Please help me with this apparent paradox: Does set of all naturals exist in {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...} ?

Status
Not open for further replies.
I've been talking around an ambiguity in your language, trying to clarify what you are referring to. Now I need to be very specific.

When you say "the set of sets" I have presumed (hoped?) that you meant, as I said here, one particular set of sets:


You never said that assumption was wrong. Since you initially asked about this particular set of sets, our answers have been about that. We've taken you at your word.

But here I think you are not talking about that particular set, but about the set of all sets. Or else you are confusing the two, and thinking that they are the same. I say that because it is utter nonsense to say "Every n that exists in [this particular] set of sets must be in some set," but it would be true (with a caveat) to say that "Every n that exists in the set of all sets must be in some set." (In either case, I'm taking your "in" to mean "an element of one of the sets constituting the big set", and not "an element of the big set"; the set's elements are sets, not numbers!)

So, please answer me directly: Are you thinking that {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...} is the set of all sets?? Or are you not really talking about this particular set of sets in the first place, but something else?

You talked from the start about the "bijection between each set and each natural number", though, as I said, that is irrelevant to the question as you asked it. Yes, there is a bijection between the set of natural numbers and {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}; but when we write {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}, we are talking about that set exactly as written, not about any set in one-to-one correspondence with it. For example, we don't say that the set {A, B, C} is the same set as {1, 2, 3}. I'm wondering if you are making that mistake, and thinking of {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...} as "the set containing any set with cardinality 1, and any set with cardinality 2, and so on -- which would be the set of all (non-empty, finite) sets.

But even this would not be the set of all sets, because it would still omit infinite sets!

Before we can solve your problem, we have to agree on exactly what you mean. That's part of the reason I emphasized definitions. You defined a particular set, but you don't seem to be talking about that set at all. Definitions and notation are taken very seriously in math; they are intended to mean exactly what they say, and no more. If you say one thing but mean another, then communication is impossible.
I am only talking about the set in the OP, {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}. Let's call it S for less confusion.

Here is the best way that I can describe why I think there is a contradiction.

Every n (natural number) that exists in S must be in some subset. If all n are elements of the subsets in S, then there can only be one of these subsets with all n. It would be identical to the set of all natural numbers.
 
Last edited:
Every n (natural number) that exists in the set of sets
"Exist" as in "belongs to" or "is a member of" ? But natural numbers belong to elements of [imath]S[/imath], not its subsets.
If all n are elements of the subsets in S, then there can only be one of these subsets.
Don't understand why you think so.

It sounds to me that you have some picture of elementary set theory in your head, and rely on it for your intuition, which leads you to some strange conclusions. You might look at some axioms and theorems of the set theory, do some exercises and learn to use formal arguments, not just what you think makes sense.
 
"Exist" as in "belongs to" or "is a member of" ? But natural numbers belong to elements of [imath]S[/imath], not its subsets.

Don't understand why you think so.

It sounds to me that you have some picture of elementary set theory in your head, and rely on it for your intuition, which leads you to some strange conclusions. You might look at some axioms and theorems of the set theory, do some exercises and learn to use formal arguments, not just what you think makes sense.
I gave you my reasoning but you pointed out some mistakes that I made while trying to give the reasoning. To correct it, by subsets of S, I meant elements of S.

Do you see anything wrong with my reasoning. If so, please explain.
 
"Exist" as in "belongs to" or "is a member of" ? But natural numbers belong to elements of [imath]S[/imath], not its subsets.

Don't understand why you think so.

It sounds to me that you have some picture of elementary set theory in your head, and rely on it for your intuition, which leads you to some strange conclusions. You might look at some axioms and theorems of the set theory, do some exercises and learn to use formal arguments, not just what you think makes sense.
I have to tweak my reasoning to make it more clear.

Every n (natural number) that exists in S must be in some subset. If all n are elements of the sets in S, then there can only be one of these sets with all n. It would be identical to the set of all natural numbers.
 
I have to tweak my reasoning to make it more clear.

