PrepShorts · Study sheet · Class 9 Mathematics · Chapter 3, The World of NumbersPrepShorts

Chapter 3 · The World of Numbers

Irrational decimals: an expansion with no stop and no repeating block

यह वीडियो हिंदी में भी · Watch in Hindi

The decimal expansion as a signature9 min

This video could not be loaded. Reload the page to try again.

Sign in with Google

9 min.

Also recorded in Hindi.Englishहिन्दी

One third never stops either, and one third is about as rational as a number gets. So “never ends” cannot be what makes a number irrational.

The idea

"Never ends" is not the mark of an irrational number — one third never ends either. The mark is never ends and never repeats, and by the remainder argument that combination is impossible for any fraction, which is what makes the decimal expansion a genuine test rather than a rough guide. Two cautions come with it, and both are printed in this chapter. A visible pattern is not the same as a recurring block, so a decimal can look orderly and still be irrational. And a terminating decimal has a second writing that ends in endless nines, so the expansion identifies a number without being unique to it. The signature is reliable; it is just not a fingerprint.

What you should be able to do

  • State the decimal signature of an irrational number, in both of its parts
  • Explain why the test works in both directions, naming the argument that supplies each direction
  • Classify a mixed list of decimals and surds as rational or irrational, and find the fraction where one exists
  • Distinguish a decimal with a describable pattern from a decimal with a recurring block, and decide which is required for rationality
  • Recognise that a perfect square under a root gives a rational number
  • State the two printed irrational expansions the chapter supplies
  • Explain why every terminating decimal has a second writing in endless nines
  • Say what the non-uniqueness of decimal writings does and does not undermine

Words to know

TermDefinition in one lineFirst introduced
Irrational Numbersnumbers no ratio of integers can expressprinted in bold in §3.5, p. 53
non-terminatingnever stoppingprinted in §3.6.1, p. 58
repeating blockthe group of digits that recursprinted in §3.6.2, p. 61
terminating decimala decimal that stopsprinted in §3.6.1, p. 58
decimal expansionthe decimal writing of a number, however longprinted in §3.6, p. 57
non-uniquenessthat a rational number may have two decimal writingsprinted in bold at the foot of p. 62
signaturethe chapter's own word for what the expansion supplies as evidence of rationalityprinted in the Chapter Summary, p. 67, sixth bullet
Real Numbersthe rationals and irrationals taken togetherprinted in bold in §3.6, p. 57
describable patternan added phrase for a rule a decimal's digits follow that is not a fixed recurring blockan added term; the chapter poses exactly this distinction in a bracketed question and gives it no name
decision procedurean added phrase for the expansion used as a two-way testan added term; the chapter treats the expansion as the easiest way to tell the two kinds apart and does not call it a procedure

Where people slip up

  • "A decimal that goes on for ever is irrational." One third goes on for ever. This is the single most common error in the whole chapter, and the section exists to correct it.
  • "√ anything is irrational." √81 is 9, and it is deliberately placed first in the chapter's own list so that students who have just met the √2 proof will fall into it. Test the number under the root for being a perfect square.
  • "If I can see the pattern, it repeats." Item (v) has a describable pattern and no recurring block. Repeating means a fixed group returning again and again unchanged, not merely being predictable.
  • "A long messy decimal must be irrational." Item (vi) is twenty-one digits long and stops, so it is rational. Length and untidiness are not evidence.
  • "0.999… is a bit less than 1." The chapter says outright that this is what many would guess, and the algebra says otherwise. Give the derivation and then the interpretation.
  • "If a number can be written two ways, the decimal test is unreliable." Both writings give the same verdict. Non-uniqueness is about the writing, not about the classification.
  • "Non-terminating and non-repeating are two ways of saying the same thing." They are independent conditions, and it takes both together to force irrationality. A student who collapses them cannot classify item (iii) correctly.
  • "π and √2 have been written out completely on p. 61." They are shown to about twenty digits with an ellipsis. Neither can be written out completely, which is the point of the section.
Transcript1,279 words

Divide one by three and the digits never stop. Nought point three three three three, going on for as long as you are willing to keep writing. So here is a question worth getting right the first time. Is that what makes a number irrational? That its decimal never ends? It is not. One third is a fraction. It is about as rational as a number gets, and its decimal runs for ever.

Never ending is half an answer at most. On its own, it is not evidence of anything at all. The real mark of an irrational number is two conditions, and both of them have to hold. The decimal never stops. And no group of digits repeats for ever after. One third fails the second one straight away. A single three, over and over, is exactly a group of digits repeating for ever. So one third lands on the rational side, which is where it belongs.

