Inference to the best explanation is a really interesting philosophical topic.
Let's start the story with Descartes. Skepticism becomes an option--what if we're all being deceived? What if our eyes and ears are lying to us? Descartes tried to answer this question by starting with some firm knowledge and slowly justifying almost all of our knowledge from a few secure facts. In this way all our purported knowledge would be justified, and we would be allowed our confidence again.
As the story goes, not everyone agreed with Descartes secure truths. So other attempts at justifying our knowledge, at securing epistemology, began. Locke and Hume tried to justify human knowledge by matching all knowledge with observation. But, famously, this only gets you so far. So empiricism doesn't work to give us complete, skeptic-free confidence in our beliefs.
(Here I fudge the story a bit.) So what do you do now? Is it a free for all just because we don't have an answer to the skeptic? No, it's not. What we do is acknowledge that justification needs to end somewhere, and as it turns out justification for our beliefs ends before the point where we would feel totally secure in our beliefs. So there isn't any answer to the absolute skeptic. But that means we need to figure out where to stop.
So the game becomes trying to build your house as close to the cliff as possible. You lose if your house falls off the cliff, and you also lose if your house ends up looking like a mess. That is, the game is to find a starting point with as few assumptions as possible (satisfying a general desire for parsimony in the absence of justification) and building and justifying as much of our common-sense and scientific beliefs as possible. You lose if you end up with skepticism, and you also lose if you end up justifying silly beliefs, like beliefs in fairies or witches.
Right now I'm reading a paper where someone tries to say "We don't have to stop the trail of justification at inference to the best explanation. We have a way to justify inference to the best explanation that makes sense." As it turns out, that way of justifying inference to the best explanation is designed to allow for other kind of "inferences" besides inference to the best explanation. Specifically, it wants to allow for inference to the only things that make deliberation possible, which would include normative facts.
As far as I can tell, the debate about this paper has to be (obviously about the details of his argument, he needs to show lots of little points in order to get his big points into play) about whether he tried to go too far back. If his stopping point is no better than Inference to the Best Explanation, I think he'll find many people saying "Hmm, yeah that's interesting. I'm going to build my house right over here on ground that's a bit more secure, right here with Inference to the Best Explanation."
[His response will be, but building your house over there you still haven't made deliberation possible, but that's the second half of his debate. If the first point is dependent on the second, then he's got a very different argument on his hands.]
There's also a debate (that I know nothing about) where people wonder if inference to the best explanation is the right place to stop for reasons that have nothing to do with ethics.
Also, an observation: there are no arguments for mathematical or ethical realism that are particular to math or ethics. That's because realism itself isn't restricted to math or ethics. You win the realist game if you define plausible methods for coming to physical truths, and then use those same methods for reaching mathematical or ethical truths. This is exactly what inference to the best explanation/indispensability arguments do.
Sunday, October 25, 2009
Monday, October 12, 2009
A weird idea about reductionism
What is reductionism? Here's an example of a reductionist approach--a reductionist approach to metaphor. You start with a sentence that seems kinda weird: "The city is a jungle; you've got to take care of yourself." So, on the surface this sentence seems to be quite similar to sentences where you identify two things, or at least apply some property to some subject, EX: "Michael is that guy" or "The eagle is a giant bird." If you classify "The city is a jungle" with those sorts of sentences you get really screwed up results. What I mean is, if you take "The city is a jungle" literally you'll either think that I'm classifying the city in the category of jungles--cities are jungles--which is just stupid. Obviously, that's not how you're supposed to read "The city is a jungle."
Quite clearly, what we need is to analyze what "The city is a jungle" means. What it really means is "The city is similar to the jungle in certain ways [e.g. it's dangerous and complex]." The point is that you start with some weird way of speaking that we don't think that we should take literally, and we reduce it to a level of discourse that we're more comfortable taking literally. This is what I just did with metaphor.
A more philosophical example: We start with the notion of causation, and we think that it's mysterious and confusing (Hume thought this). He doesn't think that a literal understanding of causation makes much sense, so he reduces it to a more down-to-Earth notion. So he takes a sentence "A causes B" and translates it as "Whenever B happens, A happens too." He's reduced discussion of causation to a discussion of correlation, the occurrence of two events at the same moment.
People do this to ethics too (Harman recommends that one has to do this in order to maintain ethical realism). "Saving lives is good" sounds a bit spooky, and it's not clear what things in the universe ethics is talking about, and if it's anything it would seem that it would have to be abstract objects or properties that you can't see, smell, touch, etc.... There's a bunch of problems. So people say, "Well, let's take a reductionist approach to ethics; we'll translate ethical statements into normal ones that we feel more comfortable with." So this, for example, can mean that ethics gets translated into the language of emotions--after all, we all agree that people's emotional reactions exist, and so everyone should feel comfortable talking about that. For example, you might reduce/translate "X is good" or "Y is bad" into "I like X" or "I don't approve of Y." (That would be called "emotivism", and it's a reductionist approach to ethics). If you're willing to translate everything into the language of emotional responses, then you no longer have to say that there's anything special about the language of ethics; truth in ethics is just truth about evaluating the way you feel. There are no ethical facts, only physical facts, and in particular, physical facts about emotional responses to situations and actions.