Every n (natural number) that exists in S must be in some subset. If all n are elements of the sets in S, then there can only be one of these sets with all n. It would be identical to the set of all natural numbers.
First, your term "exists in S" is not standard. Clearly you don't mean "is an element of S", because the elements of S are sets, not numbers. Evidently you mean "is an element of (at least one) subset of S". Is that correct? (Please stop using the term "exists in S"!)

Second, every n is in infinitely many of the elements of S. For example, 5 is an element of (1,2,3,4,5}, and of {1,2,3,4,5,6}, and of {1,2,3,4,5,6,7}, and so on.

Third, none of the elements of S contain all natural numbers n. They are all, by definition, finite; each contains only the natural numbers up through some particular n. That's what the notation means: {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}.

Again, on what grounds do you make that last statement? A statement "there can only be ..." requires a reason, not just a declaration.
 
I have to tweak my reasoning to make it more clear.

Every n (natural number) that exists in S must be in some subset. If all n are elements of the sets in S, then there can only be one of these sets with all n. It would be identical to the set of all natural numbers.

You tweaking did not make it any clearer for me. I can only repeat my previous answer:
If all n are elements of the sets in S, then there can only be one of these sets with all n.

I don't believe this is true, and I don't see any formal reasoning proving this statement (which, BTW, cannot be proven since it is not correct).
 
Again, on what grounds do you make that last statement? A statement "there can only be ..." requires a reason, not just a declaration.
I did try to explain the reasoning in my paragraph you quoted. I will correct the errors and make it clearer.

If every n (natural number) that exists must be in some subset in S, then all n must have its own set. It would be a set identical to the set of all natural numbers.
 
If every n (natural number) that exists must be in some subset in S, then all n must have its own set. It would be a set identical to the set of all natural numbers.
No. Why must they all be in the same set? This is nonsense.

You need to learn about quantifiers. Each natural number is in at least one set, but not all need be in any one set. You are misusing words like "some", "all", and "its own", and that leads to bad thinking.

But even that level of logic is not needed. None of these sets contain all natural numbers, by definition! Again, when you defined the set as {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}, you were saying that every set that is an element of the set is finite! There is no way to get around that.
 
No. Why must they all be in the same set? This is nonsense.

You need to learn about quantifiers. Each natural number is in at least one set, but not all need be in any one set. You are misusing words like "some", "all", and "its own", and that leads to bad thinking.
How can all the natural numbers be in the subsets without all the natural numbers finally being in only one set?

In other words, imagine for a moment that the set of all natural numbers were in S. We would have something that looks like {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ... {1, 2, 3, 4, 5, 6, 7, ...}}. It is only in this last set does it seem possible to me for all natural numbers to be in all sets in S.

But even that level of logic is not needed. None of these sets contain all natural numbers, by definition! Again, when you defined the set as {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}, you were saying that every set that is an element of the set is finite! There is no way to get around that.
I am here because I see that definition contradicting (indirectly) another definition (or at least the implication of another definition). Your side of the argument is what you put here, and I am trying to show you what I see as a contradiction.
 
You tweaking did not make it any clearer for me. I can only repeat my previous answer:


I don't believe this is true, and I don't see any formal reasoning proving this statement (which, BTW, cannot be proven since it is not correct).
The way we defined S is that the number of natural numbers equals that same number of sets that are in S. And each set increases by 1 from the previous set starting at the set {1}. If we only use 5 natural numbers then we would know that there must be a set with all 5 natural numbers from 1 to 5, which looks like {1, 2, 3, 4, 5}. We know this because of the definition/rules we made for what S contains. Now if we have all infinite natural numbers in S, then I would think that the same be true in that there must be a set with all infinite natural numbers in S.
 
The way we defined S is that the number of natural numbers equals that same number of sets that are in S. And each set increases by 1 from the previous set starting at the set {1}. If we only use 5 natural numbers then we would know that there must be a set with all 5 natural numbers from 1 to 5, which looks like {1, 2, 3, 4, 5}. We know this because of the definition/rules we made for what S contains. Now if we have all infinite natural numbers in S, then I would think that the same be true in that there must be a set with all infinite natural numbers in S.
There are significant differences between finite and infinite sets. E.g., a finite set cannot be "equal" to its proper subset, but an infinite set can. Do you know any examples of such infinite set/subset equivalence?
 
