An abstract object is an object that does not correspond to any pattern of matter and energy in space-time. Purported examples of abstract objects are numbers, properties, sets, etc. An object that does correspond to some concentration of matter/energy in space-time is called a concrete object.
Using my recent attempt at (partially) tabooing “exists” to translate:
Nominalism: We can’t rationally care about abstract objects.
Platonism: We can rationally care about abstract objects.
So far Platonism appears to be “winning” according to this definition since UDT is Platonist in this sense, and there isn’t really a “nominalist decision theory” that’s equivalent or seems as promising.
But what about possibilist versions of Platonism as in “Abstract objects are ones which possibly exist”?
It seems quite rational to care about things which might happen, or which might exist without conceding that they actually will happen or actually do exist.
My inclination is to say that it doesn’t, and that the disagreement is really just about how to use the word “exist”. But there are a couple of ways in which the distinction might have a bearing on anticipated experience.
One prominent argument for Platonism is the Quine-Putnam Indispensability Argument, which says that our best strategy for ontological commitment is to believe in the existence of those objects over which our best scientific theories indispensably quantify. So if one cannot dispense with quantification over mathematical objects while maintaining the integrity of our best theories, then we should believe in the existence of those mathematical objects. If one accepts this criterion, then whether or not one believes in the existence of abstract objects depends on whether our best theories indispensably quantify over such objects, so the Platonism vs. nominalism debate depends on science. Hartry Field wrote a book where he tries to axiomatize Newtonian continuum mechanics without any quantification over real numbers.
Also, it’s plausible that Tegmark’s Level IV multiverse hypothesis assumes Platonism, since it requires that mathematical structures have independent existence. So if you believe Tegmark’s hypothesis constrains anticipation, then perhaps Platonism does as well.
Nope. The prerequisite is that it is uncertain whether it does or not pay rent. See Russell ’Problems of Philosophy” for a characterization of philosophy as “That which may eventually develop into science”…
That reminds me of my aunt saying “philosophy is that thing with which or without which the world would stay the same” (the Italian words for ‘which’ and ‘same’ rhyme).
I think a number of discoveries in psychology and neuroscience are relevant to the physicalism vs. anti-physicalism question.
I think relativity basically destroys the case for A-theory. The idea of an “objective present” loses all attraction (for me at least) when you realize that there is no such thing as objective simultaneity.
I think there’s plenty of evidence that God does not exist (and there is plenty of potential evidence that would convince me that He does).
There’s two things you could mean here. First, you could mean that the notion that their is no objective simultaneity removes your motivation to accept an A-theory. Second, you could mean that it makes an A-theory untenable. I take it you meant the first of these, but if not it might be worth checking out the second half of Craig Borne’s book “A Future for Presentism”.
In your examples, evidence from other disciplines has bearing on questions in philosophy. The problem is that information rarely flows the other way. All the philosophical debate on these (real) questions did not contribute significantly to our understanding. And then useful data came in from the relevant sciences and settled them, and would have done so even without the philosophical arguments in place.
Yeah, I was responding to your original claim that none of the questions here have any link to anticipated experience. Your claim here—that philosophy does not produce any knowledge of use to other disciplines—is a different criticism, and one that my comment was not intended to address. I think this criticism is also false, by the way. Well, it may be true in the sense that as a matter of fact very few people in other disciplines pay much attention to contemporary philosophy, but it is false that there is nothing of value in philosophy for these other disciplines.
Yes, this is definitely a confused question. “Correspondence” is complicated.
People shouldn’t use “exists” to mean “corresponds to some pattern of matter and energy” (so apparently I’m a Platonist); yet they also shouldn’t ignore the ontological distinction between numbers and atoms (so I guess I’m also kind of a nominalist).
I agree with you that the distinction here gets at what people mean by “exists,” but am not sure what makes that a false dichotomy. Personally, I lean towards nominalism precisely because “exist” isn’t a verb I would apply to abstract objects… whatever it is that existence consists of, abstract objects don’t do that thing.
a type of logical fallacy that involves a situation in which only two alternatives are considered, when in fact there is at least one additional option. The options may be a position that is between the two extremes (such as when there are shades of grey) or may be a completely different alternative.