I wonder, though, what if we were to try to reduce physical facts into ethical ones? That's a bit of a loony idea. But suppose that we felt comfortable with ethical facts and uncomfortable with physical ones. What's stopping us from trying to reduce physical facts to ethical ones? If it works in one direction, it should be possible to do it in the other.
How would that project go, though? We would need to find an ethical translation of all statements that refer to physical objects of properties! Let's take an example, "There is a zebra eating grass behind the barn." Now, could we simply translate this as "It's good that zebras eat grass behind barns"? Of course not, for a bunch of reasons. First, because it's not really a translation--we're still referring to zebras, grass, and barns, and this means that I'm still committed to the truth of some physical facts. Second, because a reductionist approach works when you can capture what is meant by the original sentence (for the most part) in the translation. There will be lots of situations when we would want to say, intuitively, that there is a zebra is eating grass, but we wouldn't always want to say that that's a good thing. Come to think of it, while the second is true, the first problem I mentioned is the real problem.
So, what's the reason behind this failure? Part of the problem is that there aren't any particular ethical objects that we can employ. The language of physical objects is very rich, and the language of ethical objects is quite poor. So what could we do to correct this? Maybe expand our circle out from ethics, and let's employ all talk of values (this will include discussions of beauty, simplicity, etc.). It's still no good, I think, and I'm willing to diagnose the problem in the following way: value language has a lot of predicates, concepts, properties, but very few objects.
(So there's a few complications to what I wrote. First, for naturalists there are only scientific facts, so to ask whether I can translate physical facts into ethical ones is a question that wouldn't make sufficient sense. Also, I'd have to show how this argument works in math. Could we begin to talk about reducing physical concepts to mathematical ones? Probably not, but why? Is it because the language is so sparse. So then what's my point here?)
The absence of ethical objects in our everday ethical manner of speaking is a complication that I run into in my research. The indispensability argument in math concludes that mathematical entities exist; it's not clear that anybody really wants such an argument to work in ethics. We want the objectivity of ethics, but ethics-talk usually involves applying predicates/concepts/properties to physical things--acts, deeds, states of the world, rules, whatever.
I think that this is a more precise way of saying Harman's argument. The reason why ethics seems to be dispensable, the reason why our best explanation of the world doesn't need it, is because it doesn't make reference to any ethical objects, just ethical concepts/predicates. And part of the reason why math seems indispensable is because we're making reference to mathematical entities.
Earlier I posted about why existence--ontology--should matter to us. After all, even if we have an argument that concludes that numbers exist we won't start suddenly bumping into numbers on our way to the library, work or school. Our lives will be the same; it's our perspective on the world that is liable to change. And I put out the following idea: the reason why existence is interesting is because ontology seems to secure semantics. We know that some sentence can be true or false if it refers to objects that actually exist--if it's actually talking ABOUT something in the world. So ontology becomes a handmaiden of semantics. Some philosophers disagree with this; they think that it's interesting to investigate the world to settle the question of what exists and what doesn't. I'm not sympathetic to that view. We know what it's like to live in the world, and if we're positing the existence of acausal, abstract objects that shouldn't really change my life. But it still seems to matter whether stuff exists or not, and I argued that the only reason that I knew of was because if X exists it seems that it could be true or false to say stuff about X.
If what I've posted is right, then there might be very different problems facing philosophers of math and metaethicists. Philosophers of math need to secure the objectivity of math by saying that mathematical objects exist. But ethics doesn't really make reference, in general, to ethical objects. They just apply ethical predicates to regular, normal physical objects. So the question of objectivity in ethics might be the kind that our form of the indispensability argument can't really touch.
Quite clearly, what we need is to analyze what "The city is a jungle" means. What it really means is "The city is similar to the jungle in certain ways [e.g. it's dangerous and complex]." The point is that you start with some weird way of speaking that we don't think that we should take literally, and we reduce it to a level of discourse that we're more comfortable taking literally. This is what I just did with metaphor.
A more philosophical example: We start with the notion of causation, and we think that it's mysterious and confusing (Hume thought this). He doesn't think that a literal understanding of causation makes much sense, so he reduces it to a more down-to-Earth notion. So he takes a sentence "A causes B" and translates it as "Whenever B happens, A happens too." He's reduced discussion of causation to a discussion of correlation, the occurrence of two events at the same moment.
People do this to ethics too (Harman recommends that one has to do this in order to maintain ethical realism). "Saving lives is good" sounds a bit spooky, and it's not clear what things in the universe ethics is talking about, and if it's anything it would seem that it would have to be abstract objects or properties that you can't see, smell, touch, etc.... There's a bunch of problems. So people say, "Well, let's take a reductionist approach to ethics; we'll translate ethical statements into normal ones that we feel more comfortable with." So this, for example, can mean that ethics gets translated into the language of emotions--after all, we all agree that people's emotional reactions exist, and so everyone should feel comfortable talking about that. For example, you might reduce/translate "X is good" or "Y is bad" into "I like X" or "I don't approve of Y." (That would be called "emotivism", and it's a reductionist approach to ethics). If you're willing to translate everything into the language of emotional responses, then you no longer have to say that there's anything special about the language of ethics; truth in ethics is just truth about evaluating the way you feel. There are no ethical facts, only physical facts, and in particular, physical facts about emotional responses to situations and actions.
