousing the above might be vulnerable to the Open Question Argument, but because it includes a normative premise, it's not an issue with the is-ought gap. For a non-naturalist or non-reductionist, the property of interest isn't pain, it just is "wrongness." So, a non-naturalist deduction might look like:

    P1. X is wrong.
    P2. If X is wrong, then we ought not to X.
    C. We ought not to X.

    So yes, there is a difference between your account and a realist account. But, this difference has no effect on the relevance of the is-ought distinction.

    That black hole information paradox got solved in a way that the information wasn't actually lost, because the infalling particles have effects on the stuff happening right at the event horizon, which does eventually bring the information back to us in the form of Hawking radiation. It seems like lightspeed particles moving away from us at the edge of the universe could have some impact on the other stuff there at the edge of the universe that is still capable of communicating with us, and so information about that escaping stuff could still make its way back to us in principle.

    Yeah, I suppose that could be. In that sense, every bit of matter present at the big bang could be said to have some residual effect on us. It's still possible that such an effect could be so miniscule or hard to measure that attempts to do so would require more energy than there is in the universe, or something like that. Such effects would be unverifiable in principle, but I expect that in such a case, you would say that they were literally indeterminate.

    I have a whole thing about the contingent facts about definitions (and, hence, analytic a posteriori facts) that I'm going to do a later thread about.

    I plan to do a later thread on this topic, so I'll defer answering until then.

    Is all this information already written in your book?
  • Meta-ethics and philosophy of language


    I think it's quite irrefutable that most people are either born with, or are predisposed to develop, a basic set of what we might call moral imperatives. We sense other's pain and try to minimise it, we sense other's intentions a try to help and we are drawn to to other people who appear to act the same.

    I totally agree. Humans have innate moral sensibilities. We often perceive something, and without more than a thought about it, we will judge it as right or wrong. Humans are not unique in this regard. I've seen studies implying that chimpanzees have a sense of fairness, and will freak out if they're rewarded worse than another chimp doing the same task as them.
    Gravity is another thing that humans have innate sensibilities about. We've always known that if something goes up, it must come down. If a rock is falling, we ought to get out of the way. Our intuitions about gravity are reliable, but not always. We weren't doing exact calculations until we had theories. We can say the same about mathematics. We could count objects, find symmetry in the face of a partner, divide things into groups - without theories of mathematics.
    The fact that a crude evaluative mechanism for some subject has arisen biologically within us does not mean that there is no truth to be found in that subject.

    To me, a moral dilemma is only a dilemma because the answer is not delivered to us automatically by those same set of basic instincts.

    Our moral questions are more complicated today. If people encountered a man amidst a field of fresh corpses, bashing in skulls with a rock, holding a baby upside down by the leg, they might ask what he was doing. If he replied that he was killing people purely because he felt like it, for no reason other than his own pleasure, who among the spectators would not judge him as wrong automatically? In such a case, I believe the consensus would be as good as unanimous. Do you think it'd be a less certain consensus than those about physical intuitions?

    How do you imagine reasoning actually working here - step-by-step what does it do, do you think? Say I'm trying to decide whether to wear my coat The weather report says it's going to rain, but the sky looks clear and blue. I decide not to wear my coat and enjoy a sunny day without the extra burden. Was my decision right or wrong? How do I judge the quality of my reasoning prior to knowing the outcome?

    Reasoning consists in weighing the considerations for and against a given action or belief. Everybody reasons about things at least some of the time. How do you imagine that "judging" the quality of reasoning works? Would you say it's just looking at how it worked out for you after the fact?
    Imagine someone who regularly takes unfavorable risks, is inconsistent, and barely thinks about anything at all before he does it. When he achieves his aims, it's pure luck. But, as it turns out, he gets lucky a lot. Is a person like this a good decision-maker? Should he be in a leadership role, or working as a consultant?
    I don't deny that the outcome is important. If I left my jacket and it's sunny, I'm happy with the outcome! If I left my jacket and it rains, I am not happy. When I'm standing there wet, I might think that I made a bad decision to leave my jacket at home. This could be a psychological trick: think the poker player who goes all-in on an incredible hand and unluckily loses. He might think he made a bad decision, but did he really? If he has the same hand again, all else being equal, he should go all-in again. He made the best decision he could with the information he had available. A good decision can result in a bad outcome, and a bad decision can result in a good outcome.

    So, the point here is that we'd never know. A moral system based on reasoning would be completely indistinguishable from one based on gut feeling because we'd have absolutely no way of telling if the reasoning is post hoc rationalisation of what we we're going to do anyway, or genuine reasoning.

    Couldn't you say the same about math, or logic? At the end of the day, we only believe that the Law of Non-Contradiction is true because we really, deep down, feel like it's true. Are we just post-hoc rationalizing our gut feeling? Perhaps, but this isn't obvious.
    We could even say the same about our physical intuitions. Deep down, I feel like it's true that larger objects can't fit inside smaller ones. But how do I know? I haven't tried to fit every object in the universe inside every other. What if I met someone who claimed otherwise?

    what I find disturbing about all these "I've worked out how to decide what's moral or not" type of models (we seem to get a half dozen of them every week) is that they try to add a gloss of authority to moral resolutions which we have absolutely no way of distinguishing from gut feeling (or worse, political ideology).

    Have you read much moral philosophy?
    Given your concerns about reasoning, I'd like to introduce you to the concept of "reflective equilibrium." This is the idea that the beliefs we are most justified in holding are the ones that have, upon the most reflection, remained consistent. Reasoning is done by comparing "seemings," or "things that seem to be the case." These seemings are defeasible: a less convincing seeming is often discarded in favor of a more convincing one.
    The bedrock of this system are those seemings which, after the most reflection(which consists in examining seemings and comparing seemings), are the most stable. Take your example that larger objects cannot fit inside smaller ones. This seeming has been consistent with everything you have ever experienced in your life. Not only this, but it seems intuitively true based on what "larger" and "smaller" mean. I'm sure that the more you consider it, the more sure you are of it. What would it take to defeat this seeming?
    The more we consider our seemings, the more we approach reflective equilibrium. We do this in our own minds, and we do it with other people dialogically. The most important thing to remember is that we could always be wrong. As you rightly point out, reasoning is fallible. It falls back on judgment. But, this alone should not discourage us from our pursuit of truth.
  • Meta-ethics and philosophy of language

    I don't see any widespread agreement on those matters. Torturing someone for no reason, is just definitional,what distinguishes actions (the things to which the term 'moral' applies) are those reasons,and it's that matter over which there's so little agreement.

    I said "torturing someone for no reason other than personal pleasure."
    Consider, "murder is wrong." Since "murder" means "unjustified killing," it seems almost trivial, since you would say that what matters is whether or not people agree that a given killing is murder. While people might disagree about whether a given killing is in fact sufficiently justified, they agree that if it is unjustified, it is wrong. There seems to be a need for killing to be reasonably justified in a way that we don't need to justify, say, going for a walk.

    It is, but the evaluation includes the outcome in a way that evaluation of arguments doesn't.

    To take your lottery example. Imagine you bought those thousand rickets and won a thousand times, you do the same next week and again the week after, the same thing happens. Is it still a bad decision, simply because it 'ought to be' on the basis of the evidence? Clearly the success of the outcome must cause us to review our assessment of the decision, we must have got something wrong somewhere.

    "The success of the outcome causing us to review our assessment of the decision" is reasoning. If it seems that I can consistently win lotteries, entering lotteries might appear more reasonable. Remember, I did not say that reasoning makes a decision better. I said that more well-reasoned decisions are likely or often better decisions. I might make a well-reasoned decision to go to the bank today, and then get struck by lightning the moment I step out the door. This would be a terrible outcome, but this doesn't mean my decision to go to the bank was not likely to have a good outcome. Its likelihood to have a good outcome was probably a large part of what made it seem to be a good decision.

    But past probabilities do not affect future probabilities. In considering the lottery, none of my past wins indicate a heightened probability of a future win. This is just like how, in a series of a hundred coin flips, getting tails 99 times does not mean there is more than a 50% chance my next flip will be heads.
    The exception would be if I have been winning the lottery due to a reliable cause, like a "guardian angel" arranging that I will usually get a winning ticket. If I have reason to believe I have such a guardian angel, I may always have reason to buy another ticket. Otherwise, no matter how many times I have won in the past, buying another ticket is always a bad decision.

    In addition, what we believe are reasoned decisions are very often not. Even High Court Judges hand down longer sentences when they're hungry than they do when they're not.

    I'm sure this is the case. I don't think human beings are flawless automatons of reason. People often take themselves to have a reason to do or believe something, and then later realize they were mistaken. Sometimes, they never realize they were mistaken at all.


    No, I suspected not. Perhaps that's for another day though.

    If you do want to get into this, define what you mean by "subjective" and explain how you think it applies to reasoning.
  • Meta-ethics and philosophy of language


    I can't think if a single moral fact that everyone agrees on, and not many that are agreed even by a large majority. I can say, however, that virtually everyone in the world agrees on the physical properties of tables, or the physical functioning of a cup. That solid things cannot pass through other solid things, that large objects do not fit inside smaller ones...etc.

    You raise a good point. There are facts about physical reality that are extremely basic, agreed upon by practically everyone. There are also physical matters-of-fact that are much more contentious. Likewise, there are moral facts that enjoy the agreement of a vast majority: that torturing someone for absolutely no reason other than personal pleasure is wrong, for instance, or that committing genocide is worse than donating to charity. The consensus may not be as high as our most basic physical intuitions, but I'd bet you it's close.

    I'm interested in how you come to believe (and defend) whatever it is you believe.

    What are some arguments against, say, cognitivism, that you'd like to see me respond to?

    To answer your question though - I think more well-reasoned arguments are better arguments by definition. The measure we usually use to determine 'better' when it come to arguments is the the quality of the reasoning (a subjective judgement, I might add, but nonetheless the case).

    Funnily enough, I don't think that quality of reasoning is subjective.

    Decisions, however, are not usually judged 'better' on the strength of their reasons, they're usually judged on the evaluation of their outcome, so the two are different.

    Evaluating the outcome of a decision is reasoning about it.

    Buying a thousand lottery tickets is a poorly reasoned decision. It is a gamble which has an incredibly small chance of resulting in your desired outcome. Perhaps one of your tickets is a winner. You win more than enough to compensate what you spent on tickets. Do we judge your decision as good, based on the outcome? Or was it poor, based on the reasoning? Would you repeat your decision if you had the opportunity to buy a thousand tickets for the next lottery? Would you recommend that your friends and family do the same?
  • Meta-ethics and philosophy of language


    Then whence the notion "It is not an impasse like you would expect if people's disagreements were just brute expressions."? It seems like an impasse, and you were previously imploring that we treat things the way they seem to be until we have good reason to believe otherwise.

    Perhaps I should have said moral disagreement is not at more of an impasse than disagreement about things we all agree are matters of fact. On the individual level, views on issues like abortion are changed. There are also people at metaethical impasses, but this alone does not push us to conclude that discussion about metaethics is noncognitive. If you thought this, you wouldn't talk about "reasons for believing expressivism" at all. "There are pages full of reasons for expressivism" would just be you expressing "Woo, expressivism!"

    Do you have any reason to believe this?

    Do I have any reason to believe that well-considered decisions are often better decisions?
    Well, a reason is a consideration in favor of X. If, after deliberation, I conclude that X is what has the most/strongest considerations in favor of it, this is what I think I have the most reason to do. This means that I think it's the best thing to do. You're asking me, right now, to consider my reasons for a belief. You think I've reached the wrong conclusion about what the most reasonable thing to believe is, right? You implore me to review my beliefs by exposing them to compelling arguments. Are more well-reasoned arguments likely better arguments? Are well-reasoned positions often better positions? If so, why not believe that well-reasoned decisions are often better decisions?
  • Meta-ethics and philosophy of language


    Isn't it? You expect the dispute about abortion to be resolved any minute do you? What an endearing sense of optimism you have!

    I'm not quite that optimistic! We haven't even resolved the dispute about vaccines, or the dispute about flat earth.

    1) Decisions which we call 'moral' ones are actually a very wide range of decision-types involving (sometimes very different) areas of the brain. I'm loathe to make absolute statements, but one I think I could stand by is that moral decision-making definitely is not one unified thing. It is several disparate and possibly even mutually exclusive processes depending on the exact nature of the decision. we do not involve the same process in deciding to care for a baby as we do in deciding to give to charity.

    "The range of moral criticism, as most people understand it, is broad. Various forms of behaviour, such as premarital sex, homosexuality, idleness, and wastefulness, are often considered immoral even though they do not harm other people or violate any duties to them."
    -Scanlon, What We Owe To Each Other

    I agree with your assessment. There are many unrelated topics that we ascribe moral value(or disvalue) to. It is not surprising that the decision-making processes we employ are diverse, given the scope of what we consider to be "moral decisions."