For example, abstract objects could be considered to exist in the minds of people imagining them, and consequently in some neuronal pattern, which may or may not match between different individuals, but considered to not exist as something independent of the conscious minds imagining them. While this is a version of nominalism, it is not nearly as clear-cut as “abstract objects do not exist”.
For example, abstract objects could be considered to exist in the minds of people imagining them, and consequently in some neuronal pattern, which may or may not match between different individuals, but considered to not exist as something independent of the conscious minds imagining them. While this is a version of nominalism, it is not nearly as clear-cut as “abstract objects do not exist”.
That would be conceptualism and is a moderate anti-realist position about universals (if you’re a physicalist). Nominalism and Platonism are two poles of a continuum about realism of universals. So, you probably lean towards nominalism if you’re a physicalist and conceptualist.
I only used the word “exist” in a sentence because TheOtherDave and I agree on the meaning of it, which I doubt that any of the -ists you mention (probably including you) would agree with.
That seems to me to be pretty straightforward nominalism to me. I’m having a hard time imagining a more strict nominalist who would call your view an intermediate between his view and platonism.
I agree that if I don’t have a crisp understanding of what it means for X to exist (such that maybe X implemented as a neuronal pattern exists, and maybe it doesn’t, and no amount of data about the world could tell me which it is because I don’t know how states of the world map to the existence or nonexistence of X in the first place ) then I can’t clearly assert whether X exists or not.
For my own part, I’m fairly comfortable refusing to use “exists” to refer to that which abstract objects are doing by virtue of being represented by a particular neuronal pattern, and I’m consequently fairly comfortable identifying as a nominalist (for purposes of this question). If I were instead comfortable using “exists” to refer to that act, I would identify as a Platonist (fpotq). If I was comfortable doing both, or neither, I would choose “Other.”
Abstract objects exist when instantiated. The “form of 2” does not exist in a world of forms somewhere, but 2 billiard balls is an instance of both ‘2’ and ‘billiard ball’.
I agree this is an ‘Other’ doctrine, and ‘nominalism’ is a bad name for the doctrine of concretism (i.e., everything is spatiotemporal). But isn’t Aristotelianism a form of ‘nominalism’ as defined here? Doesn’t your Aristotelianism deny, or refuse to affirm, any “object that does not correspond to any pattern of matter and energy in space-time”? Properties are not ordinarily thought of as objects, and even so if you think that all properties are instantiated, then they do seem to be spatiotemporal, though perhaps in an informational rather than ‘matter/energy’ sense. Hm.
In the loose sense of ‘form’ (cf. Physics I.7), I think objects could: a rabbit has the form of a rabbit, and if it’s white it has that form too. That’s also consistant with the relationship between primary and secondary substances and non-substances from Categories.
On the other hand, in Metaphysics, ‘form’ is pretty much restricted to the identity of a substance. Still I think Aristotle would be happy enough with polymorphism so long as you made a distinction between substance-forms and non-substance-forms.
What do platonists say the word object means, if some objects are abstract? What is a property that is true of all objects? What are some categories of non-objects?
For most Platonists, an object is something we quantify over. So, for instance, numbers are objects, because we say things like “There is one even prime” (existential quantification), and “All multiples of 6 are also multiples of 2″ (universal quantification). Any domain over which we quantify is a domain of objects.
There are many different answers to this question. One common answer is that we should believe an object exists if a theory we accept quantifies over that object. If quantum field theory requires quantification over the integers then 2 (and other integers) exist. Since contemporary physics doesn’t quantify over a domain containing phlogiston, phlogiston does not exist.
There’s something wrong when smart people argue and disagree over a question when there are many different ideas as to what the words in the question actually mean...
Other: I hold a Tegmark-style “all mathematical objects exist” view, but additionally hold that this requires promoting the type of the “exists” property from boolean-valued to real-valued, so that this becomes “all mathematical objects have nonzero degree of existence”, and most ordinary usages of the word “exists” pick up an implied “as a substructure of” relation to some set of universes. This “as a substructure of” relation excludes platonic objects when present, so nominalism vs. platonism reduces to the word “exists” being a pair of homonyms.
I think there’s some truth in the claim that this is just a debate over the word “exists” but actually, in philosophy there’s a fairly well-regarded meaning of the word and the debate tends to be about this meaning. I think that’s one of the reasons comparisons of people from LW with philosophers will be misleading, philosophers use a lot of technical vocabulary that differs from the standard usage.