I wonder, though, what if we were to try to reduce physical facts into ethical ones? That's a bit of a loony idea. But suppose that we felt comfortable with ethical facts and uncomfortable with physical ones. What's stopping us from trying to reduce physical facts to ethical ones? If it works in one direction, it should be possible to do it in the other.
How would that project go, though? We would need to find an ethical translation of all statements that refer to physical objects of properties! Let's take an example, "There is a zebra eating grass behind the barn." Now, could we simply translate this as "It's good that zebras eat grass behind barns"? Of course not, for a bunch of reasons. First, because it's not really a translation--we're still referring to zebras, grass, and barns, and this means that I'm still committed to the truth of some physical facts. Second, because a reductionist approach works when you can capture what is meant by the original sentence (for the most part) in the translation. There will be lots of situations when we would want to say, intuitively, that there is a zebra is eating grass, but we wouldn't always want to say that that's a good thing. Come to think of it, while the second is true, the first problem I mentioned is the real problem.
So, what's the reason behind this failure? Part of the problem is that there aren't any particular ethical objects that we can employ. The language of physical objects is very rich, and the language of ethical objects is quite poor. So what could we do to correct this? Maybe expand our circle out from ethics, and let's employ all talk of values (this will include discussions of beauty, simplicity, etc.). It's still no good, I think, and I'm willing to diagnose the problem in the following way: value language has a lot of predicates, concepts, properties, but very few objects.
(So there's a few complications to what I wrote. First, for naturalists there are only scientific facts, so to ask whether I can translate physical facts into ethical ones is a question that wouldn't make sufficient sense. Also, I'd have to show how this argument works in math. Could we begin to talk about reducing physical concepts to mathematical ones? Probably not, but why? Is it because the language is so sparse. So then what's my point here?)
The absence of ethical objects in our everday ethical manner of speaking is a complication that I run into in my research. The indispensability argument in math concludes that mathematical entities exist; it's not clear that anybody really wants such an argument to work in ethics. We want the objectivity of ethics, but ethics-talk usually involves applying predicates/concepts/properties to physical things--acts, deeds, states of the world, rules, whatever.
I think that this is a more precise way of saying Harman's argument. The reason why ethics seems to be dispensable, the reason why our best explanation of the world doesn't need it, is because it doesn't make reference to any ethical objects, just ethical concepts/predicates. And part of the reason why math seems indispensable is because we're making reference to mathematical entities.
Earlier I posted about why existence--ontology--should matter to us. After all, even if we have an argument that concludes that numbers exist we won't start suddenly bumping into numbers on our way to the library, work or school. Our lives will be the same; it's our perspective on the world that is liable to change. And I put out the following idea: the reason why existence is interesting is because ontology seems to secure semantics. We know that some sentence can be true or false if it refers to objects that actually exist--if it's actually talking ABOUT something in the world. So ontology becomes a handmaiden of semantics. Some philosophers disagree with this; they think that it's interesting to investigate the world to settle the question of what exists and what doesn't. I'm not sympathetic to that view. We know what it's like to live in the world, and if we're positing the existence of acausal, abstract objects that shouldn't really change my life. But it still seems to matter whether stuff exists or not, and I argued that the only reason that I knew of was because if X exists it seems that it could be true or false to say stuff about X.
If what I've posted is right, then there might be very different problems facing philosophers of math and metaethicists. Philosophers of math need to secure the objectivity of math by saying that mathematical objects exist. But ethics doesn't really make reference, in general, to ethical objects. They just apply ethical predicates to regular, normal physical objects. So the question of objectivity in ethics might be the kind that our form of the indispensability argument can't really touch.
Thursday, October 8, 2009
Summing up the way I've been thinking so far
I'm about to dig into a more substantial phase of research now. Over the next few weeks I'm going to be trying to understand what an explanation of the world is, and what the differences are between science, math, and ethics in this regard. So before I do, I want to clarify what I'm going into this research with, what my hypothesis (of sorts) is. I can't really defend this view; it's just a starting point, a bunch of suspicions that I have.
Without a doubt there is something that is quite different about ethics, science and math. Pinning down exactly what is really the challenge, and the next challenge is trying to figure out what those differences should mean--why the differences matter.
My general hypothesis is that hard, strong, epistemological lines tend to break down under stress. For example, Quine convinced most of the philosophical world that there is no strong, philosophically useful distinction between the meaning of words in a sentence and the substance of those sentences; there's no strong dichotomy between analytic sentences and synthetic ones. The argument comes down to just applying a great deal of stress to the dichotomy, and watching it collapse under investigation. I expect, coming into a philosophical investigation, a similar thing to happen when we've set up boundaries between realms of knowledge. Facts/values, empirical/non-empirical, a priori/a posteriori, history/science, objective/subjective and of course science/math/ethics are all ways that people have tried to divide up realms of knowledge. The problem is not dividing up realms of knowledge itself--everyone does that because there are real differences between the areas--but when it comes to epistemology, how we know what we know about these realms, I expect the dichotomies to break down. I expect that the way we gain historical knowledge isn't very different from the way we gain scientific knowledge. I expect that the way we gain mathematical knowledge isn't so different from the way we gain scientific or ethical knowledge. That's my hypothesis, in general. It's heavily influenced by Quine and Putnam (and Dewey, through Putnam), I think. At the very least, it's influenced by a misinterpretation of Quine and Putnam.