    2) The processes used for for any given decision-type vary across people, ages and circumstance. At any given time a decision might be made on the basis or norm-following, rules, consequences, emotion, empathy... At any given time this decision might be something we're consciously aware of, or something we're process sub-consciously.

    I agree with this. What winds up making our choices, at the end of the day, is variable. At least some of the time, though, we make decisions based on reasons. These well-considered decisions are often better decisions.
  • Meta-ethics and philosophy of language


    Who's this 'we' and from where are you getting your assessment of usually? I certainly don't, and neither do any of my colleagues. We wouldn't get very far understanding the role of things like social influence or group identity if we just presumed everyone meant exactly what they said. If someone after work tells you the would 'kill for a beer' are you concerned for their sanity, or do you presume they didn't literally mean what they said?

    That's begging the question. Their surface level speech already matches what they claim it means, that's why they claim it means it. Why would they match their surface level speech to what cognitivists claim it means? If "You shouldn't murder" means "Boo Murder!", then the expressivist has nothing to change in their surface level speech, "You shouldn't murder" means exactly what they intend it to.

    Clearly your experience differs from mine here. My experience with people discussing morality usually consists in them putting forward claims, like "abortion should be permissible," and supporting their position with reasoned arguments. It is not an impasse like you would expect if people's disagreements were just brute expressions.

    And we have plenty of them. People have been studying this for hundreds of years, we're not coming to it fresh just now. Have you read the works of any expressivist philosophers? What is it that you think the pages are full of if not reasons? Are they to you just a series of blank pages with ",,,and therefore expressivism" at the end?

    Yes, proponents of expressivism provide pages of considered arguments for expressivism. This is just like how proponents of cognitivism provide pages of considered arguments for cognitivism, proponents of quasi-realism provide pages of considered arguments for quasi-realism...

    Pretty much every theory in philosophy has "pages full of reasons" in support of it. I can't believe every theory at once. I could suspend judgement until I have read every book arguing for or against a position, but that would leave me unproductively agnostic on many things.

    Just mentioning the name of a problem doesn't really help anyone understand why you think it's applicable. Do you think expressivist philosophers are unaware of the Frege-Geach problem? If not, then presumably they don't think it makes their position less tenable, and they presumably have their reasons. So what's relevant here is not merely the existence of a reason to find their position untenable, it's why you find that reason compelling.

    I've already talked about the Frege-Geach problem and why I find it convincing earlier in this thread. It came up in reference to Pfhorrest's prescriptivism.
    No, I don't think expressivist philosophers are unaware of it. Nor do I think they find it insignificant: as far as I'm aware, every noncognitivist theory at least attempts resolution of embedding problems, including the Frege-Geach problem. Some people think these attempts succeed, others think that they fail.

    Is there any particular literature on expressivism you recommend?

    Again, you're just making an assumption that the manner of speech dictates how the world is. when someone is determining whether the leaf is green, the ventromedial prefrontal cortex is barely involved, neither is the insular cortex. Both are heavily involved in judging something like murder being wrong. The ventromedial prefrontal cortex is responsible for regulation of emotional affect and the insular cortex is involved in feelings of disgust and visceral somatosensation. If the question "is murder wrong" was like the question "is the leaf green" then why would completely different brain regions be involved?

    So, emotional areas in our brain are more active when we make moral judgments. People certainly have moral sentiments. No cognitivist would deny that. Is the fact that people react more strongly when judging murder than when judging a leaf surprising? People care far more about wrongness than they do about greenness.

    So why would we presume now that it's grammatical structure had the same meaning? When determining whether to murder someone (or commit some other immoral act) one does not consult one's database of actions to see what property this particular one has attached to it, one consults a wide variety of emotional responses. So, if that's what's actually happening when we make moral choices, then why would our moral talk be all about assigning properties to behaviours, properties which are barely consulted when they consider one of those behaviours?

    Emotional responses and deliberative thought both play roles in deciding what actions we take. What do you make of people who claim to employ reason in moral decision-making?
  • Meta-ethics and philosophy of language


    I’m just some isolated guy who wants to actually talk to other people about the philosophical system I’ve been brewing for a decade instead of just writing it down in my book that nobody will ever read.

    Hey, don't be pessimistic about it. Are you going to publish? The only negative thing I would say about the sort of "theory-of-everything" approach you have is that it requires a reader to sacrifice more of their initial beliefs than a single, cohesive "theory-of-one-thing." It's a lot harder to convince someone to adopt a wholly unique theory-of-all-issues than, say, simply fit a meticulously argued panpsychism into their belief system. But, obviously, trying to fit it all together is an impressive undertaking.

    The distinction I make between desire and intention is important here. Being a slave to your vices is a case of your intentions not being causally effective on your actions, e.g. you mean to do something, you think you ought to, you resolve to do that, but despite that you just can't help but do otherwise, because other desires besides the desire you desire to desire override it.

    This is analogous to mirages and optical illusions. Sometimes you perceive something, and you know that perception is false, you judge that your perception is incorrect, but that doesn't stop you from perceiving it anyway. You perceive something you don't believe. Likewise you can desire something you don't intend. As just like you might not help but act on some perceptions even though you disbelieve them (e.g. recoiling from a scary hallucination you know isn't real), so too you might not help but act on some desires you don't intend to.

    This can be explained easily in terms of second-order desires. I might have a certain desire, but desire not to have that desire. In such a case, the "desire I wish I had" is not affecting my actions, but the "desire I do have" is. Being a slave to my vices might be my first-order and second-order desires conflicting. If X being good is a belief, I can sincerely hold that X is good and yet not intend to X. While I want-to-want-to-X, I unfortunately don't want-to-X.

    In the case of the gym example, if you honestly don't intend to go to the gym, rather than just not desiring to and expecting those slothful desires to win out, then that would suggest that you think there is some greater good than going to the gym that you would be neglecting, so you think you going to the gym today is not actually good (because in the full context you think doing so would be worse than not doing so), even if you think going to the gym generally is good.

    This seems to imply that, in deliberating different options, only the maximal one is "good." This precludes the idea of deliberating between multiple good options. It doesn't follow that just because going to the gym is not the most good, it is not good at all.

    "Going to the gym is good, but I don't intend to go to the gym." could reflect that I am not going to the gym because of a conflicting greater good, I agree. But, I can still cogently evaluate going to the gym as good, while not intending to do it.

    The truthmakers of moral claims, on my account, are the appetitive experiences of things seeming good or bad in the first-person. Not the sensory experiences of other people seeming, in the third person, to be having experiences of things seeming good or bad. Just like the truthmakers of factual claims, on my account, are the sensory experiences of things seeming true or false in the first-person, not the sensory experiences of other people seeming, in the third person, to be having experiences of things seeming good or bad.

    This avoids solipsism or egotism, as I expect you'll object next, because you can either trust other people that they had the first-person appetitive experiences that they claimed to have, or go have those same experiences yourself if you don't trust them. In either case, it's that first-person experience that is the truth maker, not any third-person account of a fact that someone's brain is undergoing some process.

    So, if I understand you correctly, experiencing a first-person appetite is an irreducible ought. I'm still not confident that this is the case. The fact of my appetitive experience is a physical fact, even from the first-person perspective. If I were in pain, I could, say, go through an MRI and verify that I am in fact in pain. Of course, I don't need to do this to have the experience of being in pain. Similarly, I don't have to go through an MRI to verify that I am having a perceptive experience of looking at a painting. If where I have pain, I have an ought, there seems to be a relationship between "me factually being in pain" and "what ought to be the case."
    In the same vein as above, do you think that
    "I am hungry, yet I ought not to be fed." is some kind of paradox? If "I am hungry" entails "I ought to be fed," it must be.

    Who or how many people is not relevant. But in any case bloodlust or sadism as usually defined would be desires, not appetites. Someone desires to kill or hurt someone else. That doesn't mean that a complete moral account has to grant them what they desire. But whatever raw experiences they're having that give rise to those desires, whatever kind of psychological pain or whatever may be behind it, those are appetites, and need to be satisfied.

    Sure. Bloodlust could easily be defined in terms of an appetite, though. What makes hunger an appetite? It is an experience of a certain sort that can only be satisfied by an actualization through my sensations, in the case of hunger, me being fed. You classify pain as an appetite as well. It, also, is an experience of a certain sort that can only be satisfied by a sensory actualization. In the case of pain, the appetite is to be relieved of the pain, and a physical coming-to-be in accordance with my appetite satisfies this. Similarly, bloodlust is an experience of a certain sort that can only be satisfied by a coming-to-be of seeing someone else hurt. It might feel like an anger-based appetite of a sort. People do experience things like this. Consider the phenomenon of "cute aggression," where people often report wanting to tightly squeeze and hurt adorable animals. This cannot be explained at the level of desire, as people who experience cute aggression have no desire to hurt these animals. Their experience is at a deeper level, which can be contradicted/overridden by desires, and I believe it could be accurately construed as an appetite under your view.

    Even if you don't think that humans have this bloodlust-appetite, you could imagine a being that does. We would still need to factor its appetite into our consideration under your theory. I find this to have troubling implications if sinister appetites "need to be satisfied," as you say. The idea of an appetite that necessitates the suffering of others does not seem inherently contradictory, impossible, or even hard to imagine.

    Consider natural numbers for analogy. There is no natural number that could not, in principle, be counted up to. But we can never finish counting all of the natural numbers. No matter how many we count, there will still be infinitely many that we haven't yet counted. But those are still nevertheless in principle countable: if we keep counting we will eventually count any number you'd care to name, but there will always still be more that we haven't counted yet.

    This makes sense, and does a good job at resolving the apparent contradiction I had pointed out.

    If there was truly absolutely no way in principle to ever tell anything about the object they're feeling than the things they "mistakenly" feel, then yes, those things would actually not be mistakes.

    But there is still an elephant. They may not be making methodological mistakes, but their perceptions of various objects where there is in fact an elephant represents a deviation from what is actual.

    It's important to keep distinct things that are "practically impossible" and things that are really and truly impossible in principle. There are lots of cases where it's "practically impossible" to verify something, but still actually possible in principle, and it's that in-principle that makes the difference. If you extend the "practical impossibility" to ridiculous lengths, you end up getting ridiculous-sounding conclusions, and if you take it impossibly far all the way to complete actual impossibility, you get ridiculous conclusions like these three men existing in actually separate worlds.

    Yes, this is an important distinction to make when discussing hypothetical situations. I defer to Parfit again here, where he draws the distinction between what is "deeply impossible," such as a square circle, versus merely "technically impossible," like teleportation. Your goal here seems to be to define a satisfying constraint for "in principle," so that when we say "something is true of the physical world if and only if it is verifiable in principle," we don't get lost in the weeds over what is merely practically impossible.

    In reading the message written in dead bugs, we necessarily interpret it as though it was an utterance. Part of what makes something an impression or an expression is the interpretation of the audience; really, it's more the audience's interpretation than the speaker's intention that conveys any kind of communication at all.

    I definitely don't think that we necessarily interpret it as an utterance. "Expression" and "impression" seem to be, as you have so far used them, properties related to what a speaker is intending with their words. What you're now saying implies that if I take someone to be impressing a belief on me, they are in fact doing so.
    In the case of the dead bugs, applying "expression" or "impression" in the intentionary sense is a clear error. What am I applying them to? The dead bugs? The piece of paper? If a piece of paper that was written on by nobody can cogently be claimed to impress/express beliefs, the terms become a lot less meaningful. I can similarly say that a mountain impresses on me the belief that it is large when I look at it.

    So if you read what seems to be an impression written in dead bugs that happen to have died in that pattern, and you read it as an impression, not just as a meaningless pattern of dead bugs, then to you it is an impression.

    As above, why then are impression and expression properties of utterances at all? They could be properties of anything that brings me to have a certain belief. Also, didn't you claim that the impression/expression distinction resolves Moore's paradox? With these clarifications, you will have to retreat to the weaker claim that they only resolve Moore's paradox from the perspective of someone who assigns impression/expression to the sentence in the same way that you have. If their evaluation of impression/expression differs from yours, they are still correct in taking Moore's paradox to self-contradict. You might be fine with this conclusion.

    Likewise, in my logic, every ordinary indicative descriptive sentence "x is F" can be turned into "is(X being F)" and all the same rules of logic will apply to them. But if there was another sentence "x is G" where G is a predicate meaning that what it's applied to ought to be F, instead of treating that as a completely unrelated sentence, we could render it as "be(X being F)". You can do the exact same rules of logic to that re-encoding as you would with "x is G", but you can also see relationships between that statement and the statement "x is F" and generally other statements about x being F.

    P1. be(X being F)
    P2. If be(X being F), then be(Y being E)
    C. be(Y being E)

    This restructuring alone does not resolve the Frege-Geach problem. A statement of the formal form be(P being Q) is a prescription. Its utterance prescribes a state of affairs. This is clear in the case of the atomic sentence P1: be(X being F), or "it ought not to be the case that there is stealing," if you'd rather. Prescription is the semantic content of that proposition. In the context of the antecedent in P2(IF be(X being F), then...), the semantic content of "be(X being F)" does not prescribe. We see, then, that be(X being F) in P1 and be(X being F) in P2 have different semantic content. In classical logic, we would now have to use different variables to represent them, and our modus ponens would no longer be valid.