There are significant differences between finite and infinite sets. E.g., a finite set cannot be "equal" to its proper subset, but an infinite set can. Do you know any examples of such infinite set/subset equivalence?
This is not true as phrased. If [imath] A\subsetneq B [/imath] then they are never equal, finite, or infinite. [imath] A [/imath] and [imath] B [/imath] can have the same cardinality ("number" of elements), but a proper subset cannot be equal to the bigger set, otherwise, it isn't called "proper".
 
There are significant differences between finite and infinite sets. E.g., a finite set cannot be "equal" to its proper subset, but an infinite set can. Do you know any examples of such infinite set/subset equivalence?
Yes, something like the set of all natural numbers and the set of all natural numbers divisible by 3. Why?
 
Yes, something like the set of all natural numbers and the set of all natural numbers divisible by 3. Why?
Exactly! I consider this a good illustration of the differences between finite and infinite sets. In your post #30 you jump to a conclusion about infinite sets based on the properties of finite sets.
 
Exactly! I consider this a good illustration of the differences between finite and infinite sets. In your post #30 you jump to a conclusion about infinite sets based on the properties of finite sets.
A set of the kind in the OP is either infinite or finite. Each finite number of sets is corresponded with an equal finite number of elements. There only seems to be one option for infinite, and that would be to correspond to infinite elements.

We know that infinite does not make sense to correspond to finite and vice versa.

I don't understand why it can't be like this. It seems to make more sense at least to me.
 
The way we defined S is that the number of natural numbers equals that same number of sets that are in S. And each set increases by 1 from the previous set starting at the set {1}. If we only use 5 natural numbers then we would know that there must be a set with all 5 natural numbers from 1 to 5, which looks like {1, 2, 3, 4, 5}. We know this because of the definition/rules we made for what S contains. Now if we have all infinite natural numbers in S, then I would think that the same be true in that there must be a set with all infinite natural numbers in S.
You defined S as the set containing all sets consisting of consecutive natural numbers from 1 through some natural number n. That's the meaning of your notation. If you list the contents of this set in order, the nth set ends with the number n.

That means that every set in this set ends somewhere! None of them is infinite. None of them can contain all natural numbers. As I said long ago, there is no "infinitieth element" in S. None of these sets "ends" with infinity; as I've said, infinity means that it doesn't end!

In the same way, the set of natural numbers does not include infinity; infinity is not a natural number.

Yes, there is one set in S corresponding to each natural number (which is its cardinality); there is no set corresponding to infinity: none of them contains all the natural numbers.

Of course, you could define a set that does include [imath]\mathbb{N}[/imath] (the set of all natural numbers) as one of its elements; but you would have to say so in your definition. You didn't. You'd need to write [imath]\{\{1\}, \{1, 2\}, \{1, 2, 3\}, \{1, 2, 3, 4\}, ...\}\cup \mathbb{N}[/imath].

You are still not taking your actual definition seriously. And you can't take definitions lightly when it comes to infinite things, because they just can't be expected to behave as you expect. What you mean by "think" is not thinking logically, but feeling based on experience, and you have no experience with anything actually infinite!

A set of the kind in the OP is either infinite or finite. Each finite number of sets is corresponded with an equal finite number of elements. There only seems to be one option for infinite, and that would be to correspond to infinite elements.

We know that infinite does not make sense to correspond to finite and vice versa.

I don't understand why it can't be like this. It seems to make more sense at least to me.
Your set S is clearly infinite; it contains a set corresponding to each natural number, and there are infinitely many of those.

But each of those sets that are elements of S is finite. As I've said, that's what your definition clearly says. S is a set containing infinitely many finite sets.

What you say here is very confusing, in part just because of poor grammar. "Each finite number of sets" - do you mean any finite set of the finite sets in S, or what? What "elements" are you referring to? I just can't make sense of this.
 