So by “exists” philosophers tend to mean something like “is quantified over in our fundamental theory of the universe”. So nominalists in mathematics tend to be arguing something like (and this is a dangerously rough approximate), “we can describe the universe without referring to numbers etc”. See Hartry Field for an example.
This is rough but actually, platonism begins to seem a lot more plausible when it’s understood in the way that philosophers actually use it. I suspect, then, that those from LW and philosophers are just answering different questions here.
I think I’m more comfortable with the negation of Platonism than with the positive claim of nominalism, but I suppose in this context we have ‘nominalism’ = ‘~Platonism’.
Whether what is usually meant by ‘nominalism’ is the same is as unclear to me as I am uncomfortable with the idea of making a positive claim about it.
Sorry for the very belated reply, but I was struggling to find the words to describe exactly what I meant. Luckily, Eliezer has already done most of it for me in his latest post.
Thing A exists with respect to Thing B iff Thing A and Thing B are both part of the same causal network. So ArisKatsaris was half-right, but things outside our past and future light cones can be said to exist with respect to us if they have a causal relationship with anything that is inside our past and future light cones.
Wrongbot’s idea is probably that something “exists” only in relation to something else. So things beyond the Earth’s past and future light-cones don’t actually “exist” for us, though they exist for themselves, and similarly things in a mathematical model of a different universe exist in relation to each other, but not in relation to us...
Other: Abstract objects clearly have some kind of logical existence, but it is very different from concrete existence. Also, I’m confused on this one. Also, the answer doesn’t matter, so who cares.
Tree-in-a-forest. I think that “exist” can have (at least) two different meanings and that abstract objects exist in one of those sense but not in the other.
Abstract objects: nominalism or Platonism?
[pollid:89]
An abstract object is an object that does not correspond to any pattern of matter and energy in space-time. Purported examples of abstract objects are numbers, properties, sets, etc. An object that does correspond to some concentration of matter/energy in space-time is called a concrete object.
Nominalism: Abstract objects do not exist.
Platonism: Abstract objects exist.
Still not sure what this means. Is there some sense in which this distinction pays rent in anticipated experience?
I voted other because of my confusion on this point. I think we need to taboo “exists”.
Using my recent attempt at (partially) tabooing “exists” to translate:
Nominalism: We can’t rationally care about abstract objects.
Platonism: We can rationally care about abstract objects.
So far Platonism appears to be “winning” according to this definition since UDT is Platonist in this sense, and there isn’t really a “nominalist decision theory” that’s equivalent or seems as promising.
That just shifts the ground to disagreeing about what is “rational” when arguing about different epistemologies.
Shifting the ground to an easier, more tractable problem? Awesome.
That seems a rather new argument for Platonism.
But what about possibilist versions of Platonism as in “Abstract objects are ones which possibly exist”? It seems quite rational to care about things which might happen, or which might exist without conceding that they actually will happen or actually do exist.
I voted ‘other’ to the original question. I would vote ‘accept platonism’ to this question.
My inclination is to say that it doesn’t, and that the disagreement is really just about how to use the word “exist”. But there are a couple of ways in which the distinction might have a bearing on anticipated experience.
One prominent argument for Platonism is the Quine-Putnam Indispensability Argument, which says that our best strategy for ontological commitment is to believe in the existence of those objects over which our best scientific theories indispensably quantify. So if one cannot dispense with quantification over mathematical objects while maintaining the integrity of our best theories, then we should believe in the existence of those mathematical objects. If one accepts this criterion, then whether or not one believes in the existence of abstract objects depends on whether our best theories indispensably quantify over such objects, so the Platonism vs. nominalism debate depends on science. Hartry Field wrote a book where he tries to axiomatize Newtonian continuum mechanics without any quantification over real numbers.
Also, it’s plausible that Tegmark’s Level IV multiverse hypothesis assumes Platonism, since it requires that mathematical structures have independent existence. So if you believe Tegmark’s hypothesis constrains anticipation, then perhaps Platonism does as well.
Ah, that sounds a bit like coordinate-free physics.
It’s a prerequisite that a question must not pay rent in anticipated experience before it can be part of philosophy.