    If you want to avoid this by "abstracting the prescription away" or something similar, that is where you would conflict with classical logic. The relevant semantic content of a proposition cannot just be abstracted away, or else we get things like "Close the door!"(an imperative) being logically equivalent to "The door is closed."(a description). The semantic content of a proposition is what makes it mean what it does. If you change relevant parts of that content, it becomes a new proposition.

    I don't have any problem with talking about "truth" in the broader sense that you are. I only substitute "correctness" because to some people, "truth" implies descriptiveness, and I want to avoid confusion with them. You're absolutely right that there's two concepts here, a narrower one and a broader one, and I don't have any particular attachment to the terms used for them, just so long as they are distinguished from each other.

    The importance of using "truth," for me, is the consistency with the vernacular in logic. Formal logical relationships, most crucially entailments, are based in the truth or falsity of their propositions. Claiming that something is "not truth-apt" should imply that it is not evaluable in formal logic, not usable to form valid arguments, etc. Since you use "correctness" in the same way "truth" is used in logic, I think it is important to call it truth.

    The narrower concept, though, is not specifically "physical truth", but "descriptive truth". On my physicalist account, those are identical, but for the purposes of logic, they don't have to be. Saying that there exist some abstract moral objects is a descriptive, but not physical, claim. It would be made true by reality being a certain way, by it being an accurate description of reality. Likewise any kind of claim about the existence of any nonphysical things: they're still claims that things are, really some way.

    I don't see "physical" as analogous to "descriptive," for the reasons you state. The nature of how you conflate the two is a little confusing to me. I would expect that you would entertain that nonphysical facts could be described truthfully, if they really existed, but they do not, ergo descriptions aiming at them are false. "God is real" would not be some weird quasi-proposition that we can't make sense of, it would just be a false proposition. It makes less sense to hold statements concerning nonphysical descriptions as not truth-apt at all, as you admit these things could be made true by reality being a certain way.

    There are other kinds of claims that don't purport to describe how reality is at all, though; most notable, prescriptive claims, that something ought to be, morally. I think that those kinds of claims can also be true in a sense, but a non-descriptive sense; hence I try to say "correct" instead of "true" to avoid possible implications of descriptivism

    I have your principle of descriptive truth down, I think: "X is a descriptive truth if and only if X is, in principle, empirically verifiable as true."

    What I'm after is a similar principle for your establishment of moral correctness. "X is a correct prescription if and only if..." what? You have explained your idea of prescriptive correctness, but I've found these explanations a little vague. I'm looking for a definition of what "prescriptive correctness" really is, in the form of a principle, like the above for descriptive truth. I'd have trouble trying to just throw "appetitive experiences of seeming good or bad in the first-person" into a principle like that without horribly misinterpreting and butchering what you're saying.

    (Besides its conflict with my physicalism, I would still object to moral non-naturalism because it's still a form of descriptivism, and so still tries to draw an "ought" from an "is": there are these abstract moral objects, therefore things ought to be like so-and-so.)

    Maybe some non-naturalist theories work like you say, but this is not a necessary feature of moral non-naturalism. If the non-naturalist claims "murder is wrong" is a fact, it is a normative fact. The relationship between "murder is wrong" and "you ought not to murder" is that "you ought not to do what is wrong." The non-naturalist has no problem assuming this premise, and neither do you. In fact, you seem to espouse(in agreement with many realist positions) that what is wrong is implicitly that which you ought not to do. If this counts as crossing the is-ought gap, then you cross it yourself when you hold "murder is wrong" as interchangeable with "it ought not to be the case that there is murder." Also, the is-ought gap in no way implies that there is some fundamental divide in our language between two types of diametrically opposed claims.

    Something like it yes, but not exactly that. My fork would have three tines. In the middle are the relations between ideas, without any attitude toward them, neither claiming those ideas are correct to describe reality with nor correct to prescribe morality with, just that those ideas have those relations to each other.

    If you are going to assert that mathematical truths are merely truths of relations between ideas, belonging in the middle of your fork, you need to support that. This would require an account of analyticity. Distinguishing between "things that are true by definition" and "things that are true in virtue of a contingent fact" is much harder than you might initially think.

    I think the thing I said earlier in this post about distinguishing practical possibility from in-principle possibility addresses this.

    Somewhat, but also somewhat not. Take the example of information on the edge of the observable universe. It is, even in principle, impossible for us to verify that information. This is because part of what it means to be us is that we are here, and if we cannot verify some information from here or somewhere we can get to from here, it is not verifiable at all to us. You could stipulate that if someone was on the edge of our observable universe, that information would be verifiable to someone. So, in one sense it is verifiable "in principle," but it is not verifiable "to us" in principle.

    (I spent that entire thread trying to disengage with Isaac and never got around to actually presenting an argument for those principles, but if you want to ask about them over there, I will).

    My condolences.

    I know I'm weird in not sharing the inherent sense of outrage about "infidelity" that other people have, and it's not because I want to run off and do it with other people myself, but because I'm just honestly not a possessive person. I've been in open relationships before and not felt jealous, unless it actually had negative consequences for me (like, she wants to go spend more time with him and leave me alone and lonely). If you got rid of all the negative consequences -- which may not always, in practice, actually be possible -- then I don't see how it would be a genuinely bad thing.

    The reason I chose infidelity is because it is a type of dishonesty, betraying a mutual expectation that is present in many relationships. Obviously, in a relationship where fidelity is not a value that a reasonable expectation is built around, infidelity is not a serious transgression. I found the example compelling because most people would prefer to know that their partner was cheating on them, if they could so choose, rather than maintain the blissful false belief to the contrary.

    This is where a lot of my core principles come from. When we get down to questions that we cannot possibly find an answer to, but cannot help but implicitly assume some answer or another by our actions, we have to assume whichever way is most practically useful to finding other answers.

    I think this is pretty true. I think "deep skepticism" is vastly overrated. It is easy to live in a nice house, in a first-world country, with your pick of luxuries, and wax philosophical about how we can doubt everything we believe. Someone needs to be living in a hole, with nothing, before they can truly be said to have embraced infinite doubt. Who among us truly lives as a solipsist?

    However, I do not conclude from this that we must assume something like solipsism could not possibly be true. I would just conclude that we should not afford much importance to its possible truth.

    (there are practical reasons to reject reducing "is" to "ought" or vice versa, which rules out translating moral utterances to descriptions of reality, be it a natural or non-natural part of it).

    This is not what the is-ought gap establishes. Maybe you should go into your reasoning a little more here.

    Wrapping up this post, let's return to verificationism about truth once more. This is a tangentially related epistemology argument, so I thought I'd put at the bottom. Let us focus our lens on the concept of verification itself. When we speak of verifying something, we speak of confirming it to be true. What does this mean? What is the difference between something which is truly confirmed, and something which we are merely justified in believing?
    It seems intuitive that if we have genuinely confirmed that something is true, we know it to be the case. In fact, you claim that what is true is limited by what we can know. This raises the question: what can we know? There are two ways to take this question. One is methodological:
    "How do we know that we are not mistaken in thinking we know X?"
    One is definitional:
    "What conditions would a person have to satisfy for them to have knowledge of X?"
    The latter is my question to you. If what can be true is constrained by what we can know, then before we ask what is true, we ought to ask what we can know. Before we know what we can know, we must know what it is to know at all. So, what conditions does a person have to satisfy to have knowledge of X?
  • Meta-ethics and philosophy of language


    So, on what basis are you judging 'best' here? What aspects does this explanation have which, say expressivism, doesn't have?

    Usually when the semantic content of a type of discourse is a certain way, we use that discourse in ways that match the semantic content. An example of expressivist discourse is cheering for sports teams. We see people yell "Go, Canucks! Woooo!" and take them to be expressing their approval for the Canucks. If we were going to assume that when people said "Go, Canucks! Woo!" they were actually meaning "The Canucks are the best team in the NHL." we would need a reason to do so.

    "Murder is wrong" is structured the same as "The sky is blue" or "The economy is failing." If we were to take "murder is wrong" to be a blunt expression of "Boo, murder! Grrr!" we would need a reason to do so. We usually don't assume that people are making a category error in their speech. People's moral discourse couldn't be substituted for growls and cheers and retain the same meaning. I've never seen an expressivist who puts their theory into practice and changes their surface-level speech to match what they claim it means. Presumably, people would look at them very funny, and not understand "keeping promises is good" to be equivalent to "Wooooo, keeping promises! Hell yeah, woop woop!"

    There are also formal issues with expressivism that make it less tenable, like embedding problems or the Frege-Geach problem.

    Well then those are not properties of the behaviour in question, they are properties of those regarding it. I suppose you could say that a property of slavery is that such-and-such a group of people think it's bad, but that seems like an unnecessarily clumsy way of just avoiding assigning the property to the person rather than the behaviour.

    Really? I think describing the property as being assigned to the person is far more clumsy. If anything, a relativist is assigning the property "Wrong-according-to-me" to "slavery," not assigning the property "Thinks-slavery-is-wrong" to himself. Again, compare to a color judgement. When I say that grass is green, the content of my sentence does not include myself. If somebody wanted to see if they agreed or disagreed with my judgement, they wouldn't check me for the property "thinks-grass-is-green," they'd check the grass, because that's where the alleged property "green" is. If color perception is subjective, we could disagree but both be right.

    Then on what grounds are you dismissing methods?

    On what grounds am I dismissing appealing to consensus as a guiding principle for truth?
    Well, presumably for the same reasons you dismiss it. Laypeople's views are usually uninformed and have not been subjected to serious scrutiny. Like you said,

    what you'd be setting up here is the principle that we should, no matter the absence of mechanism, no matter the lack of parsimony, take as true whatever it is that most people believe to be true. "Most people believe there are objective moral facts, therefore we should assume there are objective moral facts".
  • Meta-ethics and philosophy of language


    Maybe (although not in my experience), but you advanced this as evidence of us ascribing properties to objective entities (like slavery), so in the case of relativists they would not be ascribing the property 'wrong' but the property 'something I believe is wrong'. In order to take moral realism as prima facie true on the basis of it seeming to be how we talk about moral dilemmas, you'd have to present evidence of us mostly talking about moral dilemmas assuming realism, and that's just not the case.

    Definitely never said anything about "objective" entities. I think that, even from a relativist standpoint, assigning the property "wrong" to "slavery" best explains what is happening in a moral judgement. If I say "the movie was good," I am assigning the property "good" to the movie. I am just doing so in a subjective manner. "The movie is good" might be true when spoken by someone who liked the movie, but false when spoken by someone who disliked the movie. Similar for something like "broccoli tastes good," or if you are a certain type of moral relativist, "murder is wrong."

    We talk about moral facts assuming cognitivism, but not necessarily the kind of realism you're imagining. The case I have been advancing here is agnostic to the matter of strong moral realism. Cognitivism holds that:
    (1) Moral statements are truth-apt(can be true or false)
    (2) At least some of them are true.

    Moral relativism is a cognitivist theory. Moral relativism holds that moral statements are truth-apt, and sometimes true. Their truth is dependent on subjective facts: different types of moral relativists might believe these facts to be an individual's attitudes, or the consensus of a society, or something like that.

    But this would seem to support the opposite of what you're saying. We advance reasons whilst talking about how good certain movies despite virtually all of us being of the opinion that movie preferences are subjective. So what this example demonstrates is that our mode of conversation (reasoned argument) and our use of terms like 'best' does not in any way indicate that we consider the underlying judgment to be objective. It's just the way we talk.

    Yes, moral cognitivism is compatible with moral relativism. The way we talk about movies gives us good reason to believe, prima facie, that our statements about movies are cognitive. I wouldn't infer anything from this alone about whether movie evaluations are objective or subjective.

    I might have your position confused then, I thought you were arguing against moral relativism. I'm a relativist myself, so we're in agreement here.

    I am not personally a moral relativist, but I was not arguing against moral relativism here. I was disputing Pfhorrest's version of prescriptivism, which makes claims to its own unique sort of cognitivism, even though how he has described it so far is not compatible with the tenets of cognitivism I mentioned above.

    Then what should?(serve as a guiding principle for determining what is true)

    Our methods should probably depend on what it is we're trying to learn truth about. I can't confidently state some principle about how we can reliably come to apprehend truths in all circumstances.
  • Meta-ethics and philosophy of language


    As I raised with Pforrest earlier, this is simply not true so as a basis for believing prima facie in moral truths it's sketchy at best.

    Which part is simply not true? When I said this, referring to our use of moral language:

    "We debate, disagree, and use reason to draw conclusions from premises. When we disagree, we believe that one or both of us are incorrect."

    This has been true in my experience. Even moral relativists do this. They advance their various viewpoints, disagree about moral issues, and believe that others are incorrect(only relative to their own morality, rather than an objective one). People discuss topics we consider subjective, like how good certain movies are, in reasoned ways all the time.

