Sider on Intrinsicality
A few years ago, Sider argued against Lewis's second account of intrinsicality and in favor of Lewis's first account. In this post, I will do the following. First, I will present Lewis's two accounts of intrinsicality (I will assume the doctrine of world bound individuals in my presentation, but that is simply to make things simpler). Second, I will present Sider's objection to Lewis's second account. Third, I will suggest an amendment to Lewis's second account. And finally, I will present my own objections to Lewis's first account.
Lewis First Account of intrinsicality says that a property is intrinsic just in case it never differs between duplicates (where two things are duplicates just in case there is a one-one correspondence between their parts that preserves perfectly natural properties). Intuitively, if something looks just like me and has all the same perfectly natural properties that I have, then that thing has all the intrinsic properties that I have.
Lewis's second account starts with the notion of a lonely object. A lonely object is one that exists alone in a possible world. A property is independent of loneliness just in case (i) it is had by a lonely object, (ii) it is lacked by a lonely object, (iii) it is had by a non-lonely object, and (iv) it is lacked by a non-lonely object. Finally, Lewis's Second Account says that a property is intrinsic just in case it is both independent of loneliness and non-disjunctive.
Sider seems to favor Lewis first account over his second. Sider argues that there are counterexamples to the second account which are not also counterexamples to the first account. His counterexamples rely on the notion of maximality. A property F is maximal just in case (roughly) large parts of an F thing are not themselves Fs. Some refinements need to be made to this notion, but let's ignore those for now. According to Sider, being a rock is a maximal property because large parts of rocks are not themselves rocks. Similarly, being a house is a maximal property because large parts of houses are not themselves houses. Sider also claims that maximal properties are not intrinsic. This is because whether or not something has a maximal property will depend on whether it is a large part of something that has that maximal property. For example, something is not a house if it is a large part of a house; all of a house except one of its windows is not a house because that thing is a large part of a house. Since the exemplification of these maximal features by an object depends on what is going on outside the borders of that object, they are not intrinsic.
Now Sider's argument is as follows:
1. If Lewis's second account is true, then being a rock is intrinsic.
2. But, being a rock is not intrinsic.
3. So, Lewis's second account is not true.
The justification for (2) is in the paragraph above. The justification for (1) is as follows. Being a rock is not a disjunctive property. Moreover, being a rock is had by a lonely object and lacked by a lonely object (there are lonely rocks and lonely non-rocks). And also, being a rock is had by a non-lonely object and lacked by a non-lonely object (simple empirical investigation will verify this). So, according to Lewis's second account, being a rock is intrinsic. So, (1) is true.
Sider notes that a similar problem does not arise for the first account of intrinsicality. There could be a duplicate of a rock that is embedded in a larger rock. That duplicate is not a rock. So, bieng a rock is not shared by these two duplicates and hence is not intrinsic (according to the first account).
I like Sider's argument. I guess I think it is sound. However, I think a simple amendment to the second account will save it from this sort of objection. I wish I could say that a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) non-maximal. But, this is not quite right. Sider points out that it is not maximality that is troublesome but rather border sensitivity. Sider says that a property is border-sensitive iff "whether it is instantiated by an object depends on what is going on, intrinsically, outside that object at its border." All border sensitive properties are non-intrinsic and some are not maximal. This led me to think that the following account might be true: a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) non-border-sensitive. But, unfortunately, because of the occurrence of 'intrinsic' in the definition of 'border-sensitive' this account is circular. So, let's introduce a new notion. Let's say that a property is outwardly-sensitive just in case whether it is instantiated by an object depends on what is going on outside that object (at its border) (I'm not sure if this last little parenthetical bit is needed). If a property is outwardly-sensitive, then it is border-sensitive. Now we can say the following:
Lewis's Amended Second Account: a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) not outwardly-sensitive.
I kind of like this account. Question: are there any immediate problems with this account?
Now, I'd like to suggest that there are counter-examples to Lewis's first account which are not also counterexamples to Lewis's Amended Second Account. First, if Lewis's First Account is true, then being identical to Joshua is not an intrinsic property. This is because some duplicates of me are not identical to me. But, it is an intrinsic property. So, Lewis's First Account if false. Second, if Lewis First Account if true, then co-existing with the number 2 is an intrinsic property. This is because the number 2 is a necessary existent and no two duplicates fail to co-exist with it. But, co-existing with the number 2 is not intrinsic (at least it is not intrinsic to me). So, Lewis's First Account is false.
I like these two arguments, though, I do not think they are knock-down. David, for example, has suggested a problem for the first argument. I am not sure what I think of the problem that he suggests. But, it is interesting to note that the Amended Second Account is not so obviously subject to these counterexamples. First, as long as you do not endorse certain kinds of essentialism, we can maintain that being identical to Joshua is independent of loneliness. It is also clear that co-existing with the number 2 is not independent of loneliness.
There is a worry that I have, though. It seems that co-existing with the number 2 is an extrinsic property of me and an intrinsic property of the number 2 (thanks to Rock* for pointing out this worry). I am not sure how to amend the account further to get this result. I can make the following amendment. First, call W2 a subtraction of a world W1 iff W2 is either possible or impossible and it is just like the W1 except that something that exists in the W1 does not exist in W2. Now we can say that a lonely object* is one that exists alone in a possible world or alone in a subtraction of a possible world. We might be able to give a disjunctive account of intrinsicality where we use the notion of loneliness for contingent things and the notion of loneliness* for non-contingent things. I am not sure how this will work out. I'd prefer to come up with an account of F is intrinsic to x and wed the notion of intrinsicality to objects. I'm sorry this post has gotten a little messy toward the end, but I am still thinking through these things. If anyone has any ideas, please let me know what you think.
