MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Do you think that $\mathbb Z \subset \mathbb R$? On one hand this inclusion is quite handy. We like to write things like: $$ \sqrt{n} \quad \text{for $n\in \mathbb Z$} $$ which requires the number $n$ to be a real number (where $\sqrt\cdot$ is defined). On the other hand it is difficult to obtain such an inclusion when comes to definition. One would like to be able to define whole numbers without the need to define real numbers. This becomes more tricky when one notices that there are other inclusions which one would like to satisfy. For example I would like $1$ to be a polynomial with whole coefficients, or maybe a polynomial with complex coefficients, or maybe a real function of one variable...

I would say that it is not possible to satisfy all these inclusions. So maybe we must not insist on saying that $\mathbb Z \subset \mathbb R$ in the first place. Are there alternatives?

One possibility I see is that of having many different sets isomorphic to $\mathbb Z$. We should use the name $\mathbb Z$ for integers as we use $V$ for vector spaces. We should say: let $\mathbb Z$ be any set of integers. Or: let $\mathbb R$ be a set of reals and let$\mathbb Z$ be the set of real integers $\mathbb Z \subset \mathbb R$. And so on...

Another possibility I see (but I'm not sure if it can be really founded) is that of redefining the meaning of equality $=$ and distinguish between equality and identity. We could try to take an "object-oriented" approach where equality could be defined like any other operation. So one could define $1 = 1/1$ i.e. the integer $1$ is the same as the rational $1/1$ and one should define the sum of integers and rational by converting the integer to a rational and then performing the sum between rationals. This also modifies the concept of 'set' since the set $\{1, 1/1 \}$ is equal to $\{1/1\}$ and hence has a single element. This is, more or less, how types work in computer languages. Can this approach be made rigorous?

share|cite|improve this question

closed as off topic by Stefan Geschke, Eric Wofsey, Michael Greinecker, Dmitri Pavlov, Chandan Singh Dalawat Jan 28 '13 at 9:47

Questions on MathOverflow are expected to relate to research level mathematics within the scope defined by the community. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about reopening questions here.If this question can be reworded to fit the rules in the help center, please edit the question.

If you follow the approach of successors, you proceed from Natural numbers to integers to rational numbers and with a completion the reals. This inherently embeds integers in the reals. Alternatively, if you introduce the real numbers as a model for a certain set with some axiomatic structure, you can identify the previously mentioned subsets via their structural properties (which is more along the line of your comment about sets isomorphic to $\mathbb{Z}$. Is this in the direction you are thinking? – Daniel Spector Jan 28 '13 at 8:49
I don't understand why this is closed. Is the problem that it's too elementary? Or that it's been discussed elsewhere here? But then some comments along those lines would be nice (such as a suggestion to use math.SX or a link to another question). Instead it's closed with no useful explanation. Fortunately there were already a couple of answers worth voting up. – Toby Bartels Jan 28 '13 at 10:30
@Daniel: you speak of an embedding of $\mathbb N$ into $\mathbb R$ not of a set inclusion. This means you couldn't write $\sqrt n$ for $n \in \mathbb N$ without an abuse of notation. My question is whether it is possible to avoid the abuse of notation without having to much complications in writing formulas. Saying with other words: since we all understand what we are speaking about, why is it so difficult to find a notation to formalize it in a correct way (i.e. without abusing the notation)? This problem has been addressed in computer languages, where we don't admit abuses of notation. – Emanuele Paolini Jan 28 '13 at 11:34
Toby, I agree -- the question doesn't seem bad or MO-unworthy (although I'd get rid of the opening "Do you think..." which could suggest subjectivity). My hypothesis is that people in the math blogosphere have gotten a little tired of foundational discussions, of which there has been a recent spate. However, some explanation of closure would be nice. Should a meta thread be opened? – Todd Trimble Jan 28 '13 at 12:57
I don't know if this is the sort of thing you're looking for, but you might be interested in Chapter 7 of Mohan Ganesalingam's thesis: . (Which, it turns out, has now vanished from the web in anticipation of publication.) – HJRW Jan 28 '13 at 16:21
up vote 6 down vote accepted

We do in fact have many different sets isomophic to $\mathbb{Z}$. The correct definition of $\mathbb{Z}$ is a set with some operations on it that satisfy some axioms. This is easier to see with $\mathbb{N}$, which is a set with a 0, a "plus one" operations, that satisfies mathematical induction. There is only one such set up to oeration-preserving isomorphism.