    Currently, a significant number of surveys show most people to be moral relativists (or at least not moral absolutists).

    That's surprising to me, considering that most people are religious. If this is true, it's still compatible with what I said above. People still discuss moral facts in the same way they discuss other facts. Relativists just believe these facts are subjective.

    Even if this were not the case, however, what you'd be setting up here is the principle that we should, no matter the absence of mechanism, no matter the lack of parsimony, take as true whatever it is that most people believe to be true. "Most people believe there are objective moral facts, therefore we should assume there are objective moral facts".

    That would be a foolish principle! I am in agreement with you that "Most people believe there are objective moral facts, therefore we should assume there are objective moral facts" is a bad argument. However, I do think that people are warranted in believing what seems to be true to them, until it is defeated by a stronger reason to believe otherwise. Appealing to consensus is useful to gauge what people generally believe. It should not be used as a guiding principle for determining what is true.

    You have, however, in your assessment of those very moral facts just discarded the idea that what most people believe to be the truth about those moral facts is indeed the truth about those moral facts. This is a contradictory approach on the face of it.

    Have I? I don't think I've discarded what most people believe to be the truth about moral facts. If most people start with an intuition that "slavery is permissible," this intuition is what they have the most reason to assume until it is defeated by a reason to the contrary. This is not in contradiction with my assessment above, but in agreement with it.

    Likewise we could go the other way and ask whether most people believe prima facie that "what most people believe should be taken as being the case". I've no surveys to go on here, but I'd wager not many would agree with that, so even if it were true that "most people believe there are moral facts", adhering to our first principle would mean that we should not take this as reason to believe that there are moral facts.

    I agree with you here. What most people believe is usually pretheoretical, so it serves as a weak basis for establishing truth. Most people start from what they intuitively believe, then encounter arguments and theories that challenge their view. Once they begin this process, their intuitions will often become something they no longer have reason to believe. These prima facie beliefs will have been defeated. Until they are defeated, we have sufficient grounds to hold them.
  • Meta-ethics and philosophy of language


    Excited to get into epistemology and ontology threads when they come up. Are you a mod here or just a very involved user?

    The moral equivalent of that paradox would be the sentence "x is good but I don't intend x". It seems to me that what you intend and what you think are good are as inseparable as what you believe and what you think is true. It is possible to intend other than what is good, just like it's possible to believe other than that is true, but in saying something is good you implicitly express your intention that it be so, and so contradict the attendant explicit expression of intention otherwise.

    At face value, this seems patently false to me. Ex,
    "Going to the gym today is good but I don't intend to go to the gym today."
    Often, we recognize that things are good, but nonetheless intend otherwise. Conversely, it can't be said that we often recognize things to be true and yet believe otherwise. Alternatively, consider the guilty meat eater:
    "Eating meat is wrong but I still intend to eat meat."
    He may have come across a moral argument that convinced him that being a vegetarian is the morally right thing to do. He may still be a slave to his vices, or be insufficiently motivated by moral reasons. Nothing seems overtly paradoxical about this.


    This is the issue that I've been having trouble communicating to Kenosha and Isaac. I don't hold these "oughts" to be grounded in any "is". We can give a descriptive account of the cause of someone feeling or thinking that something ought to be, just like we can give a descriptive account of the cause of someone feeling or thinking that something is, but that account is not the reason why they feel or think that. We don't have to know anything at all about brains to go to our appetitive experiences as the ground of our "oughts", any more than we need to know about brains to go to our sensory experiences as the ground of our "is"s.

    In the latter case it's rather transparently the other way around: we learn descriptive facts about brains empirically, by relying on our sensations, our "is" experiences. If we then used our knowledge about brains, gained through sensory observation, to justify using sensory observation, that would clearly be circular reasoning. The reliance on sensory experience comes prior to any description of the brain. Likewise, on my account the reliance on appetitive experience as the ground for prescription comes prior to any description of the brain.

    What I am concerned with is not the reason why a subject thinks something ought to be the case. I am concerned with what renders them correct, independently of whether they know they're correct or not. It is not an issue of justification. It is an issue of truthmakers.

    "What would make 'murder is wrong' true?" Is a different question from "How can we know that 'murder is wrong' is true?" I am asking the former. Of course, you would say that these claims cannot be true at all, they can merely be correct. I don't think that your distinction between "correctness" and "truth" is particularly convincing, which I will address later in this post. Just substitute "true" for "correct," or "truthmakers" for "correctness-makers," wherever it is pertinent.

    Let me throw out an analogy that you may or may not consider relevant to this part of our discussion. Consider consensus relativism: the idea that what is moral is what the majority of people think is moral. I would claim that this theory jumps the is-ought gap. It claims that because a certain physical fact holds(people think X is moral), we derive from that what we ought to do(people think X is moral, therefore X is what ought to be done). You might reply that just because I can give a causal/descriptive explanation for people's agreement, that says nothing about the reasons why they come to their consensus about what is moral. I would find this to be an unconvincing rebuttal: if it is the fact that they agree that renders certain acts morally wrong, this fact is a physical fact. If we conclude something about what we ought to do from this physical fact, we are crossing the is-ought gap. The issue arises because, in consensus relativism, the truthmaker of moral claims is consensus. Consensus among humans is a natural fact. Likewise, if it is some natural fact of humans(they have certain appetites) that makes X moral, we are getting an ought from an is. Now, unlike a lot of people who misunderstand the significance of the is-ought gap, I don't think Hume had the final word on morality. But your theory in particular seems built on strictly partitioning "is" from "ought," so I think the distinction being crystal clear and incontrovertible is especially important for you.

    I'm not sure this is an accurate account of my account, because there could be differences between people that would make them have different appetites in the same situation. A moral claim is objectively true if it accounts for all of those different appetites in all different circumstances.

    So, is everybody's appetite given equal weight? Bloodlust or sadism can be considered appetites. Some people have these appetites. How does your metaethical theory account for these? Are they not ruled out for being in the minority? If half of all people had bloodlust as a base appetite, how would this change ethics?

    It's precisely that "idealized" that makes the difference. We don't, and can't, have a complete account of the way that the entire (possibly infinite) world would be experienced by all (possibly infinitely many) kinds of being. We only have the way that bits and pieces of it are experienced by beings like us. So we can't just take "the whole world, independent of all experience" (that non-existent noumena) and hold it up against "our picture of what the world should be like". We can only compare "the way this bit of the world seems to us like it is" and "the way this bit of the world seems to us like it ought to be".

    This is well put, and generally true. As a principle of epistemology, specifically constrained to physical matters, verificationism is about as good as it gets. What I don't see as plausible is the jump from "we can't possibly know everything" to "there is nothing outside what we can know." In fact, they seem to be borderline contradictory. If there is nothing outside what we can know, what are we failing to know when we can't possibly know everything?

    That last bit. The parable of the blind men and the elephant is the illustration I like to use here. Three blind men each feel different parts of an elephant (the trunk, a leg, the tail), and each concludes that he is feeling something different (a snake, a tree, a rope). All three of them are wrong about what they perceive, but the truth of the matter, that they are feeling parts of an elephant, is consistent with what all three of them sense, even though the perceptions they draw from those sensations are mutually contradictory.

    This is great. I like this parable a lot, and wish I had thought of a comparable example myself. Let's imagine that, for whatever reason, communication between these three blind men is impossible. Clearly, they are all restricted in the ways in which they can examine the elephant. If the only consistent consensus each can form is with himself, why is his judgement about the elephant not accurate? Snake-man can only verify a snake, tree-man can only verify a tree, and rope-man can only verify a rope. If verificationism about truth is correct, none of these men are wrong. Verificationism would assert that, in such an allegorical case, there would be no underlying fact of "elephant." If they cannot confirm an elephant, there is no elephant. Does this not seem as intuitively false to you as it does to me?

    But if not, then you wouldn't want to translate that second premise as "x not stealing implies x not cheating" (gerund, equally applicable to descriptions or prescriptions, just stating a relationship between those states of affairs), but rather as "'x ought to not steal' implies 'x ought to not cheat'". Symbolically, that would then be:

    P1. be(A)
    P2. be(A) -> be(B)
    C. be(B)

    This is the best interpretation of what I meant. This response does the best job at preserving formal validity. I want to preface wherever this part of the discussion goes by saying, in trying to consider an alternative account of logic itself invented by someone far more educated than me, I am getting in quite over my head. I have 5+ tabs of SEP open trying to figure out what is going on here. Miscommunication is very likely.

    On my account, it's not only moral utterances but ordinary descriptive utterances that are impressing an attitude toward the idea in question.

    It is important to remember that propositions need not deal only in utterances. We could, again, imagine that bugs have died in the shape of words on some piece of paper to form a valid modus ponens. Impression would play no role in our evaluation of it.

    My systemic objection here would be that classical logic works very well for us. An alternative account being merely internally consistent(if yours in fact is) does not give us sufficient reason to switch, universally, our understanding of classical logic to Pfhorrest logic. We would need a compelling case to conclude that when your position produces errors in classical logic, we ought to discard classical logic rather than your position.

    Attempts to eliminate the importance of tense from logic have, I believe, been made before. They are highly contentious. The work of R.M. Hare comes to mind. "Semantic content" and "meaning" are closely connected concepts. Sure, "Close the door!" and "The door ought to be closed." and "The door is closed." and "You should believe that the door is closed." all can be thought of as different forms of the same primitive idea, "the door being closed." Can we conclude from this that the meaning of all of the above is identical? Can they be freely interchanged with each other, and all be represented by the same variable in a deduction? Formal meaning must be quite distinct from semantic content if we can conclude this. "All doors are closed" and "some doors are closed" are both merely quantified versions of "the door being closed," yet, differentiation between them in formal logic is incredibly important. In avoiding the Frege-Geach problem, you might leave yourself with an account of logic that is too fuzzy to be as useful as classical logic.

    This is also where I will further explore your distinction between "truth" and "correctness." Typically, logic deals in propositions that can be true or false. Validity is a property of arguments that lies in the relationship between premises and conclusions, where an argument is valid only if the truth of the premises entail the truth of the conclusion. This relationship of entailment is core to logic. You consider the domain of what can be true to solely contain the domain of what is physical. Other domains that we might normally consider to be capable of bearing truth, such as matters of mathematics, logic itself, morality, and theology, you consider merely "correctness-apt."

    You do not seem to object to logical arguments being built around "correctness" in the same way we might normally use "truth." That you would consider a moral modus ponens to be capable of validity at all requires "correctness" to have the same relationship with logical entailment that traditional "truth" does. It seems that the concept of truth, as typically employed, you have replaced wholesale with this notion of "correctness." Under your account, "truth" has therefore been reduced in scope to refer merely to "physical truth." It appears, then, that rather than identifying a supercategory above and including our typical term "truth," you have introduced a subcategory: "physical truth." This is why I would suggest abandoning the "truth/correctness" dichotomy entirely, and just referring to "truth" and "specifically physical truth." If correctness does everything that truth does in logic, it just is truth. We can cast aside "all truths are physical truths, all else is correctness" in favor of "some truths are purely physical truths, and these are relevantly different from nonphysical truths." Since you don't espouse a correspondence theory of truth, you don't have to account for anything that nonphysical truths correspond to. This change would not be problematic for your theory at all.

    I didn't realize I was using a customized version of the word; I honestly just thought "supererogatory" was the deontic equivalent of "contingent". Doing some further research now prompted by this, I see that the word I really want is "omissible". I'll make sure to use that instead from now on, and change where I've mis-used "supererogatory" in the past where possible.

    No worries, glad I could help clear that up. "Omissible" works much better, though I might even simply suggest "Nonobligatory." It stands in clear contrast with "obligatory," and I know you love your symmetries.

    Quantum physics is full of different interpretations, so I'm cautious to speak authoritatively about what all physicists think, but as I understand it, the uncertainty principle doesn't just say that we can't know position and speed at the same time, but that to the extent that we measure one, the other becomes literally undefined. A particle with a definite momentum has no definite position; its position is actually smeared out across space.

    You make a good point. We ought to be cautious in speaking about quantum physics that we do not tread into "quantum woo" territory, given that neither of us know much about it. I was under the impression that the Heisenberg uncertainty principle established epistemic uncertainty, but if I am incorrect, I have nothing further to add to this line of argument. It may be legitimately indefinite - we might lack full information because full information does not exist to be had at at the moment of measurement.

    To other observations. We're (rightly) cautious about the accuracy of our present beliefs because we haven't made all of the observations (and never can), and there may be ones that contradict what we presently believe. To assert that something is objectively true is to assert that there won't ever be any contradictory observations. We can never know that with certainty, of course, but we can think it is so, at least tentatively.