So, in particular, I expect that the distinctions between how we gain ethical knowledge and how we gain scientific knowledge break down under pressure. I expect that all knowledge is more or less in the same boat. I suspect that the differences are all of quantity, and not quality of knowledge. So I suspect that any attempt to explain why scientific stuff is objective and ethical or mathematical stuff is not eventually breaks down. I think that the indispensability argument goes a good way towards showing how the distinction breaks down between math and science, and I suspect that something similar can be adapted for ethics. I think that there are at least two good, promising ways of doing this, but in the end every way of making the argument does it by trying to make science a little bit more modest.
But I started by saying that there are real differences between math and ethics and science? It's just a fact that there are no ethics laboratories in universities, and it doesn't strike us as a very good idea to start such labs. Why is that? An analogy from math is useful in clarifying the question. Once Quine/Putnam argue that math is actually empirical knowledge and as objectively known as science, they need to answer the question: what fooled people for so long? Why did people think that math was a priori and divorced from experience? So, if you argue that ethics is on the same par as science, you have to explain why ethics strikes people as being the sort of thing that it doesn't make sense to start a lab for.
Of course, you could say that people are wrong, and that we really should be building ethics labs of some sort. Some people--I think that I heard this in the name of Nagel or Parfit--think that ethics is just a young science, one that's bound to develop the way that other sciences have. So maybe these people think that opening ethics labs makes sense. But I'm more sympathetic to the position that there is something about ethics that makes such a notion strange. I guess this could be consistent with the view that we should open labs, but I guess the view I find most attractive is that ethics is really really really hard. (This is also the reason why I'm not swayed by an argument from disagreement that ethics is subjective. Disagreement is consistent not only with subjective views, but also with really really really hard ones.) I bet that I could even show that some things that eventually fell into the realm of science were considered subjective problems, ones that it wouldn't make sense for a lab to study. For example, I bet a lot of brain stuff fits this pattern.
How does this relate to the original program of contrasting math with ethics? Well, the idea is that by taking an argument that's found in the math literature, and seeing how it holds up under ethics we'll be able to see what math and ethics have in common. And my guess is that math and ethics can be both shown to be close to science when it comes to epistemology. And that the differences in the way they have been considered have to do with how hard/easy studying the subject is (math is more objective cuz we're able to isolate variables, ethics is REALLY hard cuz there are so many variables, math is a priori because it's an essential part of the web of beleif, ethics barely seems like knowledge because it's so hard to get secure on it, etc.).
Without a doubt there is something that is quite different about ethics, science and math. Pinning down exactly what is really the challenge, and the next challenge is trying to figure out what those differences should mean--why the differences matter.
My general hypothesis is that hard, strong, epistemological lines tend to break down under stress. For example, Quine convinced most of the philosophical world that there is no strong, philosophically useful distinction between the meaning of words in a sentence and the substance of those sentences; there's no strong dichotomy between analytic sentences and synthetic ones. The argument comes down to just applying a great deal of stress to the dichotomy, and watching it collapse under investigation. I expect, coming into a philosophical investigation, a similar thing to happen when we've set up boundaries between realms of knowledge. Facts/values, empirical/non-empirical, a priori/a posteriori, history/science, objective/subjective and of course science/math/ethics are all ways that people have tried to divide up realms of knowledge. The problem is not dividing up realms of knowledge itself--everyone does that because there are real differences between the areas--but when it comes to epistemology, how we know what we know about these realms, I expect the dichotomies to break down. I expect that the way we gain historical knowledge isn't very different from the way we gain scientific knowledge. I expect that the way we gain mathematical knowledge isn't so different from the way we gain scientific or ethical knowledge. That's my hypothesis, in general. It's heavily influenced by Quine and Putnam (and Dewey, through Putnam), I think. At the very least, it's influenced by a misinterpretation of Quine and Putnam.
So, in particular, I expect that the distinctions between how we gain ethical knowledge and how we gain scientific knowledge break down under pressure. I expect that all knowledge is more or less in the same boat. I suspect that the differences are all of quantity, and not quality of knowledge. So I suspect that any attempt to explain why scientific stuff is objective and ethical or mathematical stuff is not eventually breaks down. I think that the indispensability argument goes a good way towards showing how the distinction breaks down between math and science, and I suspect that something similar can be adapted for ethics. I think that there are at least two good, promising ways of doing this, but in the end every way of making the argument does it by trying to make science a little bit more modest.
But I started by saying that there are real differences between math and ethics and science? It's just a fact that there are no ethics laboratories in universities, and it doesn't strike us as a very good idea to start such labs. Why is that? An analogy from math is useful in clarifying the question. Once Quine/Putnam argue that math is actually empirical knowledge and as objectively known as science, they need to answer the question: what fooled people for so long? Why did people think that math was a priori and divorced from experience? So, if you argue that ethics is on the same par as science, you have to explain why ethics strikes people as being the sort of thing that it doesn't make sense to start a lab for.