Nope. The prerequisite is that it is uncertain whether it does or not pay rent. See Russell ’Problems of Philosophy” for a characterization of philosophy as “That which may eventually develop into science”…
That reminds me of my aunt saying “philosophy is that thing with which or without which the world would stay the same” (the Italian words for ‘which’ and ‘same’ rhyme).
As far as I can see, none of the questions in this survey have any relation to anticipated experience.
I think a number of discoveries in psychology and neuroscience are relevant to the physicalism vs. anti-physicalism question.
I think relativity basically destroys the case for A-theory. The idea of an “objective present” loses all attraction (for me at least) when you realize that there is no such thing as objective simultaneity.
I think there’s plenty of evidence that God does not exist (and there is plenty of potential evidence that would convince me that He does).
There’s two things you could mean here. First, you could mean that the notion that their is no objective simultaneity removes your motivation to accept an A-theory. Second, you could mean that it makes an A-theory untenable. I take it you meant the first of these, but if not it might be worth checking out the second half of Craig Borne’s book “A Future for Presentism”.
In your examples, evidence from other disciplines has bearing on questions in philosophy. The problem is that information rarely flows the other way. All the philosophical debate on these (real) questions did not contribute significantly to our understanding. And then useful data came in from the relevant sciences and settled them, and would have done so even without the philosophical arguments in place.
Yeah, I was responding to your original claim that none of the questions here have any link to anticipated experience. Your claim here—that philosophy does not produce any knowledge of use to other disciplines—is a different criticism, and one that my comment was not intended to address. I think this criticism is also false, by the way. Well, it may be true in the sense that as a matter of fact very few people in other disciplines pay much attention to contemporary philosophy, but it is false that there is nothing of value in philosophy for these other disciplines.
Another false dichotomy. The word exist means different things to different people.
Yes, this is definitely a confused question. “Correspondence” is complicated.
People shouldn’t use “exists” to mean “corresponds to some pattern of matter and energy” (so apparently I’m a Platonist); yet they also shouldn’t ignore the ontological distinction between numbers and atoms (so I guess I’m also kind of a nominalist).
Life is so much easier for an instrumentalist, for whom “exist” is quite clearly defined.
I agree with you that the distinction here gets at what people mean by “exists,” but am not sure what makes that a false dichotomy. Personally, I lean towards nominalism precisely because “exist” isn’t a verb I would apply to abstract objects… whatever it is that existence consists of, abstract objects don’t do that thing.
Wikipedia:
For example, abstract objects could be considered to exist in the minds of people imagining them, and consequently in some neuronal pattern, which may or may not match between different individuals, but considered to not exist as something independent of the conscious minds imagining them. While this is a version of nominalism, it is not nearly as clear-cut as “abstract objects do not exist”.
That would be conceptualism and is a moderate anti-realist position about universals (if you’re a physicalist). Nominalism and Platonism are two poles of a continuum about realism of universals. So, you probably lean towards nominalism if you’re a physicalist and conceptualist.
I only used the word “exist” in a sentence because TheOtherDave and I agree on the meaning of it, which I doubt that any of the -ists you mention (probably including you) would agree with.
And what meaning is that?
That seems to me to be pretty straightforward nominalism to me. I’m having a hard time imagining a more strict nominalist who would call your view an intermediate between his view and platonism.
OK, thanks for clarifying.
I agree that if I don’t have a crisp understanding of what it means for X to exist (such that maybe X implemented as a neuronal pattern exists, and maybe it doesn’t, and no amount of data about the world could tell me which it is because I don’t know how states of the world map to the existence or nonexistence of X in the first place ) then I can’t clearly assert whether X exists or not.
For my own part, I’m fairly comfortable refusing to use “exists” to refer to that which abstract objects are doing by virtue of being represented by a particular neuronal pattern, and I’m consequently fairly comfortable identifying as a nominalist (for purposes of this question). If I were instead comfortable using “exists” to refer to that act, I would identify as a Platonist (fpotq). If I was comfortable doing both, or neither, I would choose “Other.”
Other: Aristotelianism.
Abstract objects exist when instantiated. The “form of 2” does not exist in a world of forms somewhere, but 2 billiard balls is an instance of both ‘2’ and ‘billiard ball’.
