This is the talk page for discussing improvements to the Natural number article. This is not a forum for general discussion of the subject of the article.
This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.MathematicsWikipedia:WikiProject MathematicsTemplate:WikiProject Mathematicsmathematics
This article is within the scope of WikiProject Numbers, a collaborative effort to improve the coverage of Numbers on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.NumbersWikipedia:WikiProject NumbersTemplate:WikiProject NumbersNumbers
Latest comment: 9 months ago20 comments7 people in discussion
re "[the integers] are made by including [0 and negative numbers].
[The rational numbers] add [fractions], and
[the real numbers] add [all infinite decimals].
[Complex numbers] add *[the square root of −1]*.",
Z = N∪{0}∪{x: x is a negative natural}
Q = Z∪{x: x is a fraction}
R = Q∪{x: x is an infinite decimal}
¬(C = R∪{√-1})
what I mean is that saying that the the complex numbers add i, is like saying that the integers add -1. That is kinda true for a reasonable interpretation of add, but it is a different interpretation than all other uses of add in this paragraph. C\R is not {i}; C is not just one element added to the reals, it is the closure of that set under multiplication/addition/division/etc.
I don't know if this necessarily needs changed, and I can't think of a better way to express it, but it bugs me that this paragraph is expressed inconsistently in this way. EktelestesPragmaton (talk) 17:58, 17 December 2025 (UTC)Reply
It still has some issues. I'd recommend removing it from the lead and adding material in §Generalizations on , , , , , . This could include both axiomatic and constructive approaches. The material should note that while there are natural embeddings , there is no unique embedding .
I'm not sure I understand what the dispute is here and I'm not incredibly interested in figuring it out. I just want to urge that this material not be expanded to try to solve whatever the problem is supposed to be. It seems reasonable to mention more expansive number systems and how they are built up from the naturals, but just "mention", not "go on and on about". The current level of detail seems fine.
I do have one picky objection, which is formulating the reals in terms of "infinite decimals", which is not at all the best way to think of them, in particular because it suggests there's something special about base 10. That said, I definitely don't want to introduce Dedekind cuts or Cauchy sequences in this article, because that would be "going on and on about" it. I guess I can live with the mention of decimals, but it does slightly rub me the wrong way. --Trovatore (talk) 21:47, 17 December 2025 (UTC)Reply
I wouldn't say there's a "dispute". I just rewrote this paragraph from:
Many number systems are built from the natural numbers and contain them. For example, the integers are made by including 0 and negative numbers. The rational numbers add fractions, and the real numbers add all infinite decimals. Complex numbers add the square root of −1. This makes up natural numbers as foundational for all mathematics.
to
The most common number systems used throughout mathematics are extensions of the natural numbers, and can be formally defined in terms of natural numbers. If the difference of every two natural numbers is considered to be a number, the result is the integers, which include zero and negative numbers. If the quotient of every two integers is considered to be a number, the result is the rational numbers, including fractions. If every infinite decimal is considered to be a number, the result is the real numbers. If every solution of a polynomial equation is considered to be a number, the result is the complex numbers. This makes natural numbers foundational for all mathematics.
with the goal of being more precise and less potentially confusing. Hopefully this isn't "going on and on about" this topic, but it could also plausibly be boiled down to a sentence or two and unpacked in a later section. –jacobolus(t)01:20, 18 December 2025 (UTC)Reply
Anything more advanced than "infinite decimals" is not going to be usable for an elementary audience in my opinion. Someone who clicks through to real number can hopefully learn as many details as they want. –jacobolus(t)01:17, 18 December 2025 (UTC)Reply
I do think the problem is by making it "more precise" you also made it much longer, 55 words to 113 words. I would say, revert to the original paragraph in the lead, or even something more concise, and then write a new thing in the generalizations section on the precise details of how the numeric tower is constructed, and what "adding" an element means in this context (it has a precise interpretation in terms of algebraic closure, subrings, etc.). Mathnerd314159 (talk) 03:13, 18 December 2025 (UTC)Reply
Well, the issue with the lead is that it must be written at a very approachable level, while also not telling lies to children (literal children, in this context, given the likely readership). So a wikilink is about the cleanest approach I can think of, explaining that the meaning of "embed" here is a lot more rigorous than its casual connotation. Its utility is the usual utility of a wikilink, if someone wants to learn more about what is meant then they can click on it and dive down the rabbit hole. Mathnerd314159 (talk) 17:27, 18 December 2025 (UTC)Reply
The problem is that "including 0 and negative numbers", "add fractions", "add all infinite decimals", and "add the square root of −1" are of different forms, quite confusing (readers will plausibly misunderstand what "add" means), and in the last example misleadingly incorrect if following the implied pattern of the first 3 examples.
The portion in the lead could probably be shortened to something like
The most common number systems used throughout mathematics – the integers, rational numbers, real numbers, and complex numbers – are extensions of the natural numbers, and can be formally defined in terms of natural numbers.
and possibly combined with a different paragraph. Something like my rewritten paragraph (or even a longer elaboration) could be included further down the page. –jacobolus(t)06:42, 18 December 2025 (UTC)Reply
This seems very convenient for the lead, eexcept that "are extensions" seems unnecessarily technical for the kead, and would better be replaced by "contain".
To editor Mathnerd314159: the reason of my revert is that "This chain of extensions canonically embeds the natural numbers in the other number systems" is is much too technical for the intended audience of this lead. Moreover, I am not sure whether "a chain of extensions that embeds te natural numbers" is defined anywhere. Also, in the linked article there are many different concepts of embeddings, and I have not found a definition that is directly applicable (without advanced mathematical knowledge) to number systems (indeed, an embedding is a structure preserving injection, and to embed the natural numbers in some structure, one must first define the structure of the natural numbers). D.Lazard (talk) 16:43, 19 December 2025 (UTC)Reply
To editor Jacobolus: Yeah, I agree that anything "more advanced" is out of scope for this article, and honestly I think maybe the mention of the reals is already out of scope. If we are going to mention the reals then I suppose we're stuck with "infinite decimals". It's a bit grating but I don't really have a better suggestion. Do we really want to mention the reals? I'm still a little on the fence about that. --Trovatore (talk) 04:46, 18 December 2025 (UTC)Reply
These number systems can also be formally defined in terms of natural numbers (though they need not be).
with a note:
The commonly-assumed set-theoretic containment may be obtained by constructing the reals, discarding any earlier constructions, and defining the other sets as subsets of the final construction
Does this risk giving a misleading impression? My initial reading was that it was saying there are ways to define Q or R (for eg) in a general way without starting with N? TheGrifter80 (talk) 04:52, 8 February 2026 (UTC)Reply
It seems clear to me. Is there another way to read it?
just to give a possible example, you could define the integers as the free monoid generated by the two element set {1,-1} quotiented by the equality 1+-1=-1+1=0, then define the rationals, reals, complex numbers, etc. from the integers the way you normally would and you never needed to use the naturals. don't know if that is help in answering your question. EktelestesPragmaton (talk) 12:06, 10 February 2026 (UTC)Reply
Thanks for that example, that does help. I'm not too familiar with abstract algebra, but interested to understand how 1 and -1 are defined in the system. I will look around for an introduction, thanks again. TheGrifter80 (talk) 00:17, 11 February 2026 (UTC)Reply
Latest comment: 9 months ago7 comments3 people in discussion
Looking for some input on the structure of this article, particularly order of the sections and possibilities for new sections. Three suggestions to start:
1. Move the History section to the bottom of the article. I think this article would be best starting off with simple concepts, then gradually introducing more technical detail, but the History doesn't really fit into that flow, so suggest moving to the end.
