Steven Strogatz on math, from basic to baffling.
Like anything else, arithmetic has its serious side and its playful side.
The serious side is what we all learned in school: how to work with columns of numbers, adding them, subtracting them, grinding them through the spreadsheet calculations needed for tax returns and year-end reports. This side of arithmetic is important, practical and — for many people — joyless.
The playful side of arithmetic is a lot less familiar, unless you were trained in the ways of advanced mathematics. Yet there’s nothing inherently advanced about it. It’s as natural as a child’s curiosity.
In his book “A Mathematician’s Lament,” Paul Lockhart advocates an educational approach in which numbers are treated more concretely than usual: he asks us to imagine them as groups of rocks. For example, six corresponds to a group of rocks like this:
You
probably don’t see anything striking here, and that’s right — unless we
make further demands on numbers, they all look pretty much the same.
Our chance to be creative comes in what we ask of
them.
For instance, let’s focus on groups having between 1 and 10 rocks in
them, and ask which of these can be rearranged into square patterns.
Only two of them can: 4 and 9. And
that’s because 4 = 2 × 2 and 9 =
3 × 3; we get these numbers by “squaring” some other number (actually
making a square shape).
A less stringent challenge is to identify groups of rocks that can be neatly organized into rectangles with two rows that come out even. That’s possible as long as there are 2, 4, 6, 8 or 10 rocks; the number has to be “even.” All the other numbers from 1 to 10 — the “odd” numbers — always leave an odd bit sticking out.
Still, all is not lost for these misfit numbers. If we add two of them together, their protuberances match up and their sum comes out even; Odd + Odd = Even.
Yet when it comes to rectangles, some numbers, like 2, 3, 5 and 7, truly are hopeless. They can’t form any sort of rectangles at all, other than a simple line of rocks. These strangely inflexible numbers are the famous “prime” numbers.
So we see that numbers have quirks of structure that endow them with personalities. But to see the full range of their behavior, we need to go beyond individual numbers and watch what happens when they interact.
For example, instead of adding just two odd numbers together, suppose we add all the consecutive odd numbers, starting from 1:
1 + 3 = 4
1 + 3 + 5 = 9
1 + 3 + 5 + 7 = 16
1 + 3 + 5 + 7 + 9 = 25
The sums above, remarkably, always turn out to be perfect squares. (We saw 4 and 9 in the square patterns discussed earlier, and 16 = 4 × 4, and 25 = 5 × 5.) A quick check shows that this rule keeps working for larger and larger odd numbers; it apparently holds all the way out to infinity. But what possible connection could there be between odd numbers, with their ungainly appendages, and the classically symmetrical numbers that form squares? By arranging our rocks in the right way, we can make this surprising link seem obvious — the hallmark of an elegant proof.
The key is to recognize that odd numbers can make L-shapes, with their protuberances cast off into the corner. And when you stack successive L-shapes together, you get a square!
This style of thinking appears in another recent book, though for altogether different literary reasons. In Yoko Ogawa’s charming novel “The Housekeeper and the Professor,” an astute but uneducated young woman with a 10-year-old son is hired to take care of the Professor, an elderly mathematician who has suffered a traumatic brain injury that leaves him with only 80 minutes of short-term memory. Adrift in the present, and alone in his shabby cottage with nothing but his numbers, the Professor tries to connect with the Housekeeper the only way he knows how: by inquiring about her shoe size or birthday and making mathematical small talk about her statistics. The Professor also takes a special liking to the Housekeeper’s son, whom he calls Root, because the flat top of the boy’s head reminds him of the square root symbol, .
One day the Professor gives Root a little puzzle: Can he find the sum of all the numbers from 1 to 10? After Root carefully adds the numbers and returns with the answer (55), the Professor asks him to find a better way. Can he find the answer without adding the numbers? Root kicks the chair and shouts, “That’s not fair!”
But little by little the Housekeeper gets drawn into the world of numbers, and she secretly starts exploring the puzzle herself. “I’m not sure why I became so absorbed in a child’s math problem with no practical value,” she says. “At first I was conscious of wanting to please the Professor, but gradually that feeling faded and it had become a battle between the problem and me. When I woke in the morning the equation was waiting:
1 + 2 + 3 + … + 9 + 10 = 55
and it followed me all through the day, as though it had burned itself into my retina and could not be ignored.”
There are several ways to solve the Professor’s problem (see how many you can find). The Professor himself gives an argument along the lines we developed above. He interprets the sum from 1 to 10 as a triangle of rocks, with 1 rock in the first row, 2 in the second and so on, up to 10 rocks in the 10th row:
By its very appearance this picture gives a clear sense of negative space. It seems only half complete. And that suggests a creative leap. If you copy the triangle, flip it upside down and add it as the missing half to what’s already there, you get something much simpler: a rectangle with 10 rows of 11 rocks each, for a total of 110.
