You guys keep quoting me. I’m a different fool. Btw this conversation seems rather childish.

No one is evil, clearly. To reduce human strife to anything but humanity itself seems superstitious. So I agree.

The question that interests me is What makes you morally culpable? Are certain beliefs morally obligatory? If you fail to understand what actually causes the behavior you attribute to someone’s race, is that a moral or an intellectual failure? By analogy, you wouldn’t fault a deaf guy for ignoring your calls. I’m tempted to suggest goals and desires - hell, maybe just actions. But it’s all wound up with what you believe.

It would help to know what does qualify as evil, if anything.

I just saw it myself. I knew it would be Marvel/DC (or whatever it’s called) trash, but I was curious about the message and wanted to support the production for other reasons. Pretty shallow overall, but I appreciated the bad guy character, whose name I’ve already forgotten. He was the voice of pain and anger that the movie had to provide, but they avoided condoning his aggression by making him a villain.

this is where I make an ironic point about the number of posts before a Quine reference

I don’t think I have anything original to say about the philosophy around psychology, but I do have a unique perspective. I design tests to measure advertising. Everyone on my team has a copy of Experimental Design: Procedures for the Behavioral Sciences , by Roger Kirk. For anyone who doesn’t know, ‘Design of Experiments’ (DoE) is a topic-neutral synonym for ‘scientific method’. Many of the leading DoE experts are psychologists precisely because behavior is the trickiest terrain in the eyes of science. Ironically, there’s a (weak) sense in which psychology is the most scientific science.

That said, DoE got me interested in psychology, and I’ve felt increasingly disappointed following the rise of cognitivism. One hopes to exchange Skinner for some kind of methodological behaviorism, but now we’re inventing all these “systems” or hypothesizing that love can cause a starved cat to ignore food for its owner’s affection. Much psychology is just glorified market research.

@Jeremiah

If you revise the parameters to say the teacher picks one and only one of A or D to be true regardless of the content of the answer, then I will accept your analysis. Then the answer is simply 25% bc, even though it’s listed twice, it doesn’t matter what the answer choices actually say. We can observe that 50% of the students get the semantically correct answer, but we don’t go in circles because we stipulated that the criterion of correctness has nothing to do with the actual question, so the actual frequency of correct answers is irrelevant to the teacher’s designated answer choice. I appreciate the conundrum that, on some interpretations, the meaning of the question is part of the format of the question.

By the way, I appreciate your dialing down the vitriol. I am happy to debate this as friends.

I would call that more of a guideline, second to the guideline that your theory should be empirically adequate. That said, I don’t consider this a question of theory choice. It’s a straightforward exercise in chance with a self-referential but non-circular twist. I quite enjoyed it.

I think this is a meta-ethical question. Looks like we’re arguing about what is actually ethical. I expected this thread to address realism v. anti-realism. It’s right to ask, as OP did ask, what an account of objective morality would require, but I would suggest less focus on the requirements of specific theories and shift to the requirements of any theory. Not that my preferences matter, but I’d be entertained by a thoughtful challenge to expressivist/non-cognitivist views. The field is not without controversy, but I’d expect views like that from the typical naturalistic philosopher. Rational Choice Theory is a useful framework, esp as it concerns agency and behavior, but it presupposes its own adequacy.

That is interesting.

Thank you, sir/madam.

I mean, I agree that it’s a defective prompt in the pedagogical sense. I would protest as a student. But I’m gonna carry on with the analysis.

Btw, I accidentally clicked smthg on your post when I went to reply. Dunno what it was or what it did. If I somehow flagged you or smthg, I am profusely sorry.

I don’t see the distinction. If I had to choose what I’m trying to analyze, it would be the latter. I agree, we don’t care how many students successfully mark an answer. How many of them mark an answer that is the same as the correct answer to the question? Well, the correct answer isn’t listed in the prompt, so none of them sends it back.

I’m construing it as a true/false check on a send/receive transaction. The test sends a prompt and sets a “correct answer” variable on the back end. The test-taker sends a response. The program checks to see whether the response equals the correct answer. You could run that program, no problem. The frequency would be 0%, but the program runs.

In that case, broken questions can still be coherent.

Yeah. I didn’t say it would be fair. It would be coherent. The test also wouldn’t have any pedagogical use, but that’s a practical problem.

You might say the two-event analysis looks more baroque, but it’s also more adequate.

I’m not sympathetic to the “broken question” view. I assume there’s a correct answer, and then I assume the answer choice selection is random, as stipulated. The meanings of the question and answer choices don’t affect random selection. Of course, you revise to 0% after noticing the formally correct answer isn’t given, but this defect doesn’t have anything to do with the duplicative answer choice. You could imagine a teacher doing that and still successfully grading the test.

The argument shouldn’t be about the answer choice sample space. There are 4 elements and, when you de-dupe, the frequency of ‘25%’ has to be 50%. But given that weighting, we want to know how often the chosen answer is the correct answer to the question. That implies a second chance event, the (formally) unpredictable outcome of what the correct answer actually is. That sample space only has 3 elements.