2. New section on "representations of numbers" (as suggested by jacobolus last year). Including numerals and numeral systems, probably following "Intuitive concept" section.
3. Section on philosophy of natural numbers. Would only be of interest to a small minority of readers so it would be right down the bottom of the article either as short stand-alone section or part of History.
I'd recommend starting any major project by looking for sources.
I think this article would best start with some summary of elementary arithmetic of natural numbers. It would also be good to discuss divisibility and primes pretty near the top (but not starting with the current "The algebraic structure is a commutativemonoid with identity element0. It is a free monoid on one generator. This commutative monoid satisfies the cancellation property, so it can be embedded in a group. The smallest group containing the natural numbers is the integers. [...] Analogously, given that addition has been defined, a multiplication operator can be defined via a × 0 = 0 and a × S(b) = (a × b) + a. This turns into a free commutative monoid with identity element1; a generator set for this monoid is the set of prime numbers." which is ridiculously over-technical for the likely audience for this page.) –jacobolus(t)01:58, 11 February 2026 (UTC)Reply
put history between after formal statement but before generalizations, as in the theorems/conjectures section
put "Terminology" after "Intuitive concept", as in "programming constructs"
put discussion of representations after history (as in "Algorithms and data structures", "classic problems", etc. which put solution/implementation after history)
philosophy: would go into intuitive concept as far as being easily explainable, otherwise at end
put history between after formal statement but before generalizations,
Agree, good suggestion.
put "Terminology" after "Intuitive concept"
I can see the case either way for which of these first should come first.
philosophy: would go into intuitive concept as far as being easily explainable, otherwise at end
Agree. I previously had some of this material in intuitive concept but it was too dense as it was written for the start of the article. I will re-add at the end, and look for opinions if any can/should be moved into "Intuitive concept". TheGrifter80 (talk) 03:03, 13 February 2026 (UTC)Reply
I think I'd somewhat pare down the history section here, keep it toward the bottom, and direct people toward Number#History instead (or we could even have a separate article History of numbers (currently a redirect); It doesn't seem that important to prominently feature a history about the concept of 'natural number' per se here. Instead I think we should focus first on the internal structure of natural numbers, second on the relation of natural numbers to other types of numbers, and third on the applications of natural numbers (in greater depth than now). –jacobolus(t)03:11, 13 February 2026 (UTC)Reply
On the History section, I agree it should be pared down. I suggest we get rid of the "ancient roots" subsection (or summarise it very briefly), as this is just a general history of numbers. We could keep a small section of history as relates directly to what we now call "natural numbers", which is a fairly recent distinction and term, so could be of some interest.
As for applications of natural numbers, I agree and thought another section on arithmetic (at a less technical level) would be useful as you suggested above. Currently some of this is covered in "Properties" but in a formal way. Perhaps also move the paragraph on counting out of Intuitive concepts into this new section. TheGrifter80 (talk) 03:35, 13 February 2026 (UTC)Reply
Latest comment: 8 months ago12 comments5 people in discussion
you set theorist editors want to present set theory infinity like a paradise garden so explain it very verbosely but avoid stating things that are an embarrassment or misstate things so they are more rosy. Sometimes you present your own material as if it's the material of the professional and original set theorists. These are obviously contrary to Wikipedia policy. Also over verbosely presenting set theory infinity but also avoiding other things, as though there's not enough room, is obviously wrong. You do that because then set theory won't be rosy. And you present concepts that can be explained simply but present them more confusingly, presumably since people could figure out the errors of set theory if explained simply.
The present edit the editor tried to come up with his own concept of cardinality, not that of the cantorists, such that it applies to finite and infinite, and equal and not equal, and is simple, one paragraph. They don't use that because it's mathematically invalid. The natural numbers are injective bijective and surjective with the natural numbers, so claiming the natural numbers are not equal cardinality with the natural numbers because things are left over, pairing the natural numbers with their doubles, is invalid.
You would have to say if you cannot prove equal that proves not equal. Obviously invalid but that IS how set theory infinity works. But that is not what the set theorists say because that leaves out greater and they want proof not observation of not able to prove equal.
What the original and current literature says is that using the concept:
That which applies to to finite by proof also applies to the equivalent infinite without proof
But only in some cases, other cases it's 'does not apply'. This is not proof but presumption and since they cannot apply it universally presumably false. So they apply it to infinite (thus also finite) ≤ and ≥ and =, but don't allow for < or >. Likewise claim 1 + infinity equals infinity contrary to the principal, but cardinal exponentiation follows the principal.
Then they use the diagonal argument to prove their proof of (they use =, your using ≤) won't work. But then use lack of proof is proof of opposite (≠). So then using their proof of ≥ plus that to get, by a definition, >.
Likewise with ordinal size they use can prove 1 + infinity equals infinity (that's never ending infinity not as one of you claims infinity with end, in that case the proof doesn't work) , but can't prove infinity plus one equals the same infinity so claim adding a member (the set of ordered natural numbers) to the (nonexistant presumably) end of an endLess set (the natural numbers) produces an infinity with an end and larger, that is lack of proof of same length is proof of greater length.
Now in original set theory size cardinal size and ordinal size are precisely defined, not by proof or axiom but definitions. Now obviously the above (cardinality ) is not the same as actual size but quite different, and the way I explain it it is obviously erroneous. Thus the original set theorists back to cantor nor present the most professional set theorists don't claim cardinality is the same as size, nor only technically not size. Further repeating explaining cardinality instead of linking to the cardinality article is wrong
The same editor produced both edits, recently and October 8
It's not clear what you're trying to say. If you have specific suggestions I'm happy to look at them - there might be something useful to add. TheGrifter80 (talk) 14:29, 3 March 2026 (UTC)Reply
I was clear
You don't want to believe it so your mind doesn't accept it
I explained to you size is not cardinality yet you again put in an edit claiming that. The set theorists did not mistakenly use definitions for Cardinality instead of axioms for Size. Each of the original set theorists including Cantor claimed size is not cardinality and in YouTube video comments set theorists correct that size is not cardinality. Your obviously violating Wikipedia policy. I explained in detail. You also decided to invent your own set theory concept to define cardinality as if you understand better than they do. No. Your paragraph doesn't work since for example the natural numbers are not ONLY in bijection with the natural numbers but also surjection and injection, so injection does not prove >.
Instead they prove bijection cannot be proved to supposedly prove ≠ THEN use injection to prove not > but they claim ≥. Inventing your own set theory is original research and against Wikipedia policy as well as deceiving
Language models are surprisingly good at understanding even otherwise impenetrable comments, here is the list of concrete suggestions the LLM produced:
"Size of a collection" section: Explicitly distinguish "size" from "cardinality" - clarify that cardinality is a specific mathematical definition of "size" based on one-to-one correspondence (bijection), and there are other definitions of size (e.g. ordinals can be considered sizes) and also that mathematical notions of size may not correspond to our intuitive notion of size. I think some of this is in the cardinality article, but I guess the treatment here is too abbreviated.