Both conditions together. Drop either one and the test stops working. Why should two conditions be enough to decide anything? Because there are two arguments here, pointing opposite ways, and between them nothing gets through. The first direction says every fraction stops or repeats. When you divide, the remainder left at each step has to be smaller than the divisor, so there are only so many values it can take.

Keep going and one of them has to come round a second time. And when the remainder comes round, the digits have no choice but to come round with it. So a fraction cannot hand you digits that run for ever without ever settling. It has nowhere to put them. The other direction runs the opposite way. Anything that stops or repeats is a fraction. Shift the decimal along past its block, subtract the original, and the endless tail cancels exactly, leaving a whole number over a whole number.

Put the two arguments together and you have a test that decides in both directions. Every fraction with a divisor below two hundred and sixty was run. Thirty three thousand four hundred and eleven of them. Two thousand four hundred and forty one stopped, thirty thousand nine hundred and seventy repeated, not one did anything else, and every single one converted back into itself. So what does a number look like when it fails both conditions at once?

Root two. One point four one four two one three five, and it keeps going. Nineteen places written out here, twenty significant figures, with no end and no repeating block anywhere in sight. And pi. Three point one four one five nine two six five, and on it goes as well, to exactly the same length. Both of these were computed digit by digit rather than looked up. And neither of them can ever be written out completely. That is not a shortage of space. That is the whole point.

Now the trap that catches almost everybody. A root sign is not a certificate of irrationality. Root eighty one is nine. Run it through the very same machinery that produced those nineteen digits and you get nine point zero zero zero zero zero. It stops immediately. The question was never whether there is a root sign. It is whether the number underneath it is a perfect square. Below two hundred there are only fourteen of those.

Root twelve is a different story. Three squared is nine and four squared is sixteen, so twelve is not a perfect square, and searching every fraction with a divisor under four thousand finds nothing that squares to twelve either. Here is the harder trap, and it is the one this whole idea turns on. One point zero one, zero zero one, zero zero zero one, zero zero zero zero one, and onwards.

You can predict every digit of that number. A one, then a run of noughts, then another one, and each run is one longer than the run before it. There is nothing mysterious about it at all. So it certainly has a pattern. The question is whether it has a repeating block, and those are not the same thing. Look at the gaps. Between the first one and the second there are two noughts. Then three. Then four, then five, then six.

A repeating block is a fixed group of digits that comes back unchanged, for ever. Fixed is the word doing the work. If some block were repeating here, then the gaps between the ones could never grow longer than that block. But these gaps grow without ever stopping. Four thousand digits were searched, against every starting point up to two hundred and every block length up to two hundred. Nothing repeats. And in that stretch the longest run of noughts reaches eighty seven.

Being able to describe every digit and having a repeating block are two different claims. This number has the first and not the second, so it is irrational. Put six decimals side by side and sort them, because the sorting is where the traps live. Root eighty one is nine, so it is rational. Nought point three recurring is one third. Nought point one two three four five, repeating, is four thousand one hundred and fifteen over thirty three thousand three hundred and thirty three.

Root twelve is irrational, and so is the number with the lengthening runs of noughts. And then the last one. Twenty three point five six zero one eight five six, carrying on for twenty one untidy digits, and then stopping. It stops, so it is rational. The orderly looking number is the irrational one. The messy looking number is not. Length and untidiness are not evidence of anything. One more caution, and then the test is safe to use.

A decimal that stops has a second writing. Take two point four seven. Knock one off the last digit and hang endless nines on the end. Two point four six, nine nine nine nine, for ever. That is not a number a little below two point four seven. It is two point four seven. Convert the nines writing back and you land on two hundred and forty seven over a hundred, exactly.

Two thousand four hundred and thirty three terminating fractions were checked this way, and every twin came back to the number it was built from. The plainest case is the one that unsettles people most. Nought point nine nine nine, for ever. Call it x. Ten x is nine point nine nine nine, with exactly the same endless tail. Subtract, and the tails cancel completely. Nine x equals nine, so x is one.

Not nearly one. One. The two writings are two ways of spelling the same number. And notice carefully what this does not do. It does not break the test. Two point four seven stops, and two point four six nine recurring repeats, and stopping and repeating are both on the rational side. One number can carry two signatures. A signature never serves both kinds. So here is what you are holding now.

Hand me any decimal at all. If it stops, it came from a fraction and I can produce that fraction. If it repeats, it came from a fraction and I can produce that one too. And if it does neither, then no fraction made it. Not because nobody has found the right one yet, but because a fraction is incapable of behaving that way. It is not a fingerprint. Two different writings can name one number, and you have just seen that happen.

But it is a signature. And it never lies about which kind of number it came from.

Where this fits

Taken from the notes each video was made from, not from the reading order — these are the ideas this one rests on and the ones that later rest on it.

Builds on

Comes up again in

Either side of this one

The book

Open in a new tab