$\mathbb{R}$ can be axiomatized similarly, as a complete Archimedean ordered field (see the synthetic approach section at Wikipedia). Again there is only one such set up to operation-preserving isomorphism. You can identify in that set a copy of $\mathbb{Z}$, so you can treat $\mathbb{Z}$ as a subset. But this is a matter of notation. If you had a compelling reason to make it disjoint, you could require that, instead.

We say "the integers" or "the reals" because there is only one, up to isomorphism. Any theorem you prove about $\mathbb{Z}$ or $\mathbb{R}$ that doesn't use the internal representation will transfer to any isomorphic copy, so we don't need to know which one.

There is a whole mathematical notion of types that actually underpins languages with more-complicated type systems, such as ML or Haskell. You can think of $\mathbb{Z}$ and $\mathbb{R}$ as a type in a type system. Some typing systems have a notion of "subtype", but let's suppose that you don't. Then $\mathbb{Z}$ and $\mathbb{R}$ are types, and there's a designated monomorphism $i$ from $\mathbb{Z}$ to $\mathbb{R}$. Then 2 + 3.5 is an overloaded operation that is syntactic sugar for $i(2) + 3.5$. I don't know if anyone has ever worked out a clear account of what mathematicians do from this point of view, but what they do is not very complicated.

If you do want to allow subtypes, there's a notion of order-sorted algebra that allows the kind of overloading you probably have in mind. I don't know of a canonical link, but introductions are easy to find.

share|cite|improve this answer
+1 for 'I don't know if anyone has ever worked out a clear account of what mathematicians do from this point of view, but what they do is not very complicated.' – HJRW Jan 28 '13 at 17:17
Also, iirc, Chapter 7 of Ganesalingam's thesis (mentioned in a comment above) is more or less an attempt to do exactly that. – HJRW Jan 28 '13 at 17:20
I accepted this answer. I never heard of types used in mathematics, but this is in the direction of what I was thinking... Can you present some references? Is the notion of types replacing set theory or using it? – Emanuele Paolini Jan 29 '13 at 13:19
There is a gigantic mathematical theory of types, that predates the invention of the computer, and has had considerable influence on the design of programming languages such as ML and Haskell. Some people would like to replace set theory with type theory. I think the idea of one replacing the other is pointless, since for most mathematical applications I can express anything I want to say equally well in either. – arsmath Jan 29 '13 at 14:23
The best introductions to type theory are more computational oriented than math oriented. I would recommend Girard's "Proofs and Types", which is available online. I don't know a good reference explicitly for the definition of the reals inside a type theory. The main consumers of type theory tend to be more interested in constructive mathematics than in a non-constructive topic such as classical analysis. Second-order arithmetic ( is like a type theory with two types, even though it is couched in set-theoretic language. – arsmath Jan 29 '13 at 14:28

The answer to this question depends on the type of set theory that you are working in, and the way you decide to code the integers and real numbers inside set theory. For instance in material set theory, we often (it is not mandatory) define the natural numbers to be finite Von Neumann ordinals,integers as certain classes of ordered pairs of naturals the rationals as certain equivalence classes of certain ordered pairs of integers, then the reals as classes of Cauchy sequences. In this kind of set theory, in no way is the integers a subset of the reals. But their is a cannonical inclusion of the integers into the reals.

In a structural set theory what is in the set does not matter so much as how different sets relate to one another. This is inherently a categorical discussion of what set theory is. Now in category theory, a sub-object is an equivelence class of monomorphisms. In this kind of set theory the integers are a subset of the reals.

For the differences between material and structural set theory see .

A subobject is defined here:

Also a nice exposition of structural set theory is here:

Also relevant to material and structural set theory and relations to type theory are here:

share|cite|improve this answer

Expanding on the first part of Daniel Spector's comment, in all the cases you mentioned there is a canonical inclusion $\mathbb{Z}\hookrightarrow X$ for a certain set $X$ (reals, polynomials with complex coefficients, etc. ). When we write $\sqrt{n}$ for $n\in\mathbb{Z}$ (or any analogous expression), you are doing a little abuse of notation, identifying $n$ with its image under the canonical embedding of $\mathbb{Z}$ into the reals (or the appropriate set $X$). This usually doesn't cause any ambiguity and that's why we do it all the time.

share|cite|improve this answer
Ok, so you are on the side where $\mathbb Z \subset \R$ is not assumed (or is considered an abuse of notation). – Emanuele Paolini Jan 28 '13 at 11:24
Yes, I don't feel the need of having a unique object in set theory that deserves to be called "the set of integers". The relevant thing is the canonical identification referred to above. But I'm not an expert in logic and set theory, so I could be missing some subtlety... – Gian Maria Dall'Ara Jan 28 '13 at 22:20

Not the answer you're looking for? Browse other questions tagged or ask your own question.