"Position in a sequence" - so the section is right in presenting the idea that an ordinal number is whatever "comes after" a given set/sequence. But it is wrong in that in implying that all ordinal numbers are natural numbers, or that ordinal numbers behave like natural numbers - the simple example is non-commutativity of ordinal addition, 1+ω=ω but ω+1 is itself. So we need to distinguish "using natural numbers to order" (ordinal numerals) from the formal definition of "ordinal number" (which includes natural numbers but also infinite ordinals).
Over-focus of article on set theory - Victor seems to want a dedicated subsection on Finitism or Intuitionism, and how some mathematicians (like Leopold Kronecker or L.E.J. Brouwer) rejected the idea of "completed" infinities, to present a more nuanced view of the infinite nature of the natural numbers than simple "the natural numbers are a subset of the cardinals/ordinals".
Comparing between Viktor's comment and this list of suggestions, I am not sure this list will make Viktor completely satisfied, but they do seem like reasonable suggestions. Mathnerd314159 (talk) 21:43, 3 March 2026 (UTC)Reply
there are other definitions of size Cardinality gives the same comparisons as the "how many" meaning of size.
e.g. ordinals can be considered sizesOrdinal numbers carry more information than just size, e.g., and are distinct ordinals but imply the same cardinality.
But it is wrong in that in implying that all ordinal numbers are natural numbers, or that ordinal numbers behave like natural numbers Where does it imply that?
If you're thinking of A natural number used to describe a position in a sequence is called an ordinal number., I agree that the wording could be clearer.
'If two collections do not have the same cardinality, pairing will leave one of the collections with objects that are unpaired and this can be used to define a size relationship between them. The collection in which all objects are paired is said to be "smaller" and the one left with unpaired objects "larger", than the other.'
This is what you edited in.
Ignore the first sentence since you cannot say if two collections do not have the same cardinality then they do not have (thus have greater) same cardinality.
Pairing CAN leave one of the collections with unpaired objects even if it's the same collection, pairing the natural numbers with the even natural numbers leaves the odd natural numbers unpaired.
You would have to say instead if it is impossible to pair without leaving unpaired, but that's saying if you cannot prove = that proves ≠ thus >. But lack of proof is not proof of it's opposite. If you believe that your wrong. And you wouldn't have a single definition for = and ≠ thus >. Simply if one can pair all of a set with only some of a set then the latter set is proven larger using the same sort of incorrect reasoning as your proof of equal size. But then for example the set of evens would be proven larger than the set of primes. So set theory DOESNT believe that simplicity.
Now why does Wikipedia very verbosely describe set theory contrary to Wikipedia policy, but not point out their reasoning flaw, assuming that which applies by proof to the finite, perfect one to one matching proving same size, applies to the same infinite without proof but rather presumption, one to one matching of infinite sets proves same size, but only SOMETIMES assuming that which apply to the finite by proof applies to the same infinite. that is for, as per the early literature, ≤ and ≥ and they say by proof only = but said proof obviously needs = defined and the ≤ and ≥ parts are also definitions (of cardinality), so correcting them, for ≤ = and ≥ but forbidding for < or > since then all sets in Aleph0 would be < and = and > each other in infinite size, likewise but separately all sets in Alreh1. That's not math itself ('crisis in mathematics') or the foundations of all math as Wikipedia claims but an individual manipulated system of math.
Likewise to prove > they take their proof of ≥ but need proof of ≠. Thus the diagonal argument. But it ONLY proves their proof of = won't work, for example between the naturals and the reals from 0 inclusive to 1 exclusive. So they assume lack of proof of = is proof of ≠.
But no the lack of countability is from the naturals are atomic and their version of the reals is nonatomic, being instead fluid. So an atomic set cannot be atomicly paired with a nonatomic set.
Likewise for ordinal 1+∞=same ∞ (omega) but ∞+1≠same ∞ so they say is ordinal omega +1. The first is adding to the beginning of by definition beginning but never ending list of (correcting them since allowing to continue without end would obviously continue to any defined ∞s) only finite numbers, the second adding to the end of said never-ending that is without end list. They assume since they cannot prove = size that means ≠ thus larger size, but no rather cannot add to non-existent end of list so cannot prove by matching a list to a nonlist.
Now
If Wikipedia is so verbose about set theory
Why not include all of it
Why not include the reasons for what is said in Wikipedia
Obviously because that would be embarrassing
Also
Why not explain things childishly simple
Obviously since then the errors would be figured out
But if you Wikipedia editors would rather just say 'these are the rules' then set theory is not math itself ('crisis in mathematics') not the foundations of math but a private game. Victor Kosko (talk) 02:06, 4 March 2026 (UTC)Reply
Two collections have the same cardinality if it is possible to pair their constituent elements in some way; as you say, for infinite sets it is also possible to make a pairing from one set of a particular cardinality to a subset of another set of the same cardinality (e.g. to a subset of itself), which is famously counterintuitive for those used to discussing finite sets (cf. Hilbert's paradox of the Grand Hotel); some even reject the existence of "completed infinity", cf. Actual and potential infinity, but this is a philosophical rather than mathematical question.
In the context of a finite collection whose cardinality is a natural number, these counterintuitive ideas are not really relevant, putting further discussion mostly out of scope for this article. We could maybe set aside one or two sentences for a discussion of the cardinality of the whole set of natural numbers (or we could even add an entire section about it toward the end of the article), linking interested readers to Cardinality or Countable set for more information. –jacobolus(t)03:59, 4 March 2026 (UTC)Reply
The third paragraph in §Size of a collection, If two collections do not have the same cardinality, pairing will leave one of the collections with objects that are unpaired and this can be used to define a size relationship between them. The collection in which all objects are paired is said to be "smaller" and the one left with unpaired objects "larger", than the other., is only valid for finite sets; a minor edit should fix that. It might be reasonable to briefly mention infinite cardinals and ordinals in §Generalizations, with a nod to the Axiom of Choice. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 14:12, 4 March 2026 (UTC)Reply
read the article. It goes into infinite set theory heavily to the extent that someone else deleted ALOT of material from the cardinality article.
The problem obviously it the editor is not qualified to edit set theory infinity material.
I want the set theory cardinality lies in one place and likewise ordinal, BUT as I said COMPLETE, that is with the embarrassing stuff. Your claiming the natural numbers article should leave out the set theory infinity stuff. Fine, except per the problems your having with the anti set theory people it should clarify in set theory there are no infinite natural numbers. Also another set theory error, you assume by defining the natural numbers the way you do there will automatically be no infinites in the set. Not so, it has to be an axiom, without any limit the sequence obviously will go to any defined infinity, there's nothing to stop. Victor Kosko (talk) 18:37, 4 March 2026 (UTC)Reply
"read the article. It goes into infinite set theory heavily" – Which article are you talking about? The section you are complaining about says nothing beyond "Taken together the natural numbers themselves form an infinite sequence", which doesn't really seem controversial. –jacobolus(t)19:17, 4 March 2026 (UTC)Reply
I think the edits you made address the issue fully.
I used the ordinary language word "collection" rather than "set" to avoid unhelpful technicalities in this section.