    I feel like that interpretation of the word "objective" kinda came out of the blue. I don't think that to say something is objective means it is incontrovertible. I would rather say, of a truth that no knowledge could ever be obtained to contradict it, that such a truth is "indefeasible." When it comes to "objective," it is most often used in contrast with "subjective." I defer to Derek Parfit's definition here, which I like a lot and I think really captures what we mean when we talk about things being objective:

    "According to subjective theories, we have most reason to do whatever would best fulfill or achieve our present desires or aims. Some Subjectivists appeal to our actual present desires or aims; others appeal to the desires or aims that we would now have, or to the choices that we would now make, if we had carefully considered the relevant facts. Since these are all facts about us, we can call such reasons subject-given. According to objective theories, we have reason to act in some way only when, and because, what we are doing or trying to achieve is in some way good, or worth achieving. Since these facts are about the objects of these desires or aims, we can call such reasons object-given."
    -Parfit, On What Matters vol. one

    We could claim something to be objectively true(true independently of our opinion on the matter), and yet maintain that this claim is defeasible in the face of potentially undermining evidence. I do agree with you that we should never proclaim truths to be indefeasible, as this can be dangerously dogmatic. Kant argued something similar in his essay "An Answer to the Question: 'What is Enlightenment?'" which I would highly recommend.

    Causality, like physicality, is part of the background assumption of objectivity that we have to make in order to go about the process of investigating what is real. We can't empirically prove that anything is objectively real, either, but the question of whether anything is objectively real is prior to the empirical investigation. Likewise, the question of whether anything causes anything else. Empirical investigation helps us sort out what kinds of things cause what other kinds of things, on the assumption that things cause other things in the first place and all our experiences aren't willy-nilly incomprehensible.

    Yes, we assume these things to be true. That is what an assumption is: taking something to be true. That is an example of something we take to be true, and which might actually be true(things might actually cause other things, or they might merely correlate consistently), but cannot be empirically verified.

    The idea that there are only empirical truths or definitional truths has its roots in "Hume's Fork," which asserted that truth neatly divides into the two camps of "matters of fact" and "relations of ideas." If you prefer, we could accurately describe the two camps as "synthetic a posteriori" versus "analytic a priori." I take it you believe something like this, since you said at the start of our discussion that mathematical truths are merely definitionally true relations of ideas.
    Hume's Fork has since been refuted, thanks in large part to the work of Kripke, who introduced the necessary(re: analytic) a posteriori truth, and Quine, who put the analytic-synthetic distinction itself on dubious ground.

    According to current physics, that information does still exist in the universe, and so in principle those things are empirically verifiable, it's just ridiculously impractical to go about doing that verification.

    As long as information exists in the universe, it is empirically verifiable in principle? What about information that is moving away from us faster than the speed of light, at the edge of the observable universe? It is physical, yet empirically unobservable to us. Same with matters of the past - we are dealing with imperceptible matters of fact. If I say "Charlemagne had an odd number of hairs on his head," I might be correct on a guess. What would render me correct? If Charlemagne did, in fact, have an odd number of hairs on his head. Surely you wouldn't believe it impossible that Charlemagne might have had an odd number of hairs on his head? If you were to claim that it would be impossible for me to sufficiently justify a belief that he did, I would agree. However, holding these things to be undefined seems a harder bullet to bite than just admitting we can't know everything. Mystery is an aspect of the human condition, and we ought to become accustomed to it.

    This is why I only embrace verificationism explicitly within the narrow domain of descriptive truths. The principle of verificationism itself is not a description of the way the world is, but is something we settle on prior to even engaging in description.

    If everybody had to settle on verificationism about truth before even engaging in description, nobody engaging in description would disagree with verificationism. People do disagree with verificationism(in fact, it's by far the minority position), ergo, we do not need to settle on verificationism before engaging in description. If you construe "description" to mean "describing physical things," it is trivially true that empirical methods are the best methods to engage with empirical subject matter. But, not everyone makes this conflation. Most people think things can be true yet not physical. Only the most extreme skeptics doubt causality, yet even people working in the sciences doubt that verificationism about truth is legitimate.
    We have plenty of reason to believe things that cannot be empirically verified: truths of math, logic, reason, and morality especially.

    The information about the cheating still exists out there in the world threatening the harm of its knowledge to the man. If it were possible (which it's not) to change the world such that all information about the cheating was completely eradicated, then that would be equivalent to changing the world to be one in which the cheating had never happened.

    So, if we stipulate that nobody besides me and the other woman would ever find out about it, cheating on my partner would be morally permissible?

    Imagine two worlds: In W1, the holocaust never happens. In W2, the holocaust did happen, but it happened 20,000 years ago and no traces of it remain to be observed by contemporary people. Do we have reason to call W1 better than W2? Should we prefer to exist in W1? I say yes. The suffering of people still matters, whether or not we can empirically verify that they suffered.

    If it were the case that such a person could never encounter anything about the world that would be counter to this belief, then that belief would either actually be true, or just be empty.

    The existence of god is unfalsifiable, therefore god exists. This is a bad argument, but from the perspective of the man whose belief is never falsified, he may not be mistaken to make it. It is conceivable that there could be some fact which, if he learned it, he would cease to believe in god. It is equally conceivable that he could never encounter this fact. Some theses are just unfalsifiable: solipsism, simulation theory, god, Descartes' deceiving demon, and causal determinism among them. I would agree that believing in these things is rarely useful, and in that sense "empty," but any one of them may nonetheless accurately describe the condition of the universe. It may be the case that you are the only entity in existence. To argue that it could not be the case just because it could not be verified is unconvincing.

    If you mean abstract moral objects specifically, then what explanatory power do they have?

    The explanatory power lies in the observation that we discuss moral facts the same way we discuss other facts. We debate, disagree, and use reason to draw conclusions from premises. When we disagree, we believe that one or both of us are incorrect. As you are discovering, reforming our language and logic to compartmentalize moral facts is a Herculean task. Some might call it unpragmatic. Postulating that when we say "slavery is wrong," we descriptively assign a property "wrongness" to an act "slavery," and do so either correctly or incorrectly, is easy. It reflects what we appear to be doing with moral language, prima facie. The only substantive objection you've offered to seeing things this way is the argument from queerness, which just asserts that moral facts seem too weird to exist. That's not something that would deter a pragmatist.
  • Meta-ethics and philosophy of language


    Lockdown, eh? If you're in America right now, my heart goes out to you.

    Nope! Two different orthogonal divides, the distinction between which is critical to my account. Direction of fit is about the kind of opinion (descriptive or prescriptive), and impression vs expression is about what you’re doing with that opinion relative to someone else (just showing them what opinion is in your mind, or trying to change the opinion in their mind). You can impress or express either kind of opinion, descriptive or prescriptive. If anything, expression is more description-like because it merely shows what your opinions are (so is like describing your opinions), while impression is more prescription-like because it tells someone to have certain opinions (so is like prescribing your opinions). But they are still orthogonal: you can express your beliefs, impress your beliefs, express your intentions, or impress your intentions.

    I was making the mistake of necessarily associating impressions with prescriptions and expressions with descriptions. Thank you for your helpful clarification. Description(belief-level) and prescription(intention-level) are what I really care about here. Impression and expression seem unimportant. Most utterances have both an expressive and impressive aspect, do they not? If I say, "There is a truck right there!" I am both revealing my belief that there is a truck, and imploring my listener to share my opinion. Not only this, but a proposition can have neither aspect. Imagine that we find a piece of paper with "the Earth revolves around the sun" written on it. We could imagine, however unlikely it may be, that this writing appeared by chance: bugs crawled onto the paper and died in just such a pattern, or something equally implausible yet possible. There is no expression or impression involved. We can still evaluate this proposition as true or false. We could do the same with a moral sentence. What significance do impression and expression have?


    I must admit I have noted this apparent asymmetry before and struggled to reckon with it. It makes me feel like there is something I haven't fully developed right. When it comes to my approaches to assessing the correctness of either beliefs or intentions, I do end up with a nice symmetry again, but it feels like some bridge between the symmetry of meanings and the symmetry of assessment is missing, for the reasons you state. So I'm glad we're talking about it, because this is the kind of situation where I usually come up with newer, better thoughts.

    I agree with this assessment. Also, I'm glad that I'm not just running you through the same argument-counterargument loops you've probably dealt with a thousand times on this forum! I definitely have more challenges to your line of thinking in store. You're certainly challenging mine - I don't usually engage in this type of written exchange at all, and of course, the contents of your theories are almost entirely new to me.

    The symmetry I end up with for assessing the correctness of either kind of opinion is checking the opinion against experiences, where experiences come in different varieties that carry their own direction of fit: experiences with mind-to-world fit are sensations (like sight and sound), and experiences with world-to-mind fit are appetites (like pain and hunger). In both cases, assessment of the objective correctness of an opinion needs to account for not just the experiences you are actually having right here and now, but all the experiences anyone could have in any context.

    I went back in the thread to where you explained your views on feelings, and gave that section a few close readings. The trouble in walking this tightrope of symmetry is to ensure that one side of your dichotomy does not simply reduce to the other. Let us compare appetites and sensations. A complete physical description of my brain would include both my sensations and my appetites. I realize that hunger is not the only appetite, but it is the one I will use as an example here. My most basic "ought"(I ought to be fed) is grounded by an "is"(I have the brain state "hunger"). What is this relationship? Is it supervenience? Is it necessity? Is it equivalence?

    Let me be more precise: it is not that I have the brain state "hunger" that matters to us. I infer from what you've said above("all the experiences anyone could have in any context") that what matters is not my appetite, but everyone's idealized appetite were they in my situation. The fact remains that the resulting "ought" is reducible to a collection of appetites, which are themselves describable in terms of physical brain states. Depending on the relationship between the "ought" and the brain states, the below might be a valid deduction:

    P1. All beings like us would have the brain state "hunger" in situation Y.
    P2. I am a "being like us" in situation Y.
    C. I ought to be fed.

    Here, we have jumped the is-ought gap. Ordinarily, we would require a premise like:

    P3. Beings like us with the brain state "hunger" in situation Y ought to be fed.

    However, if P3 is reducible to P1, we do not need P3. The "ought" would be intrinsic to the "is."

    I think perhaps the missing bridge that avoids the asymmetry you note -- and this is just me thinking on the fly here, not recounting thoughts I've already had before, so thank you again for prompting some new thought -- is that direction of fit needs to be reckoned not so much as a relationship between the mind and the world, as it usually is, but rather as a relationship of these different descriptive and prescriptive models to our overall function from our experiences to our behavior. We don't have direct access to the world, all we have is the experience of our interactions with the world, what it does to us and what we do to it.

    According to verificationism, unknowable noumena(things-in-themselves) do not exist. Given this, it is unclear to me why you feel the need to distinguish between "the experience of our interactions with the world, what it does to us and what we do to it" and "direct access to the world." Verificationism would hold that an idealized version of the former is identical to the latter.

    Being interactions between ourselves and the world, our experiences of either kind are about both ourselves and the world: sensations tell us about how things look to beings like us in certain circumstances, appetites tell us about how things feel to beings like us in certain circumstances. The direction of fit is more between those self-regarding and world-regarding aspects of the experience, internal to the experience, than between the mind and the world itself.

    Now, let's return to sensations and perceptions. The thing of most note here is the concept of "beings like us." Let us apply your truthmaking method to its intended context: differences between the perceptions of humans. First comes the brute observation of the individual. This takes the form of me seeing a ghost. Second, we compare this to our ideal aggregate's experience. Beings like me, in the situation where I saw a ghost, would not see a ghost. This is a discrepancy between how beings like me(with proper function) perceive the world, and how I perceive it. Therefore, "I saw a ghost" is rendered false.

    Where this gets interesting is with the introduction of beings entirely unlike us. We can imagine aliens who perceive the world drastically different than we do. They could experience appetites and sensations wholly separate from the ones humans experience. Does this entail relativism about descriptive truth? Such an alien must compare their experience with beings like themselves, not beings like us. Their conjunct of all "true"(re: empirically verifiable to beings like them) propositions might overlap with ours, but there would be things we can verify that they cannot, and things they can verify which we cannot. Can a proposition like "objects take the form of shapes" be true-for-us but false-for-them, or even more confusingly, true-for-us but nonsense-for-them? Are neither of us lacking in our description of reality, since reality just is our relative ideal description? Or, are we both aiming at an underlying truth and coming away with only part of it?

    This doesn't seem to be a logically necessary premise in the same was as "if stealing is wrong, getting your little brother to steal is wrong". So it makes sense that you wouldn't reconstruct that in the same way as the brother implication, as "(stealing) implies (cheating)", because that's incorrect; there could be stealing and not cheating.

    Of course there can be stealing without cheating! My premise P2 is false, but this is inconsequential in a discussion of logical validity. An argument with false premises can still be valid, and a theory of logic has to account for why it is valid. I think there is some confusion over my point here, so I will elaborate on it further.

    Consider a symbolic modus ponens:

    P1. A
    P2. A > B
    C. B

    So long as A and B are standing in for propositions, this argument is valid. Now consider my original moral modus ponens:

    P1. Stealing is wrong.
    P2. If stealing is wrong, then getting your little brother to steal is wrong.
    C. Getting your little brother to steal is wrong.