I debated an “agnostic atheist” at length some years ago. I ended up writing a full-on essay to educate him. I stopped some way through because it wasn’t worth my time. Still got the back-and-forth and the rough draft. Thesis was that any appeal to evidence commits you to a position.

The most interesting topics addressed were the outdated empiricist assumptions behind typical atheism and the increasingly popular option to “not reject the null hypothesis.”

Notice that I never called you unintelligent, and I actually thanked you for your entertaining post.

So we agree that we're dealing with two separate chance events. We've also precisely clarified the difference in our logic. Your scenario makes sense to me. If the teacher just picks one of A-D to be correct without considering that A and D are equally justified, then the chance that the content of the teacher's designated answer choice (A-D) is the semantically "correct" answer would need to be weighted by the frequency of its appearance in the list.

Also curious how mature and constructive this conversation becomes in the absence of some contributors.

Ah, that's the difference, thanks. Sorry, wasn't going to chime in until I could re-read the whole thread, but this settles it.

Here's what I'm saying:

possible_choice <- c("25", "50", "60", "25")
possible_correct_answer <- c("25", "50", "60")
s <- 0

for (i in 1:1000000)
{

if (sample(possible_choice, 1) == sample(possible_correct_answer, 1))
{
s = s + 1
}

}

print(s / 10000)


Looks like you're correctly weighting the selection from the answer choices A-D, but I would differ with you about how you're weighting the probability of each answer choice being the correct one. I'd argue the frequency of the answer's appearance in the multiple choice list should not affect the chance of the answer being correct. My profile pic now has our two scenarios compared.

Strange, that’s exactly what I did, but I get a random variable centering on 33%. The logic must be different, but I don’t know PHP. I can work with R if you’ve got an equivalent. It sounded to me that your logic treats the two identical options somewhat differently. In your example of the classroom debate over the right answer, the teacher doesn’t acknowledge that A and D are the same, else there would be no debate. In any case, it sounded like your logic accounted for external factors, whereas my contention is purely about framing the random events. I will review your comments later today in case I missed something.

There are two random events: the random selection of the actual correct answer from the list of 3 unique possibilities (ignoring that the right answer isn’t actually in there), and the random selection of the answer choice from a list of 4.

If you took 10000 questions of the same form, disregarding what they actually ask, this is how you would analyze them. You would predict random accuracy of 33%.

Jeremiah, if you really want to prove us wrong, set up a 10k question test and show us that it’s not 33%. Now, I don’t mean run 10k samples from an array of 4 elements. That’s only half the exercise. I mean 10k questions with 3 possible answers and 4 choices with 1 answer repeated. I did it already, results in my profile image.

@Jeremiah

There he is. Good timing, was just checking one last time. I am a fool, but I know intelligence when I see it. I will keep my thoughts about you to myself. Good day, sir.

@Jeremiah

Your answer to this question will go far toward clearing things up.

I’m thinking of a number. It’s either 1, 2 or 3. If you randomly choose a number from the list below, what are the odds that it’ll be the one I have in mind?

1, 2, 3, 1


If your answer is not 33%, I must question your mathematical literacy. If it is 33%, then we’re not arguing about the same thing.

Just ran the experiment in a spreadsheet. The outcome is in my profile image. Before any iterative process of revising the actual correct answer, the abstract solution is 33%.

@VagabondSpectre I hadn't considered how to structure the iterative game. That's an interesting take.

Yeah, thought we were all on board with that.

Granted. That’s not at issue, right?

There are 4 answer choices, each 25% likely to get picked. There are 3 unique possibilities, each 33% likely to be correct. Whatever the distribution of your sample, the chance of being right is still 33%.

Newp. The chance of selecting one of the 3 choices is different from the chance of selecting the correct answer choice. Look at my profile image. The compound event is (1/4)(1/3) + (1/4)(1/3) + (1/2)(1/3) = 1/3. Circularity aside, this is an easy problem guys. We don’t need R.

They don’t. I didn’t see anyone say that. Answer choice ‘25%’ has a 50% chance , obviously. We noticed that.

Only half paying attention here, but it looks like you’re confusing the probability of selecting a unique answer choice with the probability of selecting a correct answer choice. We’re looking at the compound event.

Since some of you seem to be struggling with this. Also see my profile pic if that's easier.

First Glance Schematization
Answer Choices Chance Correct Chance Selected Chance Correct & Selected
AC 1 33% 50% 17%
AC 2 33% 25% 8%
AC 3 33% 25% 8%
Total (also the correct answer) 33%


Revision, Seeing 33% Not Listed
Answer Choices Chance Correct Chance Selected Chance Correct & Selected
AC 1 0% 50% 0%
AC 2 0% 25% 0%
AC 3 0% 25% 0%
Total (also the correct answer) 0%


Revision, Seeing 0% Not Listed
Answer Choices Chance Correct Chance Selected Chance Correct & Selected
AC 1 0% 50% 0%
AC 2 0% 25% 0%
AC 3 0% 25% 0%
Total (also the correct answer) 0%

Fun exercise. Here’s how I experienced it.