Latest comment: 7 months ago8 comments4 people in discussion
The sentence A sequence is a collection of objects with a specific order, meaning that every object has a position and comes either before or after every other object according to a given criteria. in §Position in a sequence is misleading. and match that description. I'm trying to come up with wording that is both accessible to the layman and correct, and the terms discrete and well ordered would seem to be too technical for an informal section. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:25, 3 March 2026 (UTC)Reply
It's a fair point, but this section is giving an intuitive description of natural numbers, and I think your objection is beyond the scope here. Doesn't the word "object" in this context imply discreteness?
For example, if I have bag of oranges and want to order them by weight, I don't think this means you have consider the infinite ways you might cut them into pieces first. TheGrifter80 (talk) 02:51, 6 March 2026 (UTC)Reply
Further to this: I fully agree that nothing in this section should be incorrect or misleading. The challenge is that naturals are logically prior to rationals and reals so it's difficult to give an understandable conceptual description in a way that simultaneously accounts for further extensions of the concept.
As I mentioned in a comment above I have used "collection of objects" which (IMO) means finite and discrete in ordinary language. But always keen to hear different views and suggestions for improvement. TheGrifter80 (talk) 11:14, 8 March 2026 (UTC)Reply
I tried, the key was using a definition of sequence that did not assume the natural numbers (so one borrowed from ordinal numbers for transfinite sequences). The problem is it is a very technical workaround for what is a relatively intuitive observation. I'm not saying what I did is the final form, but I think it is an improvement, both in terms of Victor's complaints and in sourcing. Mathnerd314159 (talk) 22:27, 15 March 2026 (UTC)Reply
I can see where you're going with that, but I agree it's too technical for an intuitive section. What do you think are the inaccuracies of the section as it previously was? If it is that some/all of the terms "sequence", "order" and "ordinal number" were used in a way that didn't strictly match their technical definitions, I think another approach would be to rephrase using only ordinary language, without implying any technical definitions. TheGrifter80 (talk) 00:58, 16 March 2026 (UTC)Reply
The current version does not seem correctly targeted for the expected audience. I think we should instead start with a gloss along the lines of "A sequence is an ordered list of objects". (And we could even initially skip "ordered", which is perhaps included in the concept of a "list".) –jacobolus(t)03:56, 16 March 2026 (UTC)Reply
What about something like this to start with:
A sequence is an ordered list of objects. The natural numbers form an infinite sequence, meaning they have a fixed order, specific starting point and no end point: 1, 2, 3, and so on indefinitely.
Any finite collection of objects that is ordered in some way, such by size or weight, also forms a sequence that can be described using natural numbers. Each object in the sequence is identified with a number, meaning that it occupies the same position in the sequence of objects as that number occupies in the sequence of natural numbers.
The typical use of natural numbers is actually for infinite sequences, e.g. you will see which implicitly indexes an infinite sequence . Well, see what you think of the new version. Mathnerd314159 (talk) 16:44, 19 March 2026 (UTC)Reply
Natural numbers are answers to questions like "how many..."
Natural numbers are answers to questions like "how many apples are on the table?"
you wrote "this phrasing is confusing" when reverting. Could you elaborate on what is confusing here? If others can explain why this is confusing, that would be helpful too. Thank you, Ebony Jackson (talk) 01:24, 17 March 2026 (UTC)Reply
An "answer" to the question would be a sentence such as "There are 4 apples on the table". Such a sentence is self evidently not a number. Saying that a number is an "answer" is going to confuse the hell out of a large number of readers. It confuses the hell out of me. Beyond that, even if you think most people will expext "4" to be the "answer to the question ...", you start running into nontrivial philosophical questions about the meaning/nature of numbers and other abstract objects that are well out of scope here, and I just don't think it's worth opening that can of worms. –jacobolus(t)03:25, 17 March 2026 (UTC)Reply
Let me turn it around, in case it helps to show why I find this construction potentially confusing. We might imagine a hypothetical conversation: A: "What is a natural number?" B: "It's the answer to a question about apples on a table." –jacobolus(t)15:21, 17 March 2026 (UTC)Reply
I could live with either version. but agree that saying numbers "are answers" feels a bit awkward. It seems much easier to say what numbers are used for than what they truly "are".
If the aim here is to use more direct language, perhaps "are used" instead of "can be used" is a small improvement.
Or, another option. Frege refers to:
positive whole numbers which give the answer to the question "How many?"
Latest comment: 7 months ago8 comments3 people in discussion
@TheGrifter80 thanks for your nice images of apples and oranges. Those are clear and effective. I don't think the animation part is really that essential though, and may be distracting for some readers. Each of the two animations has one frame which I think effectively conveys the same message; can we just go for those as static images instead. (Aside: if you do want to keep an animation, try to make sure that the initial frame is as close to independent as a stand-alone static image as possible, since some readers may have animation disabled and articles can be presented in contexts where an animation is frozen, e.g. turned into a PDF file.) –jacobolus(t)17:00, 20 March 2026 (UTC)Reply
jacobolus, thanks - yes agree for the first image the animation doesn't add much so I'll revert to a static image. For the second one I was trying to convey the dynamic aspect of counting. Let me amend it so that the initial frame works as a standalone and let's see how that looks. If consensus is that it's better static, I can change. TheGrifter80 (talk) 23:39, 20 March 2026 (UTC)Reply
For the "cardinality principle of counting" image, I too think that a static image would be better. Maybe instead of animation you could highlight the 3 in some way to indicate that it is the answer to the counting problem. Ebony Jackson (talk) 01:36, 21 March 2026 (UTC)Reply
@jacobolus, @Ebony Jackson - How about this? If not, give me your suggestion or feel free to edit the file.
That looks pretty good, though I might scoot things to the left and not leave the "4..." quite so pushed into the margin. Also, is it just me or is the middle arrow not quite centered between the other two? –jacobolus(t)04:15, 21 March 2026 (UTC)Reply
I too think it looks pretty good. My own personal nitpick would be to ask that 1,2,3,4 be equally spaced instead of having the distance between 3 and 4 be greater! Ebony Jackson (talk) 04:33, 21 March 2026 (UTC)Reply
Latest comment: 4 days ago22 comments6 people in discussion
I propose the following changes to the lead:
[A] Changing the definition in the first sentence:
Current: "the numbers0, 1, 2, 3, and so on, possibly excluding 0.[a]"
Proposed: "the numbers created by starting with 0 or 1,[a] and adding 1: (0), 1, 2, 3, 4, 5, and so on."
The current phrasing does not have any explanation for how to continue the sequence, instead relying on readers inferring the pattern, which can be continued in various ways (e.g. OEIS:A269303, A029931, A001590, A002981). Sure, the correct sequence is presumably obvious to most readers, but does it hurt to be explicit?
[B] Adding "(and may differ in whether they include or exclude zero)" after the list of alternate names.
For instance, the modern convention is that "positive" means greater than zero, whereas "non-negative" means greater than or equal to zero. I've sometimes heard "whole numbers" and "natural numbers" used distinctly in the same way; I don't remember which was which.
[C] Adding a link to "counting" in the second paragraph. I'm confused by the comment about common language vs technical meaning here.
[D] Linking "numerals" to numeral system in the third paragraph. This elaborates on the (IMO somewhat off-topic) statement about how numbers are written using what I'd call digits.
