Research diaries #10: The real line, gaps in the rational numbers and the axiom of completeness

in StemSocial6 months ago

Historically mathematics did not start at the true fundamentals. The fundamentals were only completed around the 19th century. All of the work done before that was done with some intuitive understanding of the basics, in particular the nature of the real line which we will denote by R in this post.

Not sure what images to paste. So here is my cat

Mathematicians like Cantor showed that our intuitive understanding of R is flawed. Hence, it requires a rigorous formulation so that we can proceed in a proof theoretical manner to obtain properties of R. Properties of R are of specific interest in relation to the rational numbers which we will denote by Q. Rational numbers are all numbers in R that can be expressed as a fraction of two whole numbers. There exist methods to construct R from Q but that is not the approach we will take as it is rather technical. For the interested you can have a look over here

What kind of fundamental properties can we assume on R? We assume that that we have the natural addition and multiplication so that we can perform all the natural algebraic operations that we did in primary school (more rigorously we are assume that R is a field). We also assume that there exist an ordering on R this means that the elements are ordered but that also the basic operations, addition and multiplication, apply to this ordering. So if a < b then also a+c < b+c (this means that it is an ordered field)

More of my cat

Although we have not proven the existence of numbers which are not in Q. We expect that there are numbers that are in Q but not in R. These numbers will create gaps in Q in relation to R. So how do we express that R is gapless. We phrase this as an assumption:

Axiom of completeness Every nonempty set of real numbers that is bounded above has a least/smallest upper bound

Here bounded above, also called upper bounded, means that there exists a largest element in R which is greater than all the elements in the set. There can exist many such bounds but the axiom guarantees that there exists a single smallest one. Finally, it important to note that the least upper bound does not necessarily need to be contained in the set. For example if we consider the set generated by the numbers -1/n where n=1,2,3, ... Then 0 is the least upper bound but 0 is not contained in the set.

To get a better understanding of the axiom let's look at it in the context of an example. Consider the set

A = { r ∊ R : r2 < 2 }

Then the axiom tells us that is has a least upper bound. However, if we consider

B = { r ∊ Q : r2 < 2 }

and imagine that we are only allowed to work on Q then there is no least upper bound in Q. Or more specifically, square root is approximately 1.4142... and consequently can be approximated from above by finer and finer rational approximation meaning 2, 15/10, 142/100 ... but there is no smallest one in Q. It goes on indefinitely. Hence, at least heuristically we observe how the axiom of completeness makes the real line complete

More cat tax


Interesting post 😁 but i have a series questions that may follow from the one i'm about to ask, it would depend on your response.

2, 15/10, 142/100 ...

Are these set of values for r² belonging to Q or r belonging to Q ?

These are values in Q. They are not contained in B. But they do bound B from above ^^

These are values in Q. They are not contained in B.

This looks convincing. In fact, that response alone cancels the other questions i would have asked.

But they do bound B from above ^^

But then this seems to be complicating your response.

The values you presented don't follow the rules of B ( for example, r = √1.5 or √(15/10) isn't a rational number). So my question, bound B in what context ?

B:={ r ∊ Q : r2 < 2 } only takes elements from Q. Indeed r2 = 1.5 or r2=15/10 are not contained in it. We are viewing B as subset of Q. And looking for a least upper bound in Q (it is the same what we did for the R story but now R is replaced by Q). For B we cannot apply axiom of completeness and (you can prove that) B has no least upper bound in Q

I have seen the Wikipedia article on this topic, it looks more explanatory. You didn't answer my first question well, probably why i was still confused.

The set of values you presented (2, 1.5, 1.42....) are actually for r belonging to Q. You said they belonged to Q but never stated if it was for r or r².

Sorry I don't understand the question. r can only exist within the definition of A or B. So we can only view it as something contained in a set.

2, 15/10, 142/100 are in Q but not in B. But they do bound B from above.

I pretty much understand you, ok.

2, 15/10, 142/100 are in Q but not in B. But they do bound B from above.

please don't repeat this again, it's making you sound like a robot. 😂

How about using the Cauchy sequence approach, I like that one, it looks more understandable.

Aww that was too adorable


Intuitive understanding is the mother of knowledge, just as important is the understanding of nature and its care.

Mothers, like intuitive understanding, can be wrong too :P

That is also true, observation and testing end up clearing up any doubt.

Thanks for your contribution to the STEMsocial community. Feel free to join us on discord to get to know the rest of us!

Please consider delegating to the @stemsocial account (85% of the curation rewards are returned).

You may also include @stemsocial as a beneficiary of the rewards of this post to get a stronger support. 

I don't pay much attention to the math... but that cat could have a post of its own :)

Congratulations @mathowl! You received a personal badge!

Happy Hive Birthday! You are on the Hive blockchain for 5 years!

You can view your badges on your board and compare yourself to others in the Ranking

Check out the last post from @hivebuzz:

HiveBuzz World Cup Contest - Quarterfinals - Recap of Day 2
HiveBuzz World Cup Contest - Quarterfinals - Recap of Day 1
Support the HiveBuzz project. Vote for our proposal!