Of course, you could say that people are wrong, and that we really should be building ethics labs of some sort. Some people--I think that I heard this in the name of Nagel or Parfit--think that ethics is just a young science, one that's bound to develop the way that other sciences have. So maybe these people think that opening ethics labs makes sense. But I'm more sympathetic to the position that there is something about ethics that makes such a notion strange. I guess this could be consistent with the view that we should open labs, but I guess the view I find most attractive is that ethics is really really really hard. (This is also the reason why I'm not swayed by an argument from disagreement that ethics is subjective. Disagreement is consistent not only with subjective views, but also with really really really hard ones.) I bet that I could even show that some things that eventually fell into the realm of science were considered subjective problems, ones that it wouldn't make sense for a lab to study. For example, I bet a lot of brain stuff fits this pattern.
How does this relate to the original program of contrasting math with ethics? Well, the idea is that by taking an argument that's found in the math literature, and seeing how it holds up under ethics we'll be able to see what math and ethics have in common. And my guess is that math and ethics can be both shown to be close to science when it comes to epistemology. And that the differences in the way they have been considered have to do with how hard/easy studying the subject is (math is more objective cuz we're able to isolate variables, ethics is REALLY hard cuz there are so many variables, math is a priori because it's an essential part of the web of beleif, ethics barely seems like knowledge because it's so hard to get secure on it, etc.).
Ethics and Observation, Harman
(I'll explain the cat picture soon enough.)
I'm having great difficulty trying to pin down the difference between the role of observation in ethics and science that Harman describes (in "The Nature of Morality"). Not sure why, but I'm just unable to state the difference between ethics and science with regards to observation in any clear way. Anyway, here's my attempt to formulate it. Hopefully this will help me get closer to understanding it.
Harman's thesis: Observational evidence plays a role in science that it doesn't play in ethics. Specifically, observations can provide evidence for scientific theories, but observations can't provide evidence for ethical theories. Ethics fails to meet the standards of science, then.
So how does observation work in science? Harman begins by telling you how observation doesn't work in science. You might think that science works like this: you, the scientist, observe a new species of animal: looks kinda like a cat, kinda like a horse. It might be tempting to think that you're getting an usullied picture of the world when you make this observation. Like you're just downloading a bunch of data into your brain. But that wouldn't be quite true. Philosophers and psychologists know that this isn't the way that people perceive the world; there's a lot of your preexisting beliefs that go into your perceptions. Put another way, how you think about the world has a lot to do with how you see the world. For example, you need to know what a "box" is before you could possible perceive a box. For another example, suppose you experienced something totally unlike anything that you had experienced before. Would you be able to describe it? So perception is more like receiving processed data than it is receiving straight data. All of our data gets processed in the process of observation.
So does that mean that we should be skeptical of our observations? Why should we think that there is anything behind our observations, if our mind and preexisting beliefs color the way that we look at the world?
The answer, for Harman, is that we have good reason to believe that our observations are true because of inference to the best explanation. What is the best explanation of your observation? And by that I mean, what's the best explanation of the fact that you had the observation that you had? Well, let's list some of the possible explanations of the fact that you had the observation that you had.
(a) You were hallucinating, causing you to have the observation of something that seemed real.
(b) Your theory, your preexisting beliefs, colored the way that you observed the world. What you really saw was something that didn't have the kind of animal you observed, but you interpreted it in that way because of your theory and beliefs.
(c) You actually saw something in the world that looked the way you thought it did.
The best explanation is the third one. So in order to explain the fact that you observed something, we need to infer that you actually did observe something. It's inference to the best explanation.
Now, clearly this is right, but I'm not sure if it all adds up the way I'm describing it. What makes (c) the best explanation? Is it the simplest? What does simple mean? Does simple mean only one sentence long? Is it simpler to assume that you were hallucinating or simpler to assume that you actually saw a new species? By simpler, do we just mean "more likely to happen to a person"? So we assume that people see real stuff all the time, and from that we reason that the best explanation of a phenomenon is that you actually saw something? But that's gonna end up being a bit circular, because what we're interested in knowing is what justifies the thought that we're not hallucinating during observation.
Let me move on to ethics. Ethics, Harman says, is quite unlike science when it comes to observation. So, having told us how science works, we should be able to see that ethics doesn't work that way. Let's give it a shot.
So, Harman discusses the example of an ethical obseravtion. You're walking down the street, and you see a bunch of kids burning a cat (his example, not mine!). You immediately come to the conclusion "It's wrong to burn a cat." Now, you didn't necessarily believe this before you saw it. It might be that life never afforded you the opportunity to consider the case of a cat lynching. So we can call this a full-fledged observation of an ethical fact. Of course, of course, your pre-existing beliefs about what's right and wrong factor into your observation, but that doesn't matter because (as Harman argued above) the case is the same for any observation, including a scientific one. Every obseration is processed through the brain's machinary before coming to your consciousness, whether it's an ethical observation or a scientific one.