[E] Changing [[Inequality (mathematics)|compared by magnitude]] to [[inequality (mathematics)|compared]] by magnitude, since "inequality (mathematics)" is about comparison, not magnitude.
[F] Changing the explanation of how they are compared:
Current: "with larger numbers coming after smaller ones in the list 1, 2, 3, ..."
Proposed: "with 0 or 1 being the smallest, and each following number being larger than the previous: 0 < 1 < 2 < 3 < 4 < …"
Using less-than signs makes it more obvious what is going on, and avoids randomly introducing some poorly-defined list.
Each attempts to redirect to a section of "arithmetic", but only the latter works properly.
[H] Clarifying how the number systems are built up:
Current: "can be formally defined in terms of natural numbers"
Proposed: "each one can be formally defined in terms of the preceding one, starting with the natural numbers, which are themselves formalizable using the Peano axioms"
As far as I understand it, each construction is generally built on the previous layer, not all directly from natural numbers.
Sure, axiomatic systems are an advanced topic, but Peano arithmetic is highly relevant here, as the bedrock underlying much of the modern understanding of natural numbers.
I tried twice to implement these changes directly (Special:Diff/1374578936, Special:Diff/1374701935), but have been reverted both times, and was told to discuss here prior to making such edits.
(A) seems like an unnecessary level of indirection that makes readers do extra interpretation for no particular benefit. We don't need to belabor this in the first sentence. I'm not really worried about people mistaking "numbers0" for a single link, but it would also be fine to take the wikilinks on numbers out of this sentence; we could link at least the words zero and one at some later point. (B) seems overcomplicated. (C) is fine either way (D) Linking 'Numeral system' doesn't really seem that helpful in this particular spot. We should somewhere on the page discuss the Hindu–Arabic numeral system (if nowhere else in the "History" section), but it doesn't have to be here. (E) I think the previous "compared by magnitude" is the more useful link text; not every type of comparison is an inequality. If you want we could change "magnitude" to "size", which might be more accessible. (F) seems overcomplicated. (G) Using "arithmetic" as an adjective can be sometimes confusing because "arithmetic" is also a noun, and the two are spelled the same but pronounced differently; "arithmetical" is unambiguous, remains very common, and seems fine. I don't care strongly here though. (Aside: I fixed the broken redirect.) (H) also like an unnecessary complication that dwells on details of the construction rather than the point here that natural numbers underly them; the specific details differ from one source to another and the choice depends on the context, audience, and author's needs. I would find a different way of mentioning Peano axioms if that seems like it needs to be in the lead. Overall, I think this is a net disimprovement. YMMV. –jacobolus(t)20:31, 13 September 2026 (UTC)Reply
[A] The current phrasing does not have any explanation for how to continue the sequence - yes, we do not want to get involved with details like positional notation and the mechanics of how decimals increment.
[A] OEIS [shows it can be] continued in various ways - it is sequences A000027 and A001477, links somewhere in the article would be good I think, but no reason for them to be in the lead
[A] rel[ies] on readers inferring the pattern - yes, this is exactly how most books introduce the natural numbers, e.g. Enderton which is referenced
[A] does it hurt to be explicit? the correct sequence is presumably obvious to most readers. - well, this is somewhat complex. For this initial lead sentence we are at primary school-level writing. Primary school students typically have learned the verbal sequence of counting by the time they are reading Wikipedia, but the concepts of numbers representing real quantities is somewhat slippery for them. That is why we have these useful apple and orange diagrams. I think both the explicit enumeration and the procedural definition are useful. But from this study, it seems the explicit enumeration will be immediately recognizable to children ages 4-5 as the counting sequence, whereas the procedural definition they will kind of nod their head but not be able to actually apply it usefully until age 6+. So my suggestion would be to put it after the explicit enumeration, like In mathematics, the natural numbers are the numbers 0, 1, 2, 3, and so on, possibly excluding 0.[current citations] These are the numbers created by starting with 0 or 1 and adding 1 successively.[OEIS]
[A] The current version also violates MOS:SEAOFBLUE with "numbers 0". - so I think you are misreading seaofblue too strictly, it is only where the link can be confused as one single link. In this case it would not make sense for "numbers 0" to be a single link.
[B] - this is explained more in the terminology and notation section, with more extensive sources
[C] so this comment is based on MOS:OVERLINK, the words "counting" and "ordering" are used in their everyday meaning and linking them draws attention away from the cardinal/ordinal number links. The words with links to pages were added 2003 and at some point someone removed the links and added the comments.
[D] Linking "numerals" to numeral system - I don't think this is right, my vote for the target would rather be Numerical digit
[E-H] I have no issue with these, they are not so important. Regarding the redirect, it seems it is already fixed.
(EC) As to point [E], I just delinked it. I don't think this is a natural place for the reader to say "say, I think I'd like to go learn more about the notion of inequality in mathematics". I agree with most of the rest of Jacobolus's points. I would put the bit about Peano arithmetic in a somewhat stronger form. The natural numbers have been studied by mathematicians for thousands of years; it's not reasonable to suggest they didn't know what they were doing till Peano came along. Peano arithmetic is an important subject to mention and this could be done in the lead, but it's important not to frame it as the basis for the concept of the natural numbers. --Trovatore (talk) 20:51, 13 September 2026 (UTC)Reply
Ok, thank you all for the feedback! Here are my updated opinions on each point:
([A]: new proposal) I do not think we need to provide an extensive explanation here of precisely how to compute the elements of the set, but I do think some hint about how to keep going would be helpful. What do you think of this alternative, that I think provides a clear definition, building on the elementary understanding of counting?
In mathematics, the natural numbers are the numbers used for counting up, starting from 0 or 1.[a] The first few natural numbers are 0 (if included), 1, 2, 3, and so on.
([B]: new proposal) How about linking to that section, like this?
([C]: accepting status quo) I do think counting is relevant, but I don't feel strongly about whether it's linked, so I'll leave it alone.
([D]: new proposal) I'm unsure what to do with this, since I'm not sure why we're discussing decimal notation here at all, but how about changing the wording from "numerals" to "decimal digits" (which redirects to the IMO-awkwardly-titled article "numerical digit")?
([E]: accepting some of your suggestions) Delinking it and changing "magnitude" to "size" makes sense to me.
([F]: new proposal) How about "Natural numbers can be compared by size: 0 < 1 < 2 < 3 < 4 < …."
([G]: accepting status quo) Thanks for fixing the redirect! I see your point, and I'm now okay with leaving this unchanged.
([H]: accepting status quo) Ok sure, I guess this is fine as-is, especially since:
The other systems' definitions in terms of each other can be reduced down to using natural numbers, and some of them may directly use natural numbers anyways (e.g. the denominator of a rational number, or the index of a term in a sequence backing a real number)
There are ways to define operations on natural numbers other than using the Peano axioms, such as the set theory approach explained in §Set-theoretic definition, computations with digits, or maybe by taking the Peano axioms but replacing the successor operation with concepts from group theory.
I don't think the changes to the first few sentence of the lead add anything useful and just add repetition. As it stands now the first paragraph starts "the natural numbers are the numbers used for counting up.." and this is followed by the second paragraph which repeats "The natural numbers are used for counting..."