    This argument seems valid in the usual way. My second example(the one with cheating) was to demonstrate that your alternative account of the above's validity(it's valid because little brother stealing entails stealing) does not actually explain why the argument is formally valid. What I showed was that even if "B" does not entail "A," the argument is still valid, and trying to apply your explanation of "B entails A"(in that case, cheating entails stealing) fails.

    There are several approaches you might take to explain moral modus ponens. First, you could accept that these propositions are truth-apt in the regular way. Then, moral modus ponens is the same as any common variety modus ponens. Second, you might hold that moral utterances are not truth-apt propositions at all. They might serve a purpose that is not purporting to report fact, such as expressing an attitude. This is the view of moral emotivists. The moral emotivist encounters the following serious problem:

    P1. Stealing is wrong(expresses Boo, stealing!)
    P2. If (Boo, stealing!), then (Boo, getting your little brother to steal!)
    C. (Boo, getting your little brother to steal!)

    The first issue here is that the atomic sentence in P1(can be represented as "A") is used to express a sentiment, while the same exact sentence does not do this in P2. By saying "If stealing is wrong, then getting your little brother to steal is wrong" I am not expressing an attitude towards the antecedent. Therefore, "A" as an atomic sentence in P1 and "A" as the antecedent of a conditional in P2 have different meanings. In P1, "stealing is wrong" expresses an attitude towards stealing, but in P2, it does not. This breaks the law of identity(A=A) which is obviously necessary for modus ponens to be valid. This is called the Frege-Geach problem, and while you are not an emotivist, I believe that your theory needs to be wary of it as well.

    The second issue is that, even if the first issue is ignored, it is not clear that the relationship between the the premises and the conclusion is one of validity. If I accept P1 and P2, then reject C, am I failing logically? It seems that I am merely violating some kind of moral rule(Don't do what you boo!) concerning my attitudes, rather than a logical one. I may hold inconsistent attitudes, but am I making a logical mistake?

    My original intention in introducing this line of analysis was to apply the first issue of the Frege-Geach problem to your theory. If you are claiming that moral propositions serve not to report fact, but to perform an act of prescription, you are vulnerable to this issue. While in an atomic moral sentence(Stealing is wrong!) I may be prescribing something, in the antecedent of a conditional(IF stealing is wrong, then..) I am not prescribing anything. My utterance of this sentence does not commit me to an attitude on stealing. This puts the law of identity on fragile ground, as explained above.

    You might say that using moral sentences in regular formal logic is some kind of mistake, since they are "correctness-apt" rather than truth-apt, and we should be using a system that you have constructed as part of your theory to explain them. Not only does this require significant justificatory legwork, it is unclear why we should be compelled by the "schmogic" of moral propositions, or hold such "schmogic" in equal standing to regular logic.


    For the same reason that "false" is contained within "contingent": supererogatory = not-obligatory, and all bad things are not-obligatory, just like contingent = not-necessary, and all false things are not-necessary. (There are some things that are necessarily false, but that just means impossible; likewise, things that are "obligatorily bad", so bad you are obliged not to do them, are just impermissible).

    If you are using supererogatory to merely mean "not obligatory," this is idiosyncratic. It means "above and beyond the call of duty" or "to a level far exceeding what is obligatory." I'm not one to argue about the definitions of words, as they mean what we want them to mean, but if you are going to use a customized version of a word, it is usually best to clarify before employing it.


    I disagree. When it comes to the limited domain of descriptive propositions, I agree completely with the verificationist theory of truth: a claim that something is true of the world yet has absolutely no empirical import is literally meaningless nonsense. If something like gods can really be said to exist, there must be something observable about them.

    Here is my challenge to the verificationist theory of truth. You hold that for something to be true, it must be empirically observable. I take you to mean observable in principle, as this is the strongest form of the claim. From our discussion it is also clear that you believe the "empirically observable" and the "physical" to be one and the same category. For something to be empirically observable, it must be physical, as our empirical methods interact only with the physical world. If something is physical, it exists, and if something exists then true claims can be made about it. For true claims to be made about something, it must be empirically verifiable, as you have stated. Therefore, if and only if something is physical, it is empirically verifiable. If I can provide an example of something that is physical yet not empirically verifiable(even in principle), verificationism becomes quite untenable. Such a damning counterexample seems to exist.

    Consider the Heisenberg uncertainty principle. We are confident that we can know either the precise position or the precise speed of a particle at a given time, but not both. A particle is a purely physical entity, and facts about particles are physical facts. I can ask this question:

    "Knowing this particle's current position, what is its current speed?"

    The answer to this question is empirically unverifiable, in principle. By measuring the particle's position, we necessarily prevent ourselves from verifying its speed. Must we conclude that the question is nonsense? This seems especially unintuitive when we consider that we could have chosen to measure the particle's speed instead. If we had done so, its position would be unknown to us. Can a question about a physical matter of fact become nonsense moments later? If we are verificationists, we must concede that at the time of measuring the particle's position, to ask of its "speed" is nonsensical, referring to a property that does not exist. This is certainly not what physicists conclude here. Rather, the Heisenberg uncertainty principle establishes that there are physical matters of fact which are empirically unverifiable.
    If verificationism is true, we have complete information about the particle each time we measure it. What would the Heisenberg uncertainty principle be stating if there was nothing to be uncertain about? I think that this alone is enough to do away with verificationism, but I will provide further reasons which I consider to be overdetermining.

    Defining verificationism without reference to a separate truth seems problematic. Why do we not appeal to the consensus of what we have, in fact, empirically verified? Well, because we know our methods are flawed. We must instead appeal to the consensus of our "idealized" methods. An idealized method is one where any mistakes we make are eliminated. A mistake is an inaccuracy. An inaccuracy to what? To the truth? How do you define our ideal methods of observation other than "the ones which accurately capture the truth?"

    Causality cannot be proven true empirically. Are questions about causality nonsense? Our investigations into the material world are predicated on all sorts of assumptions that are, themselves, empirically unverifiable. I am hard pressed to imagine us building a search for truth atop a foundation of nonsense.

    The past is empirically unverifiable. Whether or not a certain dinosaur ate in a certain spot sixty million years ago is a matter of fact that we are incapable of observing. The same can be said of whether Charlemagne had an even or odd number of hairs on his head, or whether a man in New Zealand stubbed his toe last week. Should we conclude that questions concerning these subjects are literally not truth-apt?

    Some people consider verificationism self-defeating. The claim "empirical verification is the only way to learn truths about the world" is, itself, empirically unverifiable. That means, by your account, that someone asking "Is empirical verification the only way to learn true things?" is asking an unintelligible question. The statement itself cannot be true or false.

    Moving on,

    The prescriptivity involved there is still ultimately the same kind as moral prescriptivity, though. It's basically a case of considering what the proper function of a human mind is -- proper as in good, good as in prudential good, which we've already established boils down to moral good -- and then looking upon yourself in the third person, so to speak, and thinking "Hey, there's a mind! Is it functioning properly? No no, it should be perceiving like this and desiring like that instead..." It's self-parenting. Parents teach their kids how to think, both in terms of figuring out what is true and in terms of figuring out what is good, for the moral good of those kids, and everyone they'll have an impact on, right? Likewise, making sure we ourselves are thinking correctly is ultimately for a moral good too.

    How can you establish that holding true beliefs is always morally better for us? I can't imagine such a stance being grounded in the "appetites," which you tout as the foundation for moral imperatives. A man being cheated on by his wife experiences no harm, so long as he remains unaware. Discovering the affair is what causes him hedonic harm. Here, one benefits from holding a false belief. Yet, we still believe it is morally wrong to cheat when your partner doesn't know about it.

    Consider believing in a benevolent, all-knowing god. It is plausible that I could live a life where I purely benefit from having this belief. When my life goes poorly, I can keep my chin up, believing that it's all a necessary part of god's great design. When my life goes well, I can attribute my good fortune to god looking after me, rewarding me for my virtue and compensating me for my hardships. This belief may be prudential(thus moral) for me to have, and for others to encourage me to have. Concluding otherwise requires an account of why truth is valuable separate from moral good.


    If you are concerned with pragmatism rather than existential truth, it would seem most pragmatic to simply postulate abstract objects exist and move on from there. From the pragmatist perspective, who cares if they are real? They are useful. Their explanatory power, and the lack thereof of the alternatives, speaks for itself.
  • Meta-ethics and philosophy of language


    As you are evidently a meticulous thinker, I will have to try my best to act as one in turn. The level of detail in your posts will delay my responses.

    I say instead that moral utterances impress (and so implicitly also express; you caught the part about impression vs expression earlier?) intentions. And I say that intentions can be objectively correct or incorrect ("true" and "false" also frequently have descriptivist connotations, so I try to avoid them myself, but recognize their casual use). Both intentions and beliefs are subsets of what I call "thoughts" (as distinct from "feelings", "experiences", and other mental states), so the simplest rephrasing of the above would just be to say "John thinks ... while I think ..." instead, since the permissible/impermissible already carry subtler imperative force.

    I did catch what you wrote about impression vs expression. If I understand it, it follows this neatly symmetric(quite pleasing if true!) world-to-mind vs mind-to-world dichotomy of thought that you are positing. By your account, an "impression" is a speech-act that is world-to-mind, and an "expression" is a speech-act that is mind-to-world. It is when we begin to explore the "correct/incorrect" and "true/false" distinction that I lose your train of thought. I'd like to refer back to something else you said about the importance of direction of fit: "it may be the same picture, but its intended purpose changes the criteria by which we judge it, and whether we judge the picture, or the thing it is a picture of, to be in error, should they not match."

    I'd also like to borrow a quote from Wittgenstein: "The world is all that is the case." What I take it to mean is that, essentially, if one knew the conjunct of all true propositions, they would be lacking nothing in their account of reality. What it means for something to be true is that it is the case, and what it means for something to be false is that it is not the case. As you correspond "true/false" to descriptivism, I imagine you're inclined to agree. I expect that you might respond with a mirror image of this sentiment, perhaps something like "If one knew the conjunct of all correct intentions, they would be lacking nothing in their account of what-ought-to-be. What it means for something to be "correct" is that it ought to be the case, and what it means for something to be incorrect is that it ought not to be the case."

    With that in mind, let's hop back to your quote. We have a picture. We wish to judge whether this picture is in error. If the picture is mind-to-world, it is of the usual descriptive sort. It is attempting to depict something factual. Note that, as described, this is strictly a two-way interaction between mind and world. In judging the picture as erroneous, we note that the idea is divergent from the content of the world. But what if the picture is world-to-mind? Now, it is of a prescriptive sort. We would expect that, given the symmetry yet distinction between these two types of claims, we would proceed through inverting the judgement used in the descriptive case. We would judge the world as erroneous for diverging from the content of the picture. I believe you say as much when you establish that the descriptive-prescriptive divide dictates "whether we judge the picture, or the thing it is a picture of, to be in error, should they not match." Again, this is a strictly two-way interaction between mind and world. What if we apply this to moral judgement?

    Instead of a picture of what ought to be, we are now dealing with a claim of it: "it ought to be the case that Russia launches nukes" This is a world-to-mind judgement. Therefore, if the world does not match the mind(I imagine that in this case, that would look like "Russia is not in fact launching nukes"), we judge the world to be in error. This, on its own, entails relativist conclusions. This presents a problem. I expect that you would respond to this problem by claiming that "it ought to be the case that Russia launches nukes" is an intention uttered in error, since it ought not to be the case that Russia launches nukes. At this point, mind-to-world can no longer be considered symmetrical to world-to-mind. Instead of the two-way relation between world and mind that we explicitly find in the mind-to-world case, we are forced to introduce some further arbitration to account for the fact that even in a world-to-mind judgement, the mind can be mistaken in some further way besides simply not matching the world. As opposed to mind-world, we have something like world-mind-standard. You might accept this asymmetry, but it harms the parsimony of your theory to do so. Of course, I would avoid this issue by wholeheartedly accepting that moral states of affairs exist, and utterances concerning them are beliefs which can be true or false in the regular way.


    So in your modus ponens, the logical relationship is actually between "stealing happening" and "getting your little brother to steal": getting your little brother to steal entails stealing happening. So if it-ought-to-be-the-case-that-there-is(not(stealing happening)), and (getting your little brother to steal) entails (stealing happening), then it-ought-to-be-the-case-that-there-is(not(getting your little brother to steal)).

    You can replace it-ought-to-be-the-case-that-there-is with it-is-the-case-that-that-there-is and you get the same logical relations, just with descriptive force instead of prescriptive force.