Now, in science we said that we have reason to believe that scientific facts are true because they are necessary for providing the best explanation for the fact that you had the observation that you did. Now, let's try this for ethics. What is the best explanation for the fact that you had the ethical obseration that you did, that you observed that it's wrong to burn cats? Here are a couple options:
(a) You actually did perceived something in the world that appears the way that you observed it, that is, you actually managed to perceive/see ethical properties in the world, the same way people observe that a ball is blue or that a tree is tall.
(b) You were "hallucinating." Your experience was that of an observation about something real in the world, but actually it was your brain and beliefs doing all the work.
Around here is where I get stuck. The idea is supposed to be that the best explanation of the fact that you had an ethical observation is (b). And so inference to the best explanation doesn't require any ethical facts to exist. But isn't this just to assume what what we were trying to prove? Oy. Need to get back to this.
Sunday, October 4, 2009
Can meat eaters who think they're doing something wrong teach us about metaethics?
I dunno. There are definitely people in the world who think that it's wrong to eat animals but do it anyway. Myself, I come close to it. Louis CK thinks the same thing, and if you don't mind the profanity here's a funny youtube video explaining the point.
OK. So some people are convinced that it's wrong to eat animals, but don't do it. How can this help us?
Some arguments against the objectivity of ethics go like this: if we had objective knowledge about ethics, it would be knowledge about a realm unlike anything else that we have knowledge of. Indeed, objective ethical knowledge would have to be about a realm of ethical objects that share the feature of being able to bring about obligations just through knowing them. That's unlike the rest of our knowledge: I don't immediately get any obligation just from knowing that "That tree is green." So, this knowledge of ethics would have to be unlike any other knowledge. And from here a small jump is taken--we have no secure reason to believe that such knowledge is possible or exists at all, and no reason to think that there are any objects such that knowledge of them leads to any sort of obligation.
So, there are two responses I can think of. The first I mentioned in the previous post, and it's the idea that there are actually lots of objects such that knowledge of or about them leads to obligations of some sort (the obligation to believe certain things, I pointed to in the post).
Another sort of response relates to vegetarianism, and people who think that it's wrong to eat animals but eat animals anyway. If you could get enough examples like this, you could begin to wear away at the notion that knowledge of ethical objects necessarily involves obligations--meaning, if you could know for certain, and securely, that it's wrong to eat animals while still not feeling obligated to stop eating 'em then this would be difficult for the notion that there is something intrinsically queer or weird about ethical knowledge. The weird thing, the obligation, has to come from somewhere else. But you could start to wear away at the strangeness of ethical knowledge, and through this you could resist the argument that ethical knowledge would have to be knowledge about a mysterious, curious and queer domain.
Update: As I think about this, a good analogy would be "I see the evidence, but I can't bring myself to believe it." In both cases reasons are given and recognized, and it's conceded that a rational person would probably act differently, but one doesn't believe the thing that he should, or doesn't do what he ought to. I guess, then, my point is complementary to the previous post. Together, I argue that the situation in ethics isn't so different from truth in general, and there are two responses one can make to the argument from queerness about ethics. One is that oughtness, or ought-to-ness is present in discussions of truth in general just as much as it is in ethics. So it's not queer unless truth is. But the other response is that reasonable people can know stuff about ethics/truth without acting/believing, so that knowledge isn't really knowledge of something that is essentially normative.
OK. So some people are convinced that it's wrong to eat animals, but don't do it. How can this help us?
Some arguments against the objectivity of ethics go like this: if we had objective knowledge about ethics, it would be knowledge about a realm unlike anything else that we have knowledge of. Indeed, objective ethical knowledge would have to be about a realm of ethical objects that share the feature of being able to bring about obligations just through knowing them. That's unlike the rest of our knowledge: I don't immediately get any obligation just from knowing that "That tree is green." So, this knowledge of ethics would have to be unlike any other knowledge. And from here a small jump is taken--we have no secure reason to believe that such knowledge is possible or exists at all, and no reason to think that there are any objects such that knowledge of them leads to any sort of obligation.
So, there are two responses I can think of. The first I mentioned in the previous post, and it's the idea that there are actually lots of objects such that knowledge of or about them leads to obligations of some sort (the obligation to believe certain things, I pointed to in the post).
Another sort of response relates to vegetarianism, and people who think that it's wrong to eat animals but eat animals anyway. If you could get enough examples like this, you could begin to wear away at the notion that knowledge of ethical objects necessarily involves obligations--meaning, if you could know for certain, and securely, that it's wrong to eat animals while still not feeling obligated to stop eating 'em then this would be difficult for the notion that there is something intrinsically queer or weird about ethical knowledge. The weird thing, the obligation, has to come from somewhere else. But you could start to wear away at the strangeness of ethical knowledge, and through this you could resist the argument that ethical knowledge would have to be knowledge about a mysterious, curious and queer domain.