I think the very first line of the lead should identify the topic in the most basic and direct way possible (which will always require some basic assumed knowledge), and most authors just list the first few numbers to do this. For example:
Mayberry: "The natural numbers 0, 1,2, ... , as we now understand them...."
Quine "The condition upon all acceptable explications of number ( that is,of the natural numbers 0, 1, 2, ... )"
Russell: "...he obvious starting-point of mathematics would be the series of whole numbers, 1, 2, 3, . . . etc."
How about changing the second paragraph instead of reverting the first one? This is what I propose: A natural number can be used as a quantity (e.g. "there are seven days in a week"), in which it is called a cardinal number, or to label a place in an ordered series (e.g. "the third day of the month"), in which case it is called an ordinal number. FWIW, I consider both of those cases to be "counting up", either how many there are or what position it has, respectively, and I think that's fundamental to how they are formed and what they are used for.
As for repeating the word "numbers", MOS:REDUNDANCY says "Sometimes a little redundancy is unavoidable", and provides "The Oxford English Dictionary is the principal historical dictionary of the English language" as an example, which repeats both "dictionary" and "English".
I think you make some fair points and perhaps they belong elsewhere in the article, just not in the very first line. From previous comments I see you were concerned that it might not be obvious how to continue the sequence 0, 1, 2, 3.... hence we should add something about counting or adding 1 to be clear. But counting or adding 1 pre-supposes knowledge of the sequence of natural numbers, so it's really the same inference. IMO if someone reads "0, 1, 2, 3, and so on" and truly does not know that 4 comes next, then they will not know what counting or adding 1 means. To me, the phrase "the natural numbers are the numbers used for counting up" is indirect (saying what they are "used for", not just showing what they are). TheGrifter80 (talk) 10:48, 10 November 2026 (UTC)Reply
But counting or adding 1 pre-supposes knowledge of the sequence of natural numbers, so it's really the same inference. How so? Someone could understand the concept of counting without being aware of what the underlying set of numbers is called. Adding 1 can be done in other groups, such as the real numbers.
For direct, how about one of these alternatives?
"In mathematics, the natural numbers are the numbers formed by starting with 0 or 1 and counting up: 0 (if included), 1, 2, 3, and so on"
"In mathematics, the natural numbers are 0 (if included), 1, 2, 3, and so on, counting up to higher and higher numbers"
Isn't the question of terminology covered by the second sentence which says that "counting numbers" is another term used for natural numbers? I'm not really seeing the problem you're trying to solve here, or where genuine confusion could exist with the original sentence. TheGrifter80 (talk) 09:22, 11 November 2026 (UTC)Reply
I explained my complaints about the first sentence in point A of my initial comment. Another problem I forgot to mention there is that I find tacking on "possibly excluding 0" at the end to be terribly awkward phrasing.
Yes, most readers will be able to infer how to continue the sequence based on their own intuition and the context, but I think it's better to be explicit than implicit unless there's a substantial cost to doing so.
Do you have any problems with my proposed alternatives? I'm open to other wording if you have any ideas for a different compromise.
It's actually very difficult to succinctly "define" natural numbers. The things you suggest we should include to make the initial definition more explicit, such as "counting" or "adding 1" are themselves defined in terms of natural numbers, so it's a circular definition. Yes the definition relies on the reader knowing how the sequence continues, that the next number is 4, which is reasonable. To say the sequence continues on by "adding 1" is to rely on this exact same knowledge, because adding 1 just means going to the next number in the sequence, ie 4.
I don't really like either proposal. The first implies the numbers are "formed" by some logically pre-existing process of counting which I dont agree with. In the second, the phrase "counting up to higher and higher numbers", is redundant, because that's exactly what is meant by "and so on".
I do agree that the phrase "possibly excluding 0" is a bit awkward, so if you can suggest an alternative feel fres. But we dont want to get bogged down.in any confusing discussion about whether not 0 is a natural number or not in the first line. TheGrifter80 (talk) 11:41, 11 November 2026 (UTC)Reply
I agree that this is a difficult concept to define. However, the first sentence is not required to have a perfectly formal, axiomatically-grounded definition. WP:EXPLAINLEAD says we should try to make the lead accessible to a general audience rather than being too technical. Therefore, beginning with an informal definition may be best.
Furthermore, I disagree with your analysis of what's circular or redundant. Counting up, without further qualification, means repeatedly applying the successor function, which is the crux of the Peano axioms and the set-theoretic definition. Adding 1 means repeatedly applying the operation + with the identity element of the operation ×. I don't see how either of those inherently relies on the natural numbers being predefined. For instance, the latter could be modeled geometrically by placing unit-length line segments end-to-end, and declaring the set of possible resulting total lengths to be the natural numbers.
(Remember, Wikipedia is supposed to be a reference work for general audiences, not a rigorous step-by-step lesson plan for robots with no external experience. Most readers of this article will already know how numbers work, but just not be familiar with what the various sets are called. Thus, even definitions that might technically be circular can still be very helpful for someone who intuitively understands the derived concepts but not the formal foundations.)
"and so on" could be continued with any matching pattern, which may be spaced differently or even non-monotonic. "counting up to higher and higher numbers" specifically indicates that the sequence is increasing by a constant difference.
I indeed think we should try to address the issue of 0 clearly and concisely up-front, not with any long explanation there of when each convention is used, but rather by briefly making it clear that its smallest element may be 0 or 1 (with a clarifying footnote like the current one), and leaving any further discussion for the body.
I agree with most of what you've said about formality etc. But i still struggle to see a huge problem with it as is. Realistically who would read the first paragraph and think, "could this be referring to the sequence of Numbers k such that (266*10^k + 1)/3 is prime?"
True, the intended continuation is pretty obvious; I just don't want to leave needless ambiguity in the definition itself.
I'm not sure about 'for ordering', since rational and real numbers work just as well for that, but I think 'for ranking' would make sense.
So how about this slight tweak?: "In mathematics, the natural numbers are the numbers used for counting and ranking: 0 (if included), 1, 2, 3, 4, 5 and so on."
If you want to put a countable list of arbirary things in order, using real numbers as indices seems pretty unhelpful and confusing to me. YMMV. –jacobolus(t)04:21, 13 November 2026 (UTC)Reply
Indices are generally natural numbers, for good reason, but there are various situations where allowing real numbers (or at least rationals) can be invaluable. In particular, using fractional values allows inserting items in between other ones without restriction.
Here are some examples:
In the Windows operating system, each minifilter driver has an altitude that "is an infinite-precision string interpreted as a decimal number. A filter driver that has a low numerical altitude is loaded into the I/O stack below a filter driver that has a higher numerical value." (source)
On reflection I wonder if the link I removed that piped "compare by magnitude" to inequality (mathematics) was really trying to get at the ordering on the naturals, which is one of the most important things about them. I added a sentence to link to total order for this purpose; I think that's more on-point than inequalities. (Bonus: It doesn't need a pipe.) --Trovatore (talk) 04:29, 14 September 2026 (UTC)Reply
I'd entirely skip the jargon term "total order" from the lead section here, but if we're going to include it with a wikilink, a substantial rewrite of the first few sentences of Total order, aiming to make it much more accessible, is called for. I think a lot of the readers of this article (Natural number) are going to have great difficulty making sense of:
In mathematics, a total order or linear order is a partial order in which any two elements are comparable. That is, a total order is a binary relation on some set , which satisfies the following for all , , and in : [...]
