Irrational 2.0

True or False?

Given any six irrational numbers, there always exist three such that the sum of any two of them is still irrational.

True False

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Áron Bán-Szabó
Jun 21, 2017

Let the six numbers be six points, called: A , B , C , D , E A, B, C, D, E and F F . If two numebr's amount is rational, then we link them with a red line segment, if the amount is irrational, then we link with a blue line segmnent. Since there should be a triangle which sides are colored the same ( Ramsey-theorem ), we will show that there will be a blue triangle. Suppose there is a red triangle, let the numbers be x x , y y and z z . Then x + y , x + z , y + z x+y, x+z, y+z would be rational numbers, so the sum of them is rational too: 2 x + 2 y + 2 z 2x+2y+2z . From that x + y + z x+y+z would has to be rational, but since y + z y+z is rational, x + y + z ( y + z ) = x x+y+z-(y+z)=x would have to be rational, but x is irrational. So there can't be a red triangle, and there must be a blue triangle.

Is the "distinct" condition needed?

Calvin Lin Staff - 3 years, 11 months ago

Log in to reply

You are right! It is not compulsory, I corrected it.

Áron Bán-Szabó - 3 years, 11 months ago

Log in to reply

Indeed.

Note that out of 6 points in 2 colors, there must be at least 2 chromatic triangles. Can you find the case where we have exactly 2 chromatic triangles?

Calvin Lin Staff - 3 years, 11 months ago

Log in to reply

@Calvin Lin Yes, that happens for sets of the form { α , α + r 1 , α + r 2 , α + r 3 , α + r 4 , α + r 5 } \{\alpha , \alpha + r_1 , \alpha + r_2 , - \alpha + r_3 , -\alpha + r_4 , -\alpha + r_5 \} , where α \alpha is an irrational and r i r_i are all rationals.

Shourya Pandey - 3 years, 11 months ago

Ok. So this has little to do with rational numbers. It is true of any two mutually exclusive options for classifying 6 or more objects.

Carl White - 3 years, 11 months ago

Log in to reply

For this problem you also have to show that one of the two mutually exclusive options is actually not possible; so it is specialized for rational numbers

Matt McNabb - 3 years, 11 months ago

Actually the closure of rational numbers is an essential part. Here we specifically need a blue triangle and not just any monochromatic triangle.

Shourya Pandey - 3 years, 11 months ago

This is a beautiful proof

Ian Prado - 3 years, 11 months ago

It looks like a stronger claim can be made: 5 irrational numbers are enough to give that property.

Marco Coden - 3 years, 7 months ago

Log in to reply

Can you prove that claim?

Calvin Lin Staff - 3 years, 7 months ago

Log in to reply

I'll try to prove it by showing that we reach a contradiction if we try to construct 5 number such that, taken any 3 of those, at least one of their pairwise sums gives a rational result.

This is based on the property that 3 irrational numbers can't have all 3 pairwise sums result in rationals.

Let's call the 5 numbers A, B, C, D, E. If we pick 3 of those (ex A,B,C) at least a couple will have a rational sum (else the statement would already be verified); we can say it was A+B. We apply the same reasoning to the 3 left numbers and without a loss of generality we can now say that C+D is rational. Now let's consider the triplet A,C,E: either A+C is rational, or it's one of A or C plus E; in that second case we can now consider B+D+E...

What matters is that now we have constructed a chain of at least 3 rational sums, let's say it was A+B, B+C and C+D. That means that A+D must also be rational, since A+D = (A+B) + (C+D) - B+C. Now that we have the four numbers on a "square" relation, we can compare the fifth (in this case E) with the 2 numbers on opposite sides of a diagonal. E must have a rational sum with either A or C, or else A,B,C would all have rational sums, which is impossible for irrational numbers. Same reasoning goes for E with B and C.. but now we see that A has a rational sum with one of (A or C) and one of (B or D), which again creates an impossible triangle of rational sums. QED.

Marco Coden - 3 years, 7 months ago

Log in to reply

@Marco Coden Very nicely done. The argument could be cleaned up slightly, but you have a correct approach.

Calvin Lin Staff - 3 years, 7 months ago
Tom Verhoeff
Jul 4, 2017

Let the six irrational numbers be a i a_i for i = 1 , , 6 i=1,\ldots,6 . First, observe that if a i + a j a_i+a_j and a i + a k a_i+a_k are both rational, then their sum 2 a i + a j + a k 2a_i+a_j+a_k is rational as well; hence, a j + a k a_j+a_k is irrational (for otherwise 2 a i 2a_i and thus a i a_i would be rational too). Consequently, there do not exist triples all of whose pairwise sums are rational. We call this property T T .

Consider the five sums a 1 + a i a_1+a_i for i = 2 , , 6 i=2,\ldots,6 . Either at least three of these are rational or at least three are irrational. If three are rational, then those three a i a_i have pairwise irrational sums (otherwise T T would be violated). If three are irrational, then because of T T not all pairwise sums of the three a i a_i can be rational. Suppose a i + a j a_i+a_j is irrational. Then a 1 , a i , a j a_1, a_i, a_j have pairwise irrational sums.

Fahim Saikat
Jul 16, 2017

Let assume that the statement is f a l s e false .

So, there always exist three such that the sum of any two of them is rational.

Let the six irrational numbers are A , B , C , D , E A, B, C, D, E and F F .

Here the three numbers are A , B , C A, B,C , the sum of any two of these numbers is rational.

So,

A+B=p \tag{1}

B+C=q \tag2

C+A=r \tag{3}

Here , p , q , r p,q,r are rational.

( 1 ) ( 2 ) + ( 3 ) p q + r = 2 a (1)-(2)+(3)\rightarrow p-q+r=2a

2 a 2a is irrational but p q + r p-q+r is rational .

So, p q + r 2 a p-q+r\neq2a Therefore , it is proved that , the statement is true.

It looks like you don't need 6 numbers at all, three will suffice, no?

Roman Sologub - 3 years, 9 months ago

Log in to reply

No, there was a mistake in the proof: "there always exist three such that the sum of any two of them is rational" is not the logical negation of the initial statement. (the correct negation is: at least one instance of six irrational numbers exists such that, whichever three you choose, the sum of at least two of them is rational)

Marco Coden - 3 years, 7 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...