Lewis First Account of intrinsicality says that a property is intrinsic just in case it never differs between duplicates (where two things are duplicates just in case there is a one-one correspondence between their parts that preserves perfectly natural properties). Intuitively, if something looks just like me and has all the same perfectly natural properties that I have, then that thing has all the intrinsic properties that I have.
Lewis's second account starts with the notion of a lonely object. A lonely object is one that exists alone in a possible world. A property is independent of loneliness just in case (i) it is had by a lonely object, (ii) it is lacked by a lonely object, (iii) it is had by a non-lonely object, and (iv) it is lacked by a non-lonely object. Finally, Lewis's Second Account says that a property is intrinsic just in case it is both independent of loneliness and non-disjunctive.
Sider seems to favor Lewis first account over his second. Sider argues that there are counterexamples to the second account which are not also counterexamples to the first account. His counterexamples rely on the notion of maximality. A property F is maximal just in case (roughly) large parts of an F thing are not themselves Fs. Some refinements need to be made to this notion, but let's ignore those for now. According to Sider, being a rock is a maximal property because large parts of rocks are not themselves rocks. Similarly, being a house is a maximal property because large parts of houses are not themselves houses. Sider also claims that maximal properties are not intrinsic. This is because whether or not something has a maximal property will depend on whether it is a large part of something that has that maximal property. For example, something is not a house if it is a large part of a house; all of a house except one of its windows is not a house because that thing is a large part of a house. Since the exemplification of these maximal features by an object depends on what is going on outside the borders of that object, they are not intrinsic.
Now Sider's argument is as follows:
1. If Lewis's second account is true, then being a rock is intrinsic.
2. But, being a rock is not intrinsic.
3. So, Lewis's second account is not true.
The justification for (2) is in the paragraph above. The justification for (1) is as follows. Being a rock is not a disjunctive property. Moreover, being a rock is had by a lonely object and lacked by a lonely object (there are lonely rocks and lonely non-rocks). And also, being a rock is had by a non-lonely object and lacked by a non-lonely object (simple empirical investigation will verify this). So, according to Lewis's second account, being a rock is intrinsic. So, (1) is true.
Sider notes that a similar problem does not arise for the first account of intrinsicality. There could be a duplicate of a rock that is embedded in a larger rock. That duplicate is not a rock. So, bieng a rock is not shared by these two duplicates and hence is not intrinsic (according to the first account).
I like Sider's argument. I guess I think it is sound. However, I think a simple amendment to the second account will save it from this sort of objection. I wish I could say that a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) non-maximal. But, this is not quite right. Sider points out that it is not maximality that is troublesome but rather border sensitivity. Sider says that a property is border-sensitive iff "whether it is instantiated by an object depends on what is going on, intrinsically, outside that object at its border." All border sensitive properties are non-intrinsic and some are not maximal. This led me to think that the following account might be true: a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) non-border-sensitive. But, unfortunately, because of the occurrence of 'intrinsic' in the definition of 'border-sensitive' this account is circular. So, let's introduce a new notion. Let's say that a property is outwardly-sensitive just in case whether it is instantiated by an object depends on what is going on outside that object (at its border) (I'm not sure if this last little parenthetical bit is needed). If a property is outwardly-sensitive, then it is border-sensitive. Now we can say the following:
Lewis's Amended Second Account: a property F is intrinsic just in case it is (i) independent of loneliness, (ii) non-disjunctive and (iii) not outwardly-sensitive.
I kind of like this account. Question: are there any immediate problems with this account?
Now, I'd like to suggest that there are counter-examples to Lewis's first account which are not also counterexamples to Lewis's Amended Second Account. First, if Lewis's First Account is true, then being identical to Joshua is not an intrinsic property. This is because some duplicates of me are not identical to me. But, it is an intrinsic property. So, Lewis's First Account if false. Second, if Lewis First Account if true, then co-existing with the number 2 is an intrinsic property. This is because the number 2 is a necessary existent and no two duplicates fail to co-exist with it. But, co-existing with the number 2 is not intrinsic (at least it is not intrinsic to me). So, Lewis's First Account is false.
I like these two arguments, though, I do not think they are knock-down. David, for example, has suggested a problem for the first argument. I am not sure what I think of the problem that he suggests. But, it is interesting to note that the Amended Second Account is not so obviously subject to these counterexamples. First, as long as you do not endorse certain kinds of essentialism, we can maintain that being identical to Joshua is independent of loneliness. It is also clear that co-existing with the number 2 is not independent of loneliness.
There is a worry that I have, though. It seems that co-existing with the number 2 is an extrinsic property of me and an intrinsic property of the number 2 (thanks to Rock* for pointing out this worry). I am not sure how to amend the account further to get this result. I can make the following amendment. First, call W2 a subtraction of a world W1 iff W2 is either possible or impossible and it is just like the W1 except that something that exists in the W1 does not exist in W2. Now we can say that a lonely object* is one that exists alone in a possible world or alone in a subtraction of a possible world. We might be able to give a disjunctive account of intrinsicality where we use the notion of loneliness for contingent things and the notion of loneliness* for non-contingent things. I am not sure how this will work out. I'd prefer to come up with an account of F is intrinsic to x and wed the notion of intrinsicality to objects. I'm sorry this post has gotten a little messy toward the end, but I am still thinking through these things. If anyone has any ideas, please let me know what you think.
posted by Joshua at 11:56 AM
0 Comments:
Post a Comment
<< Home