This kind of formal definition would be better to put second, after first leading with a plain language explanation and maybe a couple of examples. –jacobolus(t)07:32, 14 September 2026 (UTC)Reply
Instead of the current
Natural numbers can be compared by magnitude, with larger numbers coming after smaller ones in the list 1, 2, 3, .... This gives a total order on the natural numbers.
I'd suggest we stick with something simpler, along the lines of
Natural numbers can be compared by size; when listed in order, 1, 2, 3, ..., larger numbers come after smaller ones."
Instead of a wikilink to some other article, we could wikilink 'compared by size' to the section §Order on the same page. (We could perhaps also somewhat rewrite that section to be a bit more accessible, mentioning again that numbers can be compared, and describing how and work for natural numbers.) –jacobolus(t)07:40, 14 September 2026 (UTC)Reply
I would strain hard to avoid the EGG-y sort of piped link in your parenthetical suggestion. It doesn't give enough notice to the reader as to what to expect when following the link, and it's getting too close to making the semantics different for readers who see a printed version.
The point of my addition was to mention one of the key aspects of the natural numbers, namely considering them as an ordered set. (Arguably that's the main fact about the natural numbers, given that everything else about them is determined by that.) If that's too technical for that spot, then I'd prefer to just remove it and not replace it with anything. --Trovatore (talk) 16:32, 14 September 2026 (UTC)Reply
Points [D], [G] and [H] correspond to real issues, even if, in two cases, I disagree with the proposed solutions.
[G] (arithmetic vs. arithmetical): as "arithmetic" is used in the linked article, I changed "arithmetical" to "arithmetic", as suggested by Sollyucko
[D] (numerals vs. digits): The sentence was very confusing, as "numeral" refered gramatically to the 10 symbols, while the context shows that it is intended to refer to the symbols built from digits. I have fixed this. This gives:
Natural numbers are commonly expressed in writing using ten symbols called digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9). These expressions, called numerals, can also ...)
[H] (other number systems): The main issue here is the use of the jargon term "formally". But I agree also with Solomon Ucko that these number systems are not defined directly from the natural numbers, but in terms of the preceding one. So, I changed the sentence into
Latest comment: 1 day ago18 comments6 people in discussion
Starting a new discussion because having Lazard's comment at the end is somewhat confusing. So first MOS:LEADSENTENCE. So note there "a" concise definition, it should only be one. At the primary-school level there are roughly three definitions of natural numbers:
Enumeration:
"Natural numbers are denoted in mathematics by the symbol N. In set notation, we write N={1,2,3,4,...}."
"1, 2, 3, ... are natural numbers."
"The symbols 1, 2, 3, 4, 5, ... are called the natural numbers."
"The natural numbers are 1, 2, 3, 4, 5, and so on"
"Natural numbers are 1, 2, 3, 4, 5, 6, and go on till infinity."
Peano-style:
"Natural numbers start at 1 and increase in ones up to infinity"
"Natural numbers are the positive whole numbers starting from 1."
"Natural numbers are all positive and start with the number 1 and go up from there."
Counting-based:
"The numbers that express the result of our counting are called natural numbers."
"Natural numbers are the numbers we use when we count things."
"Natural numbers can be viewed as counting numbers"
"Natural numbers are counting numbers"
"The counting numbers are the numbers that you first learned as a little child, being the numbers that you used to count things... The natural or whole numbers are the counting numbers, together with zero"
Mayberry, Quine, Russell that TheGrifter80 cites are not at the primary school level and are not good sources for this, e.g. Russell calls it "obvious", this is just bad style (MOS:MATH#OBVIOUS). Also the claim "most authors" is really subjective. I went with a counting-based one for the lead sentence, in particular the short description, based mostly on style - we can link numbers and counting and they're nice links, no sea of blue problem, easy to click. Also it seems like emphasizing the association between natural numbers and counting is the primary educational approach (hence also the use of the term "counting numbers" to make it explicit in the name), while the enumeration and the Peano properties are more like mnemonics to help children remember. In terms of "the most basic and direct way possible" going back to MOS:LEADSENTENCE the focus is on concision and I don't think you can get more concise, they're numbers and they're the ones used for counting, even the enumeration definition is roughly the same syllable-wise. Mathnerd314159 (talk) 20:31, 14 November 2026 (UTC)Reply
The recent changes to the first few sentences seem to me to make it more cumbersome and wordier, and somewhat confusing, for no apparent benefit. I think your quoted example above, "The natural numbers are 1, 2, 3, 4, 5, and so on", seems about right. If the main worry was with "possibly excluding 0", then perhaps instead of the previous "In mathematics, the natural numbers are the numbers 0, 1, 2, 3, and so on, possibly excluding 0." we could switch to something like "In mathematics, the natural numbers are the numbers 1, 2, 3, 4, and so on; sometimes 0 is also included, before 1." The current phrasing ("... consists of 0 (optionally), 1, 2, 3, and so on – starting at 0 or 1 depending on convention and increasing by one indefinitely.") seems terrible to me. –jacobolus(t)04:32, 15 November 2026 (UTC)Reply
I think retroactively removing an element is less confusing than inserting one, but having a list with zero directly marked as conditional is clearest to me, because I see it as having a fuzzy starting point, rather than modifying a list. Do you agree or disagree, and why?
Would changing "indefinitely" to "forever" help with making it more accessible to younger readers? The question of which one is more accurate is a question of mathematical philosophy, and I can see arguments for both, so I think either one is fine.
One reason to lead with the version starting from 1 is that it seems much more common in reliable sources, especially so in less technical sources. –jacobolus(t)17:17, 15 November 2026 (UTC)Reply
@Mathnerd314159. Useful to have that survey. I think your final comment is about right. The initial description should enumerate the first natural numbers and make some reference to counting.
But I do agree with @Jacobolus, the current phrasing doesn't seem quite right. The second sentence feels like it's trying to get too many ideas into one sentence.
My suggestion, based on your edits and @Jacobolus comment above:
"In mathematics, the natural numbers are the numbers used for counting: 1, 2, 3, 4, and so on indefinitely. Sometimes 0 is also included, before 1." TheGrifter80 (talk) 08:56, 15 November 2026 (UTC)Reply
This is still worse than the previous version, because it reads awkwardly. Mentioning that the numbers are used for counting is fine to put at a point after the first sentence. Adding the word "indefinitely" after "and so on" is not too helpful: readers who already know this well don't need it and readers who don't won't necessarily know what it means. It would be better to add a full sentence later in the lead section explaining that the sequence of natural numbers is infinite. –jacobolus(t)13:35, 15 November 2026 (UTC)Reply
I agree that "indefinitely" is unnecessary after "and so on", and that discussion of the set being infinite should be done further down.
However, what's wrong with mentioning counting at the start? For comparison, a real number is described as "a number that can be used to measure a continuous one-dimensional quantity such as a length, duration or temperature". When a formal construction would be too complicated to use as a definition, we define the concept based on its fundamental purpose instead.
If you really want to have just the list of elements in the first sentence, please identify another Wikipedia article that defines its subject primarily based on examples, so that we can compare this with it.