    Sure, you can explain that modus ponens that way. I actually found this response quite persuasive on the first reading, and thought I might have to abandon this line of analysis. On further reflection, I found your explanation lacking when applied to other moral modus ponens. First, let me structure your account of the logical relation in argumentative form:

    P1. It ought-to-be-the-case-that-there-is(not(stealing happening))
    P2. If (getting your little brother to steal) then (stealing happening)
    C. It ought-to-be-the-case-that-there-is(not(getting your little brother to steal))

    It certainly seems that if one were to accept P1 and P2, yet reject C, they would be succumbing to a logical failing. This accounts for the logical force of modus ponens. It is also intuitive that yes, if I get my little brother to steal, stealing is happening! Your alternative P2 seems synonymous with my original P2. In that sense, it can be argued to be semantically identical to the original syllogism. That is why I considered your response sufficient to resolve that particular modus ponens. However:

    P1. Stealing is wrong
    P2. If stealing is wrong, then cheating on a significant other is wrong
    C. Cheating on a significant other is wrong

    Let's translate into your proposed logical language.

    P1*. It ought-to-be-the-case-that-there-is(not(cheating happening))
    P2*. If (cheating happening) then (stealing happening)
    C*. It ought-to-be-the-case-that-there-is(not(stealing happening))

    Something isn't right here. P2 is translated incorrectly. The meaning of "If stealing is wrong, then cheating on a significant other is wrong" is totally different than the meaning of "If (cheating happening) then (stealing happening)." The relation that you originally proposed(getting my little brother to steal entails stealing happening) does not accurately explain WHY moral modus ponens is valid. Let's translate P2 more faithfully.

    P2**. If it ought-to-be-the-case-that-there-is(not(cheating happening)), then (it ought-to-be-the-case-that-there-is(not(stealing happening))

    After this translation, I'm unsure why you'd have to invent a different logic for dealing with prescriptive claims at all. "Stealing is wrong" and "It ought not to be the case that there is stealing happening" are identical claims by your view anyhow. In response to the bolded text, I concur that we can treat "is" and "ought" interchangeably in syllogisms as far as logical validity is concerned. Where we diverge, however, is that you seem to think that this is in every case meaningful. I believe that many, if not most atomic sentences produced by swapping "is" with "ought"("the sky ought to be blue," "the earth ought to be round," "a certain star a million light years away ought to be mostly hydrogen" "a triangle ought to have three sides") are nonsense in normative terms.

    I see prudential oughts as boiling down to a kind of moral ought. Taking care of yourself is a kind of moral good -- not necessarily an obligatory one, but still a moral one even if only supererogatory, you matter just like everybody else matters -- and instrumentally seeing to moral ends is still a kind of moral good. So you should stop smoking because if you don't you'll probably suffer and die, and people suffering and dying is bad.

    This satisfies me. Well put. If there are in fact any prudential reasons that do not collapse to moral reasons, I can't think of one right now.

    Rational "oughts" I think can be better rephrased descriptively. "If you proportion your belief to the evidence your belief is more likely to be accurate." You might ask "but should beliefs be accurate?" and the answer to that is a trivial yes, because believing something just is thinking it's an accurate description of reality. If you didn't care to have an accurate description of reality, you wouldn't bother forming beliefs.

    This, I find much more problematic. You could rephrase moral oughts descriptively in the same way: "If you don't lie to people, you are more likely to be morally correct." The fact remains that rational normativity is used extensively in language, and not just by term-confused laypeople. We tend to believe that people ought to be rational - that they ought to aim to believe only true things, they ought not to hold obviously contradictory beliefs, they ought to afford greater consideration to good reasons rather than bad ones. You might suggest that these collapse to moral oughts, holding that we only value rational norms because to forsake them would result in great unpleasantness for us, or something like that. I think this is defensible, but dubious. I suggest the opposite - that what is moral is merely a subset of what is rational. When I consider your arguments for your position, I am considering reasons. I might have good moral reason to agree with everything you say and shower you with compliments. This would certainly make you happy, if you believed it to be sincere. It may be supererogatory, but if all reasons are moral reasons, what does it really mean for something to be supererogatory?(I saw your chart, but that raised more questions than it answered. Why is "bad" contained within "supererogatory?") The thing that I have the most moral reason to do would always be the thing that I have the most overall reason to do, by definition. Why do people disagree at all in cases when agreement would result in a morally preferable state of affairs? Perhaps they are responding to reasons which are not moral in nature, and indeed may be stronger than whatever small moral reason they have to abandon their argument and agree.
    For instance, take this principle: "If another person is meaningfully disagreeing with you, and you know that they cannot be convinced of your position, you ought to concede." Morally, this is justifiable. Continued disagreement without even the possibility of agreement is pointless. It serves only to create strife and tension, which is not a preferable moral state of affairs. Would it not be advisable to change your beliefs and eliminate this tension? Surely you have no moral reason to choose a true belief over a false one, unless holding this false belief itself somehow harms people. This could even be avoided. I could become a "round-earth constructivist," believing something like "the earth is truly flat, but we must act as though it is round for the best results." I could adopt this belief because I have good moral reason to do so: dissolving tension with flat-earthers who cannot be convinced. Is a principle like this the one you have the most reason to follow? Is concession, changing your beliefs when you are certain your interlocutor will not change theirs, the thing you have the most reason to do?

    beliefs are to be judged by appeal to the senses, everyone's senses in all circumstances if they are to be judged objectively, and intentions are to be judged by appeal to the appetites, everyone's appetites in all circumstances if they are to be judged objectively.

    But yet, everybody's senses in all circumstances could still lead to something false. There might be things which are true of the world and yet empirically unverifiable, even in principle, like the existence of a god. As an epistemological verificationist principle, I suppose this is really as good as it gets. But as a principle for defining what is actually true/false, correct/incorrect, this lacks in the same way that a traditional "ideal rational observer" account does.

    They are instead pushing a considered thought about how everyone should think things ought to be, in the same way that descriptive statements are pushing a considered thought about how everyone should think things are.

    Woah, woah. Descriptive statements are pushing a considered thought about how everyone should think things are? That itself is prescriptive! I also think it's inaccurate. Consider a color judgement. I have a degree of red-green colorblindness. If I make a descriptive claim such as "this chair is brown," I would sure hope I'm not implying that everybody else ought to see the chair the way I do! I am merely reporting how I observe the chair to be.
  • What are people here's views on the self?

    All of these positions can be seen as reductionist, in that they treat personal identity as a product or manifestation of something more real or fundamental:

    Psychological continuity
    Physical (worldline) continuity
    Structural similarity
    Hidden essence (soul, etc.)
    I don't think you can accurately construe a soul-based theory as reductionism about the self. In what I've read, these types of theories don't say that the self is "reducible to" the soul, rather, it just IS the soul. This is less of a reduction and more of a plain definition. The self would not be a "product or manifestation of something more fundamental," as it would just BE the soul, and the soul would be brute and irreducible. I see an argument for construing those other positions as reductionist.

    I rather think that personal identity is a psychosocial construct. Consequently, it doesn't have a strict definition and delineation, but rather relies on intuitions and conventions that are to some degree fluid and diverse. This is why even those people who don't already have a favorite philosophical theory of self never seem to have a common opinion on such esoteric thought experiments as Davidson's Swampman, teleportation, duplication, etc.
    I agree with you to some extent. Our society places a lot of importance on personal identity, and this leads us to form the conceptions about it that we do. I believe many people hold false beliefs about the nature of themselves, due in large part to this sort of conditioning.

    And since there is neither a prevailing philosophy nor a prevailing intuition or convention that would apply to such cases, answers vary.
    Which is why it is so interesting to ask the questions!
  • Meta-ethics and philosophy of language
    First, I want to commend you for the substantial time and effort you have clearly put into philosophizing. You have managed to put something together which is, on the whole, innovative, and impressive. As to whether or not it does the best job among competing theories by metrics of truth and explanatory power, I hold my reservations. But, I expect a good time investigating it.

    I find this Humeanism vs Kantianism to be a false dichotomy, and moral expressivism vs moral realism to be a false dichotomy as well. I think, like the Humean, that there are no such things as "moral beliefs" per se, and that moral utterances do not have any meaning to be found in some description of reality; but I also think, like the Kantian, that moral utterances do much more than just express desires incapable of being correct or incorrect.

    "John believes that abortion is impermissible, while I believe abortion is permissible." This is a statement which makes sense, and would not be strange to hear. In a sentence such as this, do you hold that the use of the word "believes" is a category error? If moral utterances do not express beliefs, how can the above be true or false? Must it be false? How would you reform this sentence to preserve its meaning?

    Furthermore, atomic moral sentences can be used to construct valid arguments. Example,
    P1. Stealing is wrong.
    P2. If stealing is wrong, then getting your little brother to steal is wrong.
    C. Getting your little brother to steal is wrong.

    If moral statements express beliefs about matters of fact, this is no different from a common variety modus ponens. Otherwise, there is something strange going on here. If I understand you correctly, when I utter "stealing is wrong," I am impressing on my audience some imperative not to steal. This normative evaluation is capable of being correct or incorrect, by your account, but the meaning of the sentence is nonetheless to impress a particular intent. With this in mind, let us semantically dissect the above modus ponens.

    In P1, the atomic sentence "stealing is wrong" impresses an intent by its very utterance. This is due to the type of speech-act you claim it is. In a sense, the statement "stealing is wrong" cannot be disentangled from this force of impression.
    Yet, in P2, I state that "IF stealing is wrong," then this other thing is wrong. In this case, I am not committing to impress anything on my audience. The meaning of "stealing is wrong" as an atomic sentence appears different than its use as the antecedent of a conditional. This is problematic for the validity of moral modus ponens. Your theory will have to account for this in some way to be successful.

    One important difference in attitude toward an idea is sometimes called "direction of fit", in reference to the terms "mind-to-world fit" and "world-to-mind fit". In a "mind-to-world fit", the mind (i.e. the idea) is meant to fit the world, in that if the two don't fit (if the idea in the mind differs from the world), then the mind is meant to be changed to fit the world better, because the idea is being employed as a representation of the world. In a "world-to-mind" fit, on the other hand, the world is meant to fit the mind (i.e. the idea), in that if the two don't fit (if the world differs from the idea in the mind), the world is meant to be changed to fit the mind better, because the idea is being employed as a guide for the world. It is the difference between a picture drawn as a representation of something that already exists, and a picture drawn as a blueprint of something that is to be brought into existence: it may be the same picture, but its intended purpose changes the criteria by which we judge it, and whether we judge the picture, or the thing it is a picture of, to be in error, should they not match.

    Alright, I'm following so far. Whether or not this conception is accurate, I like it a lot.

    given the idea of a world where some people kill other people, I expect most will agree that that idea is "right" in the sense that they agree with it as a description (most people, I expect, will agree that the world really is like that, and an idea of the world that doesn't feature such a thing is descriptively wrong), but simultaneously that it is "wrong" in the sense that they will disagree with it as a prescription (most people, I expect, will agree that the world morally oughtn't be like that — whatever "morally oughtn't" means to them, which we're getting to — and that a world that features such a thing is prescriptively wrong). Same idea, two different attitudes toward it: the world is that way, yes; but no, it oughtn't be that way. Two different opinions, but about the exact same thing, different not in the idea that they are about, but in the attitude toward that idea.

    I take all kind of moral language, "good", "ought", "should", etc, to be conveying this kind of world-to-mind fit.

    To this point, I will simply offer an alternative explanation. I don't think that "same idea, two different attitudes toward it" captures what is going on in this case. Rather, I hold that these are two distinct ideas. One is that people do murder, the other is that people ought to murder. Somebody could agree or disagree(correctly or incorrectly) with either, in any combination. In accordance with the is-ought distinction, my agreement with one cannot logically follow merely from my agreement from the other.
    I find it interesting that you classify "ought" and "should" as distinctly moral language. I have more to say about this below.

    Saying that someone "ought" to do something, or that something "should" be or that it would be "good", doesn't necessarily mean it is obligatory. It could very well be a supererogatory good.

    I was thinking of supererogatory action when I typed that out, so I'm glad you caught it. While you are correct that "he is obligated to X" is much normatively weightier than "he ought to X," I still think that using "ought" to refer to supererogatory acts is a butchery of our use of the word. If you were to tell me "everybody on the planet ought to live life in constant ecstasy" or "you ought to sacrifice your life to save mine," I would strongly disagree with you. If you then told me that when you used "ought," you really just meant that these things would be good, I would come to agree with you, still believing that you had originally misused the word "ought." I will provide more reasons why I think this below.

    Correct ones of these non-descriptive but still cognitive opinions are not "facts" in the narrow sense, the sense that excludes mathematical claims. They could be called "facts" in a broader sense, but I find that that sense introduces unnecessary confusion, as "fact" seems to have inherently descriptivist connotations. The moral analogue of a "fact" is a "norm"; but NB that "norm" does not imply subjectivism, because "normal", "normative", etc, in their oldest senses, meant "correct" first and foremost, and it's only subjectivist assumptions that whatever everybody else is doing is correct that lead "normal" etc to take on the connotation of "what everyone else is doing". I don't mean it in that sense at all: a norm is just something that ought to be the case, exactly like a fact is something that is the case.