Since the original triangle is half of this rectangle, the desired sum must be half of 110, or 55.
Looking at numbers as groups of rocks may seem unusual, but actually it’s as old as math itself. The word “calculate” reflects that legacy — it comes from the Latin word “calculus,” meaning a pebble used for counting. To enjoy working with numbers you don’t have to be Einstein (German for “one stone”), but it might help to have rocks in your head.
NOTES:
As I hope I’ve made clear, this piece owes much to two books — one a polemic, the other a novel, both of them brilliant.
The rock metaphor and many of the other ideas and examples above have been borrowed from: Paul Lockhart, “A Mathematician’s Lament: How School Cheats Us Out of Our Most Fascinating and Imaginative Art Form” (Bellevue Literary Press, 2009).
The final example is from: Yoko Ogawa, “The Housekeeper and the Professor” (Picador, 2009).
For young readers who like exploring numbers and the patterns they make, see:
Hans Magnus Enzensberger, “The Number Devil: A Mathematical Adventure” (Holt Paperbacks, 2000).
For elegant but more advanced examples of visualization in mathematics, see:
Roger B. Nelsen, “Proofs without Words: Exercises in Visual Thinking” (Mathematical Association of America, 1997).
Thanks
to Carole Schiffman and Tim Novikoff for their comments and
suggestions, and to Margaret Nelson for preparing the illustrations.
Response to Comments:
Thanks to the many readers who posted such appreciative comments about the vision for this series.
Regarding some frequently asked questions:
The pieces are scheduled to appear each Monday, for a total of 15 in all.
An RSS feed for the series is available here.
Finally, a correction: in last week’s column I referred to 6 + 6 as an equation, but I should have called it an expression.
Sorry about that, and thank you to the readers who spotted this error.
Editors’ note: The name Hans Magnus Enzensberger was misspelled in the Notes section of an earlier version of this column; it has been corrected.
Need to print this column? Here is a print-friendly PDF version, with images.
177 Comments
The comments section is closed. To submit a letter to the editor for publication, write to letters@nytimes.com.
Jillian Spaulding
whittier January 9, 2015In my life I had never looked at math more than just a class to pass, and ofcourse I stil feel that way for the most part. However this article has shown me the different ways math can produce patterns and ways of thinking that differ in everyone.
Peter Lynn
New York, NY November 26, 2012Since this is still open for comment...
When I was about seven, my parents gave me the task of taking a carton of interlocking squares of a sort of interior Astroturf and assembling them into a rug. I started with one, then made a two by two square, then a three by three square, and so on. I noticed that I had to add the next odd number each time. My very first arithmetical pattern.
Lana Stephenson
Seattle January 19, 2012Why can't one be a square number?
jean
Berkeley, CA August 29, 2011Once again I am dazzled by your ability to make a subject which I have always avoided as much as possible seem magical, fun, and even understandable. However, there's one thing in this article which confuses me. Shouldn't you have said one through ten, instead of one to ten?
E. Means
Portland, OR June 2, 2011Melissa:
I'm with you -- I didn't see the difference between the primes and other odd numbers. After reading the other comments, however, it seems that the original author's full comment misled us. He said:
"A less stringent challenge is to identify groups of rocks that can be neatly organized into rectangles with two rows that come out even. "
His stated requirement of TWO ROWS was ignored by other contributors. The rectangles they formed have more than two rows.
Maybe you and I pay too much attention to the details ...
Nicholas
Boston, MA March 9, 2011I really have no idea why I feel the need to add my commentary to this thread, but I find this series to be captivating. Reading it during lunch seems to be having a negative impact on my daily work, however...
Santyjnu
New Delhi August 6, 2010Also to the mellisa's query... i think that only prime numbers(odd) don't make rectangle. All other odd numbers make rectangles.....to make a rectangle we need at least three multiples of a number(one itself, one, and other). I think even 2 which is an even prime number has some semblance of rectangle. after all nothing sticks out oddly from it.. :-)
RS
Seattle April 26, 2010I love your column and articles- thank you for presenting this in such a clear way to the most challenged of us but who love mathematics and its concepts. I had punitive teachers in grade school that made me have blinders on when it came to math. Yet I knew I loved doing math. I would love it if wouldyou present calculus concepts in such a clear way as well.
thank you
John
San Francisco April 19, 2010I like the column but can't figure out any shortcuts other than adding. The few procedures I've tried against other sequences do not check out. I've realized that I should not end on a prime number but it would be great if someone would just post the answers. Some people just have no sense of mathematics at all, even though we have successfully crossed the street for years.
akiwak
RI April 13, 2010Re: Pat F's (the "reluctant elitist) comment (#23):
My concern is not that today's Gauss would be sent back to his desk, but that even if his (or her) talent were recognized, were would we find a teacher to provide that "additional instruction and attention in mathematics"? Sadly, the Strogatzes of this world are not teaching middle school.