The first sentence ("In mathematics, the natural numbers are the numbers used for counting.") is misleading, since it excludes implicitly the other uses of the natural numbers (ranking, labelling the result of a count, etc. Defining the natural numbers by extension, as it was the case before your first edit is convenient for many reasons: for example, 4-years-old kids learn individual numbers before learning the meaning of the word counting, and Peano axioms define the nonzero natural numbers as succdssors each of the preceding one.
So, I strongly disagree with the current version of the two first paragraph, and suggest to restore the two first paragraphs of the version of the 11 August 2026 (before first Solomon Ucko's edit). This version is much better than every version proposed since then. (The third paragraph has been improved since this date). D.Lazard (talk) 16:13, 15 November 2026 (UTC)Reply
It does not state that the natural numbers are only used for counting; it merely implies that counting is the primary use. Ranking items and labeling the result of a count are both based on counting.
I don't mind a proper extensional definition, i.e. specifying the whole set of values rather than the required properties, but just listing the first few without specifying how to continue the sequence is instead an ostensive definition.
Do you really think that 4-year-olds who don't know what counting is are going to be looking up the definition of "natural number" on Wikipedia?
I would be happy to have a definition that is based on the Peano axioms. For instance, I have proposed referring to (iteratively) applying the successor function as "adding one" or "counting up", but I'm open to other phrasing that implies the same meaning. However, "and so on" does not imply this, unless further clarification is provided shortly before or after it, because by default the continuation could, for example, repeat previous elements.
What do you disagree with in the second paragraph? I concur that the third paragraph (about numerals) has been improved.
The purpose of the first sentence is not to be a definition. So if the first sentence is inadequate as a definition, that is entirely fine.
"Do you really think that 4-year-olds ..." – Exactly the same argument applies for how readers will understand "and so on". This is especially not a problem if we add an explicit sentence somewhere in the lead section that explicitly says that the sequence of natural numbers is infinite. No plausible reader, who already understands what counting is, is going to expect that "the continuation could [...] repeat previous elements". –jacobolus(t)20:24, 15 November 2026 (UTC)Reply
If this was an article on square numbers and the first sentence was something like: "square numbers are the numbers 0, 1, 4 and so on" then your objections would be valid.
But this is not an article on one particular set of numbers (N) in which we define that set taking the underlying understanding of numbers as given. It's an article on that underlying concept itself. So IMO an "ostensive" description that points to the numbers in a way that clearly identifies what we are talking about is totally appropriate. TheGrifter80 (talk) 08:44, 16 November 2026 (UTC)Reply
Yes I do agree with you there and to me the original description was fine. I can see the case for some reference to counting very early in the lead, but not at the cost of awkward or confusing phrasing. TheGrifter80 (talk) 06:18, 16 November 2026 (UTC)Reply
It is worth looking at the page history, it went from a bare enumeration to two enumerations (arguing about zero) to counting-first back to enumeration and now back to counting-first in my most recent edit. I do think the enumeration-first definition has some merits but looking at WP:BETTER/GRAF1 it is clear that a concise intensional definition is better than an extensional definition because it puts the topic in context more quickly - it is like defining chair as "4 legs and a platform" vs. "something you sit on". The issue though with the previous counting-first definition was it had these examples mixed in instead of being concise. So now those are in the second paragraph though, so a concise intensional definition seems possible and more attractive. I extended it with ordering and labeling, I fixed the 'terrible' second sentence, and I even managed to fit the Peano axioms in. I'm happier with this version, maybe it is still not ideal but it addresses the issues that have been brought up. I think it is important to have all 3 of the definitions (intensional/counting, extensional, Peano/formal) in the lead paragraph. Mathnerd314159 (talk) 22:33, 15 November 2026 (UTC)Reply
The new version continues to be worse. It's awkward and confusing. Saying that the natural numbers are 1, 2, ... is better than saying that they "consist of" 1, 2, ..., and the parenthetical "(in some conventions)" is awkward. The phrase "increasing with no upper bound" does not belong in this sentence. –jacobolus(t)00:17, 16 November 2026 (UTC)Reply
Oh gosh. I haven't been following this discussion, unfortunately. But I don't agree with putting the Peano axioms in the opening paragraph. The natural numbers were understood for millennia before Peano. Also saying the axioms "define" the objects is problematic; the axioms say how the objects behave but do not define them. --Trovatore (talk) 22:38, 15 November 2026 (UTC)Reply
Ah, my haste to finish up the edit shows. It's definitely a hard phrasing problem because we have both "Peano axioms" listed as subsection of Formal definitions and then also the sentence "The Peano axioms do not explicitly define what the natural numbers are". I still think some kind of formal definition is warranted in the lead, even if it's not attributed to the Peano axioms. Mathnerd314159 (talk) 22:44, 15 November 2026 (UTC)Reply
Please, stop editing the lead for implementing your own preferences, without any consensus. The lead is discussed here since several months, and no consensus has emerged. So, I'll apply the policy WP:NOCONSENSUS and restore the result of a previous consensus, that is, the version of the 11 August 2026 of the first two paragraphs. If you have concerns with this version, please open a new discussion, separately for each concern, in order that it will possible to reach a consensus. D.Lazard (talk) 11:04, 16 November 2026 (UTC)Reply
Requesting feedback on a proposal for the first sentence
I disagree: it is much worse than the previous stable version hat I have just restored. The Jacobolus' comment that you cite applies perfecly to this new proposal: "it is more cumbersome and wordier, and somewhat confusing, for no apparent benefit". Please, before proposing anything new, state clearly which issue of the current version you are trying to fix, and try to get a consensus that the issue needs really to be fixed. D.Lazard (talk) 16:14, 16 November 2026 (UTC)Reply
The issues have been discussed - covering all of the 3 types of definitions (extensional/enumeration, intensional/"counting, ordering, labeling", formal/Peano), making the intensional definition first in the because it puts the topic in context, and phrasing the inclusion of 0 properly. The problem is doing it a concise way that is not confusing or awkward. I think we were close before your revert, it seems it needs only a few stylistic fixes. I don't think 11 August is any better than 21 September and that stood for several months before TheGrifter80 reopened the discussion. Mathnerd314159 (talk) 02:33, 17 November 2026 (UTC)Reply
Calling it worse with no explanation is still not a valid argument, as explained in the essay Wikipedia:I just don't like it. I think the "0, 1, ... possibly excluding 0" phrasing sucks, but I don't think saying so is going to convince anyone either.
I do see your and D.Lazard's point that this proposal is somewhat wordy, though I think that the additional words improve the clarity and flow rather than being confusing.
I don't think it's perfect, but I'm not sure there is a perfect way to address this slippery subject, so we should instead strive to find a good enough compromise that balances the various goals we're trying to achieve.
I have suggested various wordings in the talk page and article history and would be happy to adopt any of them. I am also open to considering other alternatives.
I think this one seems fairly confusing. I don't really like the previous "possibly excluding 0", but my proposal about it wasn't too great either. I think we should keep trying. –jacobolus(t)16:06, 16 November 2026 (UTC)Reply
@Sollyucko I dont like that one, as it's a very awkward sentence. If the objective is to more clearly emphasise the importance of counting, why not try putting that in the second sentence? TheGrifter80 (talk) 21:12, 16 November 2026 (UTC)Reply