Update: As I think about this, a good analogy would be "I see the evidence, but I can't bring myself to believe it." In both cases reasons are given and recognized, and it's conceded that a rational person would probably act differently, but one doesn't believe the thing that he should, or doesn't do what he ought to. I guess, then, my point is complementary to the previous post. Together, I argue that the situation in ethics isn't so different from truth in general, and there are two responses one can make to the argument from queerness about ethics. One is that oughtness, or ought-to-ness is present in discussions of truth in general just as much as it is in ethics. So it's not queer unless truth is. But the other response is that reasonable people can know stuff about ethics/truth without acting/believing, so that knowledge isn't really knowledge of something that is essentially normative.
You can't derive an "ought" from an "is"
Here's something that we hear all the time: you can't derive an "ought" from an "is." What's an example of this? You can't derive any obligations or requirements from any fact. So, you can't derive an obligation to do something about poverty from the statement "That person is poor." Just from believing that "That tree is causing trouble" you can't derive any ought statement, such as "We ought to cut down that tree" or "We ought not to cut down that tree."
I don't know much about this area of philosophy, but it seems to me that there's a fairly simple counter-example to this. The fact that it seems simple to me is evidence that it's probably wrong, but I'm not sure where the fault is. So if anybody can figure it out, let me know.
So, "truth" has a certain normativity, ought-ness about it, no? If you think that p is true, then you ought to believe p. For example, imagine encountering a person who thought that it was true that the world was round, but believed that the world was flat. What would you be able to say to such a person? You would say: "you don't really believe that it's true that the world is round. Because if you believe that the world is round, you OUGHT to believe that the world isn't flat, that it's round."
So, isn't there a simple "ought" that you can derive from an "is"? Namely, from "I think that p is true" you can derive that "I ought to believe p."
If anyone can tell me where this reasoning fails, I would definitely appreciate it.
I don't know much about this area of philosophy, but it seems to me that there's a fairly simple counter-example to this. The fact that it seems simple to me is evidence that it's probably wrong, but I'm not sure where the fault is. So if anybody can figure it out, let me know.
So, "truth" has a certain normativity, ought-ness about it, no? If you think that p is true, then you ought to believe p. For example, imagine encountering a person who thought that it was true that the world was round, but believed that the world was flat. What would you be able to say to such a person? You would say: "you don't really believe that it's true that the world is round. Because if you believe that the world is round, you OUGHT to believe that the world isn't flat, that it's round."
So, isn't there a simple "ought" that you can derive from an "is"? Namely, from "I think that p is true" you can derive that "I ought to believe p."
If anyone can tell me where this reasoning fails, I would definitely appreciate it.
Thursday, October 1, 2009
"is true" and "ought to"
Here's something I believe to be correct, but I'm not positive.
I took a course on Frege last year with Prof Goldfarb, and we discussed Frege's views on the predicate "is true." This predicate/property seems to be unlike any other property. Usually, when a predicate (concept) is tagged onto some subject (object) there is information conveyed. For example, "The table" might be any color in the world--red, blue, murple, etc. When I say, "The table is blue" I've really added some information that wasn't there before. Now, consider the property "is true" added on to some sentence. For example, consider the sentence "The table is blue". That sentence seems to tell me what color the table is. Now, apply the predicate "is true", and you get " 'The table is blue' is true." This seems to tell me that "The table is blue" is true. But that doesn't really add any information to the first sentence! Of course, if I asserted "The table is blue" seriously, I meant for you to think that it was true. So what does "is true" add? Seemingly nothing.
This is why Frege writes (in "Thought"), "The reference of the world 'true' seems to be altogether sui generis. May we not be dealing here with something which cannot be called a property in the ordinary sense at all? ... I will begin by expressing myself in accordance with ordinary usage, as if truth were a property until some more appropriate way of speaking is found." It sure sounds like Frege is saying that "is true" isn't a normal property. (Goldfarb told us that this point is controversial, but I can only see one side of it in Frege.)
So, if "is true" isn't a real property--in the sense that it's unlike every other property because it doesn't add any information, doesn't ascribe anything to the object or subject of the sentence--then what is its function in a sentence? As far as I can tell, "is true" basically serves to clarify that you are truly asserting the sentence. Allow me to clarify. We can all imagine contexts in which we could utter the words "The table is blue" but not mean for them to be taken seriously. For example, we could just be playing around with words. We could be describing a fiction, a way that the table could be but actually isn't. In all these cases we don't mean to state a fact or to make an objective assertion, we don't mean to say something that can be true. So, my sense is that "is true" serves to clarify that you're truly asserting a sentence with the intent for it to be taken objectivley and seriously. You want it to be objective--either true or false.
Now, let me shift gears for a minute. Colyvan, in his book in the Indispensability argument, is careful to distinguish between two different arguments. The first of them has as its conclusion "We ought to believe that numbers exist." The second of them has as its conclusion "Numbers exist." He thinks that the first one is better than the second, because the second is more controversial. What's the difference? The first is a statement about what we should believe, but it's careful to avoid talking about the world. Why? Well, because you might doubt the strength of your evidence. But then are you really justified in beliving that numbers exist in the first place? I consider this to be an instance of something that Michael Resnick mentions in his article on this: there's a kind of incoherence in saying "I'm justified in believing X, but I don't think X is true." Either you have evidence--and you're justified--or you don't. Yes, of course, there are levels of certainty, and everyting is revisable. But if you're justified in believing X--whatever it takes to make you justified--then certainly you should believe that numbers exist.