Smith
Fredericksburt, Texas March 16, 2010Wonderous stuff indeed! Thank you for thinking this way. I am coaching a 12 year old neighbor in math in exchange for yard work (his idea) and will teach him the Rock Trick today. He seems unable to grasp multiplication and this may help. Your "Square Dancing" seemed to oootch him into multiplication a bit - he is definitely a visual learner. I am searching for more ways to help him - rote memorization has not been the key. Any suggestions or other wonderous works?
cornelius
new york March 13, 2010Melissa,
I think you actually have a point. If you look just above reference to prime numbers you see 5 rocks and 7 rocks making 2 x 6 rectangle. The primeness of the numbers is a separate matter from odd plus odd becomes even. Primeness does have something to do with arranging stones into rectangles in that the columns of stones much have equal numbers. That is why 15, while odd, is not prime.
A prime number does not admit this potential, it is a number divisible only by itself and one. In this detail the discussion was confusing and I wish more had been said about prime numbers if in fact more is to be said that the simple assertion I made above.
That said, the entire enterprise is simply wonderful.
S.M.
Tucson March 10, 2010In response to Mr. Bill
(3) I pose a related problem for us all that I have used teaching mathematics to people ranging from 2nd-grade children to doctoral students in mathematics education: Please figure out how many gifts in all my true love gave to me during the twelve days of Christmas.
My husband and I once discussed this problem on a 5 hour drive back from New Mexico. The problem wasn't the sum, it was the phrasing of the song! We argued (in a friendly way) about whether each day, all of the previous days' gifts were included, or whether each day a unique gift was given. It makes a big difference! When we got home, we researched the song, and agreed that each day, all of the previous days' gifts were included. The answer: that is a lot of gifts!
It's all in fun. I am delighted by the column, as well as the positive response to it. This is the first on-line column that I actually read many of the comments posted by the readers and find them intelligent and fascinating.
Dr. Strogatz gave a talk at the University of Arizona when I was a math student there a couple of years ago. It was a pleasure to attend his talk, and it is great to share in so many readers' delight in this column.
Who knew math could be so fun? ;)
alphachapmtl
Boston March 8, 2010"brain injury that leaves him with only 80 minutes of short-term memory"
Reminds me of "The Man with the 7 Second Memory" .
Clive Wearing developed a profound case of total amnesia as a result of his illness.
The story of Clive Wearing was told in a 2005 ITV documentary.
http://en.wikipedia.org/wiki/Clive_Wearing
http://www.youtube.com/watch?v=WmzU47i2xgw
Subramanian
Bombay March 8, 2010The area of an isosceles (the length of the sides forming the right angle being equal) right triangle is 1/2 x b x h. Thus the area of a rt. triangle with the equal sides of 10 is 1/2 x 10 x 10 = 50 not 55.
What gives???
JS
USA March 4, 2010I love this series, but I have encountered a problem. I want to share it with my mother who, being 82, would have a hard time reading it on the computer. I tried to print it out with the print option at the end of the article, but that option does not include the illustrations of the rocks which I feel is integral to understanding the article. So I printed the whole webpage out and it required 21 pages, only 5 of which were necessary to print the substance of the article. I wish someone would add the illustrations to the print option so I can print the rest of the sessions without using a ream of paper.
Basu
India March 3, 2010There is a story/legend that a class of young students was asked by the teacher to add up all the numbers from 1 to 100. The teacher thought he could relax for a while.
Unfortunately, a student quickly came to him with number scrawled on his little black board. The kid has realised he just needed to add 1 and 100, 2 and 99, etc 50 times. It was the start of the mathematical life of Gauss!
My school teacher told us this story before he started the proof, and I recently read it in the remarkable:
Measuring the World
thayward
New Hampshire February 25, 2010I know this is a late post, however I am a middle school math teacher and I have a snow day here in NH and thus time to catch up on this series. I noticed that 10! is five-five(55) and 100! is fifty -fifty(5050) from the talented Gauss and so had to immediately check 1000! and what do you imagine happened?? Yes five hundred - five hundred (500,500) Nothing like a great mathematical pattern to start the day!
David Bee
Brooklyn February 19, 2010Poster 127 writes:
"For successive sums solve the difference equation: S(n+1) - S(n) = n + 1. Difference equations are the easiest way to get sums of squares, cubes, etc."