*read*

“Hm. Multiple choice random accuracy is 1/p for p choices, so the correct answer to this question is 1/p.”

“Oh, but 1/p shows up twice. More accurately, it’s 1/(degrees of freedom). If you know the full list of answers and you’ve picked 3, then you know what the 4th one is before you look at it, so it’s 1/(p-1)=1/3.”

“So the answer to this particular question is 1/3. But wait, no 1/3. This question’s correct answer is not listed, so there’s no chance of picking it. But wait, looks like 0% is actually the correct answer. Also not listed. The probability is still 0.”

“They tried to get me with the two 25%, tempting me to think the answer is 2*25%=50%, but that’s assuming the repeated answer choice doesn’t affect the odds. As an exercise in probability, it’s def 1/3, because we are picking at random and without assumptions about the meaning of the question. Strangely, I did have to know the meaning of the question in order to know the answer wasn’t listed, but now I’m just clarifying. My psychology does not affect the structure of the question.”

“If 0% were listed (once) alongside the two 25%, you would indeed go in circles. Finding no ‘1/3’, you revise to 0. Seeing 0 once, you revise again. Now thought processes divide. The meta view returns to 1/3, alternating with 0. The semantic view alternates between 1/2 and 1/4.”

“Does the fact that 25% is repeated change the odds? That’s double the risk of picking an answer that’s only 33% likely to be right. Well, no. The weighted average is still 33% because each hoice, being one of the three possible answers, has a 1/3 chance of winning.”

Very pleasant. Thank you.

I’m not optimistic this will go anywhere before everyone agrees on the definition of meaning.

That said, this seems simple to me. We have a naturalistic account of life which centers on reproduction. Get yourself some self-copying organic molecules and everything else follows. One end of it is human agency. We have (descriptively) rational goals, purposes, intentions or whatever. Some are built in, e.g. pleasure and survival. Finally, we have linguistic meaning, interpretation and narration, which is important for community and socially-derived sense of self.

Somewhere in the mixture of mortality and social instincts, we get the emotional drive for fulfillment we vaguely call meaning. I’m skeptical that it somehow transcends the ordinary varieties listed above.

@StreetlightX

I hesitate to opine on this bc I know nothing about neuroscience, but I feel very sympathetic to what you said.

Political problems are inherently normative, so it’s hard to see how you can naturalize them, i.e. address them scientifically. Sure, there are relevant empirical “how” questions, mostly economic, but the fundamental problems are “why” and “should” questions.

That said, Dewey had an interesting pragmatist’s view of democracy. He considered it a kind of experimental government and thought the main virtue of democracy is the same fallibilistc, self-correcting element that drives progress in science. It’s an attractive idea, but I still see a big descriptive/normative gap between politics and science.

The most intelligent approach to political philosophy I’ve yet seen is Rawls’s Theory of Justice. It’s been heavily criticized, but I still buy the main thrust of the argument, which is that these normative questions might be objectively decidable within a decision theoretic framework. In other words, Rawls argues for objectively correct political solutions using principles of rational choice theory.

To take a stance at long last, I would attempt to naturalize politics by framing the normative problems within a rational framework like Rawls’s, thus reducing the “should” problems to empirical “how” problems where we can lean on the tools of natural science. On the other hand, I don’t see how the implementation of rational, science-based policy is compatible with democracy.

(How would you distinguish a tiny systematic bias in the data from noise? Noise doesn't have to be perfectly unbiased!) — SophistiCat

Nice catch. I was going to say the bias is the giveaway. If the acceleration were independent of distance, it would make no difference, I agree. Maybe G would be slightly incorrect, but we’d never know. The acceleration is proportional to distance, though, so error would increase with distance. To increase accuracy beyond a certain point, we’d have to correct G for distance. Im imagining an elliptical orbit with constant acceleration (0 or >0) away from its barycenter. That path would look different from one where the accleration varies with distance. As someone with a lot of quantitative experience, I would look at patterned error as evidence that my model is fundamentally wrong. Obv, some models involve patterned error by definition, but I’m thinking of bias you can’t account for. Ultimately, this comes down to precision of measurement, where you seem very doubtful any inquirers could reach the necessary precision to notice the issue. The evidence is on your side here. I just hesitate to rule it out for all observers at all times.

I don't think much about Krauss's philosophizing, but on this question he is undoubtedly an authority that ought to be taken seriously. He may be wrong, but it is far more likely for you or I to be wrong about this than it is for him.

Yes. Just yes. It’s easy to impersonally challenge a worldwide name on an Internet forum. I would not want to publicly challenge him on scientific questions. The only thing that reassures me is his universal quantification over observers and times, at a moment when we still don’t know what our own science will eventually tell us.

There is no good reason to think that scientific theories inevitably converge towards some fixed final shape.

Not sure. I’d like to debate this sometime, after we get some definitions out of the way.

A civilization with no access to the same observations that we have might nevertheless entertain theories that predict those same observations, and might even in some cases find strong indirect support for such predictions.

Agree. I’m just saying I don’t think Krauss is right to rule it out so categorically. Of course, I accept that it’ll most likely happen way, way less often.