I submit that there is no difference between the conclusion of an argument that reads "Numbers exist" and one that concludes "We ought to believe that numbers exist". This is because "We ought" just means that you're making a normative statement, that you're asserting something normatively. But if the sentence "numbers exist" comes at the end of some argument, then we KNOW that you're asserting the sentence normatively, and there isn't any difference between them. This is parallel to the picture I described above for "is true." "I ought" or "you ought" or "one ought" just indicates that a normative statement has been made, completely parallel to the funciton of "is true." One more pass at it: to say "I ought to believe in p and not p" is just as irrational as to say "p is true and not p." (I was a little bit sloppy here. There's a bit of room for doubt, because we often say "I should believe that X, but I just can't bring myself to believe it. But that just means that you're not able to be rational for whateve reason. What's at issue is whether it would be rational to say "I ought but don't.")
If this picture if right, then certainly there's no point in differentiating between the two types of arguments. One is just a wussier version of the other.
Why would "We ought" and "is true" have similar functions? Likely because normativity is built into objective claims. When I make an assertion about the world I am saying "You ought to believe that the world is this way." So truth might be a kind of normativity. That's pure speculation.
I took a course on Frege last year with Prof Goldfarb, and we discussed Frege's views on the predicate "is true." This predicate/property seems to be unlike any other property. Usually, when a predicate (concept) is tagged onto some subject (object) there is information conveyed. For example, "The table" might be any color in the world--red, blue, murple, etc. When I say, "The table is blue" I've really added some information that wasn't there before. Now, consider the property "is true" added on to some sentence. For example, consider the sentence "The table is blue". That sentence seems to tell me what color the table is. Now, apply the predicate "is true", and you get " 'The table is blue' is true." This seems to tell me that "The table is blue" is true. But that doesn't really add any information to the first sentence! Of course, if I asserted "The table is blue" seriously, I meant for you to think that it was true. So what does "is true" add? Seemingly nothing.
This is why Frege writes (in "Thought"), "The reference of the world 'true' seems to be altogether sui generis. May we not be dealing here with something which cannot be called a property in the ordinary sense at all? ... I will begin by expressing myself in accordance with ordinary usage, as if truth were a property until some more appropriate way of speaking is found." It sure sounds like Frege is saying that "is true" isn't a normal property. (Goldfarb told us that this point is controversial, but I can only see one side of it in Frege.)
So, if "is true" isn't a real property--in the sense that it's unlike every other property because it doesn't add any information, doesn't ascribe anything to the object or subject of the sentence--then what is its function in a sentence? As far as I can tell, "is true" basically serves to clarify that you are truly asserting the sentence. Allow me to clarify. We can all imagine contexts in which we could utter the words "The table is blue" but not mean for them to be taken seriously. For example, we could just be playing around with words. We could be describing a fiction, a way that the table could be but actually isn't. In all these cases we don't mean to state a fact or to make an objective assertion, we don't mean to say something that can be true. So, my sense is that "is true" serves to clarify that you're truly asserting a sentence with the intent for it to be taken objectivley and seriously. You want it to be objective--either true or false.
Now, let me shift gears for a minute. Colyvan, in his book in the Indispensability argument, is careful to distinguish between two different arguments. The first of them has as its conclusion "We ought to believe that numbers exist." The second of them has as its conclusion "Numbers exist." He thinks that the first one is better than the second, because the second is more controversial. What's the difference? The first is a statement about what we should believe, but it's careful to avoid talking about the world. Why? Well, because you might doubt the strength of your evidence. But then are you really justified in beliving that numbers exist in the first place? I consider this to be an instance of something that Michael Resnick mentions in his article on this: there's a kind of incoherence in saying "I'm justified in believing X, but I don't think X is true." Either you have evidence--and you're justified--or you don't. Yes, of course, there are levels of certainty, and everyting is revisable. But if you're justified in believing X--whatever it takes to make you justified--then certainly you should believe that numbers exist.
I submit that there is no difference between the conclusion of an argument that reads "Numbers exist" and one that concludes "We ought to believe that numbers exist". This is because "We ought" just means that you're making a normative statement, that you're asserting something normatively. But if the sentence "numbers exist" comes at the end of some argument, then we KNOW that you're asserting the sentence normatively, and there isn't any difference between them. This is parallel to the picture I described above for "is true." "I ought" or "you ought" or "one ought" just indicates that a normative statement has been made, completely parallel to the funciton of "is true." One more pass at it: to say "I ought to believe in p and not p" is just as irrational as to say "p is true and not p." (I was a little bit sloppy here. There's a bit of room for doubt, because we often say "I should believe that X, but I just can't bring myself to believe it. But that just means that you're not able to be rational for whateve reason. What's at issue is whether it would be rational to say "I ought but don't.")
If this picture if right, then certainly there's no point in differentiating between the two types of arguments. One is just a wussier version of the other.
Why would "We ought" and "is true" have similar functions? Likely because normativity is built into objective claims. When I make an assertion about the world I am saying "You ought to believe that the world is this way." So truth might be a kind of normativity. That's pure speculation.
Subscribe to:
Posts (Atom)