You defined S as the set containing all sets consisting of consecutive natural numbers from 1 through some natural number n. That's the meaning of your notation. If you list the contents of this set in order, the nth set ends with the number n.
Here is maybe the heart of my confusion. You say, "the nth set ends with the number n". Why does n ever have to end on the nth set? I keep mentioning this like a broken record, but I will say it again anyways. If we are talking about all n, why does n ever have to stop?

That means that every set in this set ends somewhere! None of them is infinite. None of them can contain all natural numbers. As I said long ago, there is no "infinitieth element" in S. None of these sets "ends" with infinity; as I've said, infinity means that it doesn't end!
Yes, I understand that.

In the same way, the set of natural numbers does not include infinity; infinity is not a natural number.
I know this too.

You are still not taking your actual definition seriously. And you can't take definitions lightly when it comes to infinite things, because they just can't be expected to behave as you expect. What you mean by "think" is not thinking logically, but feeling based on experience, and you have no experience with anything actually infinite!
If it were only that easy, except my brain keeps telling me that my definition and other mathematical definitions are clashing. I am very interesting in either clearing this up for myself or pursuing it further.

Your set S is clearly infinite; it contains a set corresponding to each natural number, and there are infinitely many of those.
I have a question regarding notation. What is the notation for S? I imagine it is something like this but I am not sure: {{1}, {1, 2}, ... {1, 2, 3, ... n}, {1, 2, 3, ... n ...}...}

This might help me look at this issue a different way.
 
Here is maybe the heart of my confusion. You say, "the nth set ends with the number n". Why does n ever have to end on the nth set? I keep mentioning this like a broken record, but I will say it again anyways. If we are talking about all n, why does n ever have to stop?
Do you not see the difference between "the nth set ends with the number n" and "n ever [has] to end on the nth set"? Words and grammar matter, which is why I keep returning you to definitions.

The definition of S deals with each set in S individually. The nth set listed is {1, 2, 3, ..., n}. That set ends.

I said nothing about n ending! Each set is defined for one value of n. There are an infinite set of sets, one for each natural number n.
  • For a given set {1, 2, 3, ..., n}, n is a fixed number; you can't talk about it ending.
  • For all of S, n increases without bound; it never ends.
But you appear to be confusing these two ideas.

If it were only that easy, except my brain keeps telling me that my definition and other mathematical definitions are clashing. I am very interesting in either clearing this up for myself or pursuing it further.
In order to understand mathematical statements, you must learn to focus on exactly what is said, and not imagine relationships that are not mentioned. (You gave one definition for that set; there is no other definition! You are the master of what that set is.)

I have a question regarding notation. What is the notation for S? I imagine it is something like this but I am not sure: {{1}, {1, 2}, ... {1, 2, 3, ... n}, {1, 2, 3, ... n ...}...}

This might help me look at this issue a different way.
When I talk about "the notation for S", I am talking about what you said at the very start:
I wanted to know if the set of natural numbers exists in the following set.

[S =] {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4} ...}
That notation means what I have described: a set containing finite sets. That is your definition. It determines what we are talking about.

Now you are talking about imagining a different notation:

{{1}, {1, 2}, ... {1, 2, 3, ... n}, {1, 2, 3, ... n ...}...}​

You are probably right that this explains your confusion. This does not mean the same thing! Here you have added a rather nonsensical "set", {1, 2, 3, ... n ...}; this notation means nothing, because n appears just to be one element in the middle of an infinite set, and in itself plays no role. That set would just be {1, 2, 3, ...}, the set of all natural numbers. More important, you then follow this with another ellipsis (...), as if there is more beyond that "set". There can't be.

To put it another way, you really can't use n in such a definition of S, because n is not defined anywhere. It would normally be some fixed (but unknown) number; for example, if n=5, then your "set" would be {{1}, {1, 2}, ... {1, 2, 3, ... 5}, {1, 2, 3, ... 5 ...}...}, which would be just {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4}, {1, 2, 3, 4, 5}, {1, 2, 3, ... 5 ...}...}, and, again, that last part would just be one more set, the set of all integers.

So the notation you are imagining is really bad notation, and it does not mean what you wrote in the first place.