    Alright, this is the place where I want to talk about reasons and rationality. In all your theorizing on normativity thus far, I have only seen you touch on moral normativity. This is not the only type. I can make claims like:
    "You should stop smoking."(Prudential ought)
    "You ought to proportion your belief to the evidence."(Rational ought)
    "You ought not to beat your wife."(Moral ought)
    "You ought to do the thing that you have the most reason to do."
    I would like to touch on that last example a little more. "You ought to do the thing that you have the most reason to do." If you are anything like me, you find that incredibly intuitive. I have come to believe that it is essentially the definition of ought. The thing that a person ought to do is the thing that they have the most reason to do. Moral reasons are a type of reason, but often the thing that we have the most moral reason to do is not the thing that we have the most total reason to do(ergo supererogatory actions).
    Since moral oughts are not the only type of oughts we use, I am surprised to not see theorizing on broader normativity in your work. Questions like "Is X rational" or "Is P a good reason to Q" need to be answerable, or at least explainable, by any plausible theory of normativity.
  • Meta-ethics and philosophy of language
    Describing and prescribing are different kinds of speech-actions, and so each have different criteria for doing them correctly, successfully, which when fulfilled make the statements true.

    Interesting take. I look forward to reading your expansion on it. It seems to capture some of our evaluative language well, while to describe other forms, it appears to be more of a stretch. You raised the excellent example yourself of historical moral evaluation. Since ought does imply can, I find it unlikely that by saying something like "The Holocaust was wrong," we are literally prescribing a course of action for the dead perpetrators. Rather, I believe that we are presenting an evaluation on the moral fact of the matter.

    It cannot be correct to prescribe something that is logically contradictory, any more than it is correct to prescribe something that is logically contradictory. ("Ought" implies "can", as you say). Mathematics is about exploring the possibilities of different abstract structures, and the limits of that possibility limit what could be moral as much as they limit what could be real.

    I agree with this. Well put(assuming one of those instances of "prescribe" was meant to read "describe.")

    We can very well say that everybody ought to have access to adequate food and water and shelter and medicine etc, and nobody ever ought to have to die; those would be good states of affairs. Them being good states of affairs has some implications on what actions agents ought to do, but it's not directly a statement about what anyone ought to do.

    I agree with what you are saying here, somewhat. Some states of affairs are better than others. It is sometimes the case that agents ought to act in ways that actualize these states of affairs. I could rephrase your statement as "everybody having access to adequate food and water is good." The fact that this is good provides agents with a defeasible reason to bring it about, in other words, ceteris paribus, they ought to act towards this end. You could describe a good state of affairs as either "a state that is good" or "a state that ought to be the case," but of the two, I think that the former does a better job at preserving what we mean when we use moral language. I think there are some circumstances where a state of affairs is good, but yet, it ought not to be the case - no individual or group ought to(or, more strongly, is obligated to) act in such a way as to bring it about. Note here that I believe oughts can apply to groups of agents just as well as single agents, take for instance, "The senate ought to pass law L."

    I can't say I understand your opposition to non-naturalism. The argument from queerness(What would a moral fact even look like?) equally defeats all abstract facts. Yet, given your views on modal realism, logic, and mathematics, you have no issue dealing in objective facts that cannot be seen or touched. You seem not to consider them spooky. Do you consider the argument from queerness to be uniquely efficacious against moral facts for some reason? If all we can say about moral facts is that they have equally substantial grounding to other abstract facts(math, logic), that seems good enough to justify all the types of moral reasoning we like to use.
  • Meta-ethics and philosophy of language
    The broad sense is just the sense of “a statement that is correct” in any sense. The narrow sense is the sense of “a statement describing the world, that is correct”. Mathematical statements have implications about what can be real (which descriptions of the world can be correct), but they also have implications about what can be moral (which prescriptions of the world can be correct). They are more abstract than either description or prescription, and no more directly say what is real than they directly say what is moral.


    I see. I think that what is different between our ontologies must hinge on this notion of what it means for a proposition to be correct. What is your account of truth? I endorse a correspondence theory: P is true just in case P actually obtains. What does it mean for P to obtain? Well, that the real state of affairs is such that P. What else might it mean for a statement to be correct?

    I'm interested in how you suggest that mathematics has moral implications. Could you elaborate on this idea?

    A classic example of a formal logical inference is that from the propositions "all men are mortal" and "Socrates is a man" we can logically infer the proposition "Socrates is mortal". But, I hold, we could equally well infer from the propositions "all men ought to be mortal" and "Socrates ought to be a man" that "Socrates ought to be mortal". I say that it is really just the ideas of "all men being mortal" and "Socrates being a man" that entail the idea of "Socrates being mortal", and whether we hold descriptive, mind-to-fit-world attitudes about those ideas, or prescriptive, world-to-fit-mind attitudes about them, whether we're impressing or expressing those attitudes, even whether we're making statements or asking questions about them, does not affect the logical relations between the ideas at all.

    Sure, those arguments are both of deductively valid form. I find that to be a strange usage of the word "ought," however. Ought claims are made about agents, prescribing that they should take a certain action or accept a particular belief. I wouldn't consider, for example, "all men ought to be made of atoms" to be an ought claim that makes sense. Ought implies can, and whether or not I am made of atoms, or am mortal, is far outside the jurisdiction of one's agency. You cannot provide a reason, normatively speaking, that I should be made of atoms. It's unclear what this could even consist in. You could provide a causal explanation, but of course, this is distinct.

    I do actually support something like this, in the form of mathematicism (or the mathematical universe hypothesis), but for me that’s really an extension of modal realism: any world that can possibly be, is, both in terms of configurations of a universe with the mathematical structure that ours has, and in terms of other mathematical structures. But claims of possibility is not made true by their accurately describing these other possible worlds, but rather by the internal consistency of the structures they posit, and the other possible worlds are limited by that same requirement of self-consistency, so they necessarily coincide.

    I am completely unfamiliar with modal realism, so this does not make a lot of sense to me. I think the best approach here is simply for me to try and paraphrase what you've said and wrap my head around it. I take what you're proposing to be something like this:

    We often speak in terms of possible worlds, in many different senses. We can talk about physically possible worlds(consistent with the laws of physics), morally possible worlds(consistent with our world's moral truths), mathematically or logically possible worlds, and impossible worlds in every such sense. All logically and mathematically possible worlds(considering the relationship between logic and math, they may be one and the same set) actually exist. The only feature that can disqualify a world from existence is its instantiation of self-contradictory properties - these worlds are contained within the set of the logically and mathematically impossible worlds(a world where the law of non-contradiction is true, for example). As long as a world fulfills this single criterion, it is modally real. That is to say, we can conceive of it, discuss it meaningfully, and make true claims about it. Correct me if I'm mistaken in my understanding of your beliefs.
  • Meta-ethics and philosophy of language

    Not Pfhorrest, but I'll give my 2 cents. This isn't a problem unique to moral reasoning. Different people could use the utterances,

    "The woman who gave birth to me"
    "The nurse employed by me"
    "The one who is scared of spiders"
    "The person who stood in that Starbucks at 5:00 PM"
    "Ma-ma!"
    "La madre de este nino"
    "Hey, YOU!"

    All to refer to the same woman, my mother. We can attempt to remove ambiguity by providing further clarification. We can never fully resolve this problem, as a person can really attach any internal meaning to whatever words they like. But, most of the time, we seem to understand what other people mean and they seem to understand us. It's not the type of thing that can be or needs to be systematized: there is no hard and fast rule to figure out the meaning of words, one must just attempt to resolve it with their interlocutor.
  • Meta-ethics and philosophy of language

    What differentiates the narrow sense and the broad sense as you speak of them, other than the fact that the narrow sense is physical and the broad sense is not? Mathematical statements make claims that are sometimes true. What does it mean for something to be true if not for the proposition to correspond to reality? Of course, I agree with you that neither mathematical nor moral facts correspond to something physical. But if all you mean by "the narrow sense of existence" IS the physical sense, then the moral non-naturalist does not disagree with you at all. The moral non-naturalist also agrees that moral facts are not physical.

    If mathematical claims are not descriptive(describing a feature of reality), nor normative(counting in favor of an agent's doing or believing X), what kind of claims are they? To hold that moral claims are made true in virtue of "moral stuff," if we mean "stuff" in the usual sense, is naturalism. Non-naturalism would deny this, as do you, so I don't see where exactly you disagree with it.

    Mathematical principles do appear to be a part of reality. At the very least, some mathematical principles appear to accurately describe physical reality, while others do not. One principle may provide a model for how planets orbit stars, while another may describe nothing that we see. But, the truth of both follows from the same axioms. They are both true for the same reasons. "Physical" versus "abstract" is a useful distinction, but is it the very same distinction as "real" and "not real?"

    Does something seem so objectionable about referring to "mathematical reality," of which our present physical reality is only one of many possible subsets? We could imagine a world with different fundamental forces, or distinct physical composition, before we could imagine one with square circles and finite primes. The concept of a three-sided triangle corresponds to reality in a way that a four-sided triangle does not. When I say "a four-sided triangle does not exist," I am saying something MORE than that a four-sided triangle is merely not physical. After all, a three-sided triangle is also not physical, yet there is a difference between them.

    Apologies for the wall of text, but I'm very interested in this topic.
  • Meta-ethics and philosophy of language
    For those like me whose physicalist ontology cannot admit of some strange non-natural but nevertheless real moral facts, non-naturalism is a non-option. (This is also called the Argument from Queerness: wtf is a non-natural moral fact like?).


    I think you may be overvaluing the argument from queerness. Do you have decisive reason to believe that only physical, natural facts exist? Take the true statement that there is an infinite amount of prime numbers. What is it true about? Is there something physical, or tangible, that you can present to me as an instantiation of the infinity of prime numbers? Mathematical truths are certainly not prescriptive, they are descriptive, but what do they describe?

    The simplest answer would be "abstract objects." Do abstract objects exist? Well, if "P exists" just MEANS "P is physical," the answer is no. But, this may be an unnecessarily limiting conception of existence. Perhaps what it is for something to exist is, instead, for there to be a state of affairs that can ground factual claims about that thing. The infinity of the real numbers between 0 and 1 is greater than the infinity of the natural numbers. This statement is true. Is it spooky because it corresponds to a state of affairs that is not physical? I think not.
  • What are people here's views on the self?

    I don't think there is a significant difference, as I am a reductionist about personal identity. Many people are not, and would believe that there is a meaningful distinction to be drawn between the "real them" and a copy of them. They might account for this difference as:

    1. The real me is the body that contains my soul, essence, or ego, while the copy does not.
    2. The real me is that body from which an unbroken spatio-temporal line can be drawn from it to my origin(in a copy's case, one cannot).
    3. In the case of a copy and an original, there is some special property that is only attributable to the original. This special property is what we should be concerned with in preserving our consciousness.

    Would your attitude towards the case change if the teletransporter malfunctioned, leaving Earth-you alive? Weeks later, when Mars-you returns to inhabit your home, does he have the same claim to it that you do? Does one of you have a stronger claim?
  • What are people here's views on the self?

    To be fair, if we are considering a scenario with teletransportation, we aren't having a discussion that's constrained by the limits of technology. We could imagine flawless organ transplants, and these questions would remain about what effects they could have on our identity, if any. Questions about body transplants can be dismissed as unlikely, as most hypotheticals can, but maybe they are worthy of greater consideration.

    Also, my point in bringing up cell replacement was this: at 65 years of age, you have undergone so much cell replacement that very little of your physical 20-year-old self remains. If this degree of replacement is equivalent to death, we would have to regard people as dying simply within the process of aging.
  • What are people here's views on the self?

    If just your body's matter changing gradually leads to death, you die in this sense multiple times throughout your life, inevitably. If you are not concerned about these deaths, once every eight years or so, do you have reason to hold special concern about teletransportation?

    While you do need an organ, it's less clear that you need your specific organ. You could live a perfectly fine life with a donor lung. Why not two donor lungs? Why not a donor heart, a donor liver and a donor stomach? What about just taking out your brain and moving it to a whole donor body?
  • What are people here's views on the self?

    I feel like that's very broad. It could be said that my kidneys are required to function "most" effectively, yet, we usually don't think of people as having died/lost their identity when they have a kidney removed.


    I am inclined to agree with you: your account of the scenario is a complete one. P at T2 experiences sufficient psychological continuity with P at T1. That is what I think matters, and why I am not concerned by teletransportation. The material we are composed of changes gradually over the course of regular life. We may not have reason to think teletransportation is any worse than this. It might be viewed as merely an unusual process.
  • What are people here's views on the self?

    You mention it being imperative that the body is held in one piece. Do you believe this to the most extreme extent? Say, that I cannot lose an arm without losing my self? Could I have all my limbs transplanted and remain "me"? Which parts of my body do you hold to be necessary for the survival of my self?
  • What are people here's views on the self?


    When you say your mind, do you mean your brain, or something else? Do you think that the self is something separate from the physical, yet dependent on it, or something reducible to the physical?