Let's try it for the sum of the squares of the first n natural numbers: 1^2 + 2^2 + 3^2 + ... +n^2
Thus, we have S(1) = 1^2 = 1, S(2) = 1^2 + 2^2 = 5, S(3) = 1^2 + 2^2 + 3^2 = 14, ... , S(n) = ???
Using Poster 127's suggestion, the corresponding difference equation would be
S(n+1) - S(n) = (n+1)^2 = n^2 + 2n + 1
Letting S(n) = an^3 + bn^2 + cn + d, we get (after some cancellation),
S(n+1) - S(n) = a(3n^2 + 3n + 1) + b(2n+1) + c
Setting this equal to n^2 + 2n + 1 and equating corresponding coefficients yields
a= 1/3, b=1/2, and c=1/6
Thus, substituting above, we have
S(n) = (1/3)n^3 + (1/2)n^2 + (1/6)n = (n/6)(2n^2 + 3n + 1) = (n/6)(n+1)(2n+1)
and so
1^2 + 2^2 + 3^2 + ... + n^2 = (n/6)(n+1)(2n+1)
as expected by the math-oriented posters.
Generally, difference equations follow from "finite calculus" (or the calculus of finite differences), which, as a college course, is sort of a "backwater" one that few students take. Correspondingly, there are courses in differential equations, with probably most math majors taking a first course in such.
Despite the usefulness of difference equations (even though probably most math majors aren't familiar with them), Poster 127 may be overdoing it when he writes such are the "easiest way to get sums of squares, cubes, etc." because the above derivation requires some background in solving difference
(or differential) equations. [Note: If I am off-base with this comment, then please feel free to comment...]
For example, let's re-derive the formula for the sum of 1^2 + 2^2 + 3^2 + ... + n^2:
Since (k+1)^3 = k^3 + 3k^2 + 3k + 1, we have
3k^2 + 3k + 1 = (k+1)^3 - k^3
Summing both sides from k=1 to k=n, we have
3SUM(k^2) + 3SUM(k) + SUM(1) = SUM(k+1)^3 - SUM(k^3)
Solving for SUM(k^2), we have
SUM(k^2) = (1/3)[SUM(k+1)^3 - SUM(k^3) - 3SUM(k) - SUM(1)]
= (1/3)[(n+1)^3 - 1 - 3n(n+1)/2 - n]
= (1/6)(n+1)[2(n+1)^2 - 3n - 2]
= (1/6)(n+1)[2n^2 + n]
= (n/6)(n+1)(2n+1)
Finally, for those non-math-oriented ones reading this far, let's use this to quickly find the sum of the
squares of 1,2,3,...,98,99, and 100 quickly:
1^2 + 2^2 + 3^2 + ... + 98^2 + 99^2 + 100^2 = (100/6)(100+1)(2(100)+1) = (100/6)(101)(201)
= 338 350
Margaret
Dallas, TX February 16, 2010@ekwas #153 - when you combine prime numbers, they certainly are able to form rectangles, as you demonstrated. However, the only complete rectangle a prime number can form on its own is a line. I believe that is the intent of the sentence you quoted.
Take the number 7:
You can have three rows of two with a leftover:
* *
* *
* *
*
or two rows of three with a leftover:
* * *
* * *
*
but the only way to make a rectangle is one row of seven.
* * * * * * *
Mwendah
Bowling Green, OH February 15, 2010This is great stuff. If all media--including TV and radio--could do something like this, then heaven would be here.
Reader in Houston
Houston, TX February 15, 2010Love this column!
mibd
Atlanta February 15, 2010Euclid--- and one would presume, the Greeks who preceded him--- looked at arithmetic this way, that is, geometrically.
ekwas
usa February 15, 2010"Yet when it comes to rectangles, some numbers, like 2, 3, 5 and 7, truly are hopeless. They can’t form any sort of rectangles at all, other than a simple line of rocks. These strangely inflexible numbers are the famous “prime” numbers."
I didn't get this one either, because it seems that,
- 3 and 5 form a rectangle of 2x4=8 xxoo
xooo
- 3 and 7 form a rectangle of 2x5=10 xxooo
xoooo
- 5 and 7 form a rectangle of 2x12=12 xxxooo
xxoooo
ekwas
usa February 15, 2010"Yet when it comes to rectangles, some numbers, like 2, 3, 5 and 7, truly are hopeless. They can’t form any sort of rectangles at all, other than a simple line of rocks. These strangely inflexible numbers are the famous “prime” numbers."
I didn't get this one either, because it seems that,
- 3 and 5 form a rectangle of 2x4=8;
- 3 and 7 form a rectangle of 2x5=10;
- 5 and 7 form a rectangle of 2x12=12.
177 Comments
The comments section is closed. To submit a letter to the editor for publication, write to letters@nytimes.com.