Part of the problem is that this sort of notation with ellipses is a little ambiguous. We have other notations to clarify such ideas. Your set might be defined in set-builder notation as

[imath]S = \{\{1, 2, ..., n\} | n\in\mathbb{N}\}[/imath]​

or, for even more clarity (since the reader might be confused by "{1, 2, ..., 1}" when n=1, which just means {1}),

[imath]S = \{\{k\in\mathbb{N}\ |\ 1\le k\le n\}\ |\ n\in\mathbb{N}\}[/imath]​

In this form, n has a defined meaning, for each element (set) in S. We aren't using ellipses to imply a pattern.

This is why I say that you must take the details of whatever you write seriously -- because we will take you at your word! Mathematicians are very careful to write exactly what they mean, and to read exactly what is written. If you don't really mean what you write, then you are just going to have endless arguments about things no one agrees on.
 
Do you not see the difference between "the nth set ends with the number n" and "n ever [has] to end on the nth set"? Words and grammar matter, which is why I keep returning you to definitions.

The definition of S deals with each set in S individually. The nth set listed is {1, 2, 3, ..., n}. That set ends.

I said nothing about n ending! Each set is defined for one value of n. There are an infinite set of sets, one for each natural number n.
  • For a given set {1, 2, 3, ..., n}, n is a fixed number; you can't talk about it ending.
  • For all of S, n increases without bound; it never ends.
But you appear to be confusing these two ideas.


In order to understand mathematical statements, you must learn to focus on exactly what is said, and not imagine relationships that are not mentioned. (You gave one definition for that set; there is no other definition! You are the master of what that set is.)


When I talk about "the notation for S", I am talking about what you said at the very start:

That notation means what I have described: a set containing finite sets. That is your definition. It determines what we are talking about.

Now you are talking about imagining a different notation:

{{1}, {1, 2}, ... {1, 2, 3, ... n}, {1, 2, 3, ... n ...}...}​

You are probably right that this explains your confusion. This does not mean the same thing! Here you have added a rather nonsensical "set", {1, 2, 3, ... n ...}; this notation means nothing, because n appears just to be one element in the middle of an infinite set, and in itself plays no role. That set would just be {1, 2, 3, ...}, the set of all natural numbers. More important, you then follow this with another ellipsis (...), as if there is more beyond that "set". There can't be.

To put it another way, you really can't use n in such a definition of S, because n is not defined anywhere. It would normally be some fixed (but unknown) number; for example, if n=5, then your "set" would be {{1}, {1, 2}, ... {1, 2, 3, ... 5}, {1, 2, 3, ... 5 ...}...}, which would be just {{1}, {1, 2}, {1, 2, 3}, {1, 2, 3, 4}, {1, 2, 3, 4, 5}, {1, 2, 3, ... 5 ...}...}, and, again, that last part would just be one more set, the set of all integers.

So the notation you are imagining is really bad notation, and it does not mean what you wrote in the first place.

Part of the problem is that this sort of notation with ellipses is a little ambiguous. We have other notations to clarify such ideas. Your set might be defined in set-builder notation as

[imath]S = \{\{1, 2, ..., n\} | n\in\mathbb{N}\}[/imath]​

or, for even more clarity (since the reader might be confused by "{1, 2, ..., 1}" when n=1, which just means {1}),

[imath]S = \{\{k\in\mathbb{N}\ |\ 1\le k\le n\}\ |\ n\in\mathbb{N}\}[/imath]​

In this form, n has a defined meaning, for each element (set) in S. We aren't using ellipses to imply a pattern.

This is why I say that you must take the details of whatever you write seriously -- because we will take you at your word! Mathematicians are very careful to write exactly what they mean, and to read exactly what is written. If you don't really mean what you write, then you are just going to have endless arguments about things no one agrees on.
Okay, I learnt some new things. Thanks!

However, let's go back to my issue about set S not containing the set of all natural numbers (N). If there is no set for all natural numbers, then doesn't that mean that there must be natural numbers without a set?

I know that I defined S to have a set for every natural number, but this seems to cause contradictory implications with other definitions of natural numbers, infinity, etc.
 
Last edited:
Status
Not open for further replies.
Top