I agree this is an ‘Other’ doctrine, and ‘nominalism’ is a bad name for the doctrine of concretism (i.e., everything is spatiotemporal). But isn’t Aristotelianism a form of ‘nominalism’ as defined here? Doesn’t your Aristotelianism deny, or refuse to affirm, any “object that does not correspond to any pattern of matter and energy in space-time”? Properties are not ordinarily thought of as objects, and even so if you think that all properties are instantiated, then they do seem to be spatiotemporal, though perhaps in an informational rather than ‘matter/energy’ sense. Hm.
I’m not sure if Aristotle actually believed an object could be an instance of two forms. That’s pretty advanced polymorphism.
In the loose sense of ‘form’ (cf. Physics I.7), I think objects could: a rabbit has the form of a rabbit, and if it’s white it has that form too. That’s also consistant with the relationship between primary and secondary substances and non-substances from Categories.
On the other hand, in Metaphysics, ‘form’ is pretty much restricted to the identity of a substance. Still I think Aristotle would be happy enough with polymorphism so long as you made a distinction between substance-forms and non-substance-forms.
Sounds about right to me.
What do platonists say the word object means, if some objects are abstract? What is a property that is true of all objects? What are some categories of non-objects?
For most Platonists, an object is something we quantify over. So, for instance, numbers are objects, because we say things like “There is one even prime” (existential quantification), and “All multiples of 6 are also multiples of 2″ (universal quantification). Any domain over which we quantify is a domain of objects.
Then what does it mean for an abstract object like 2 to “exist”?
There are many different answers to this question. One common answer is that we should believe an object exists if a theory we accept quantifies over that object. If quantum field theory requires quantification over the integers then 2 (and other integers) exist. Since contemporary physics doesn’t quantify over a domain containing phlogiston, phlogiston does not exist.
There’s something wrong when smart people argue and disagree over a question when there are many different ideas as to what the words in the question actually mean...
Other: I hold a Tegmark-style “all mathematical objects exist” view, but additionally hold that this requires promoting the type of the “exists” property from boolean-valued to real-valued, so that this becomes “all mathematical objects have nonzero degree of existence”, and most ordinary usages of the word “exists” pick up an implied “as a substructure of” relation to some set of universes. This “as a substructure of” relation excludes platonic objects when present, so nominalism vs. platonism reduces to the word “exists” being a pair of homonyms.
Augustinian!!!!!! hssss
I think there’s some truth in the claim that this is just a debate over the word “exists” but actually, in philosophy there’s a fairly well-regarded meaning of the word and the debate tends to be about this meaning. I think that’s one of the reasons comparisons of people from LW with philosophers will be misleading, philosophers use a lot of technical vocabulary that differs from the standard usage.
So by “exists” philosophers tend to mean something like “is quantified over in our fundamental theory of the universe”. So nominalists in mathematics tend to be arguing something like (and this is a dangerously rough approximate), “we can describe the universe without referring to numbers etc”. See Hartry Field for an example.
This is rough but actually, platonism begins to seem a lot more plausible when it’s understood in the way that philosophers actually use it. I suspect, then, that those from LW and philosophers are just answering different questions here.
I think I’m more comfortable with the negation of Platonism than with the positive claim of nominalism, but I suppose in this context we have ‘nominalism’ = ‘~Platonism’.
Whether what is usually meant by ‘nominalism’ is the same is as unclear to me as I am uncomfortable with the idea of making a positive claim about it.
Other: Existence is a two-valued function, not one-valued.
Could you elaborate, please?
Sorry for the very belated reply, but I was struggling to find the words to describe exactly what I meant. Luckily, Eliezer has already done most of it for me in his latest post.
Thing A exists with respect to Thing B iff Thing A and Thing B are both part of the same causal network. So ArisKatsaris was half-right, but things outside our past and future light cones can be said to exist with respect to us if they have a causal relationship with anything that is inside our past and future light cones.
Wrongbot’s idea is probably that something “exists” only in relation to something else. So things beyond the Earth’s past and future light-cones don’t actually “exist” for us, though they exist for themselves, and similarly things in a mathematical model of a different universe exist in relation to each other, but not in relation to us...
Other: Abstract objects clearly have some kind of logical existence, but it is very different from concrete existence. Also, I’m confused on this one. Also, the answer doesn’t matter, so who cares.
Basically Taboo exists
Tree-in-a-forest. I think that “exist” can have (at least) two different meanings and that abstract objects exist in one of those sense but not in the other.