ALGORITHMS: Through the Ages and Around the World

Joyce F. Fischer
Jean F. Davis

Webster's Collegiate Dictionary defines an algorithm as "a rule of procedure

for solving a mathematical problem that frequently involves repetition of an
operation." (p. 22) The word “algorithm” comes from the name of the 8th
century Arab mathematician Mohammed al-Khowarizmi. He wrote two
books, one about arithmetic and one about algebra, that explained the Hindu-
Arabic numerals and methods of calculation. Readers of the Latin
translations began to attribute the system to him and to call the new methods
of calculation “algorismi.” Over time the word became “algorithm,” and
came to mean an orderly, repetitive scheme. (Long, p. 186)

The phrase “alternative algorithm” refers to procedures different from those

currently taught in most public school classrooms in the United States.
Standard or traditional algorithms provide a first look at the four basic
operations. As the numbers become very large or more difficult, alternative
algorithms provide an interesting and generally, logically diverse, method of
approach for students who are having trouble learning the basic school
algorithms. Many alternative algorithms are found in the public domain and
additional methods are being devised each day. Some references to
Mathematics Education texts that contain various alternative algorithms are
cited at the end of this paper.

Here are some reasons why teachers should teach and use alternative
algorithms in their classrooms:

1. Alternative algorithms accommodate different learning styles.

2. Alternative algorithms demonstrate that there IS more than just one
way to solve a problem.
3. Alternative algorithms are fun!
4. Alternative algorithms provide a means so that we can appreciate
the efforts of other people in other times and places.
Texas State University Joyce Fischer & Jean Davis

Addition

Adding by Place Values

This procedure uses a “left-to-right” approach (Bennett, p. 141), which was
used by the early Hindus and later by the Europeans. Add the number in the
highest place value first, then each lower place value until you reach the
ones place value.

124

+ 392
} 100 + 300

400
+

+
20 + 90

110
+

+
4 +2

516

Across, Down, and Across

Here is a different addition algorithm, which uses the expanded form of a
number and the distributive law to simplify computation. Write each
number in expanded form horizontally first. Then add the numbers in the
appropriate place values (vertically). Then add these numbers horizontally.

124 = 1 x 100 + 2 x 10 +4x1

+ 392 = 3 x 100 + 9 x 10 +2x1
4 x 100 + 11 x 10 +6x1

400 + 110 + 6 = 516

Texas State University Joyce Fischer & Jean Davis

Count and Strike: Base Ten

This algorithm gives us a way to keep track of groups of the base without
having to remember multi-digit numbers. Starting at the top right place
value (the ones), add down until you get one group of the base or more.
Make a strike and write the remainder small and to the right of the number
that was struck out. At the very bottom of the column write the last
remainder. Count the strikes and put that number at the top of the next place
value. Always start counting at the top. Keep going until there are no more
place values. This method works well for non-standard bases.

Base Ten Problem Base Ten Problem rewritten with

more space for clarity
1 1 2
1124 1 1 2 4
1321 1 3 2 1
1109 1 1 0 94
204 2 0 4
+ 392 + 31 95 20
4 1 5 0

Count and Strike: Other Bases

This algorithm also translates easily to addition and multiplication (repeated

addition) in other bases. For other bases, the number in the group (the base)
is the only thing that changes.

Base Five with more space Base Eight with more space
for clarity. for clarity.

1 1 1 2 1 2 1 2
1 1 1 4 2 3 1 5 4
2 1 1 32 2 72 1 21 3
10 1 21 1 1 60 1 4 54
+ 1 10 1 20 + 32 1 2 1 73
1 1 0 2 0 1 2 1 6 6 3
Texas State University Joyce Fischer & Jean Davis

Subtraction

Equal Addition

The equal addition method is taught in schools in Australia and some Latin American
countries. It is also sometimes taught in the northeastern part of the United States.
Add the same number to the minuend and the subtrahend, but in a different form, then
subtract. This method never involves borrowing.

Original Problem New Problem

with space with space

1
3 1 Add ten ones 3 1
3

- 2 9 Add one ten - 2 9

2

1
2 0 7 Add ten tens 2 0 7
2

- 1 4 2 Add one hundred - 1 4 2

6 5

If more than one place value needs to be adjusted, do all of them at once, then subtract.

1 1
2 1 1 Add ten ones, Add ten tens 2 1 1
1 3

- 2 3 Add one ten, Add one hundred - 2 3

1 8 8
Texas State University Joyce Fischer & Jean Davis

Austrian Method for Subtraction

The Austrian method is really the same idea as Equal Addition, but written in a different
horizontally oriented form.

Consider the problem 764 – 348.

764 – 348 = (700 + 60 + 4) – ( 300 + 40 + 8)

= (700 + 60 + 10 + 4) – (300 + 40 + 10 + 8) Add 10

= (700 + 60 + 14) – (300 + 50 + 8)

= (700 – 300) + (60 – 50) + (14 – 8)

= 400 + 10 + 6

= 416

In the second step, we added 10 to each number. In the third step, the 10 was combined
with the 4 ones in the minuend and with the 4 tens in the subtrahend, making the
subtraction possible.

Here is another example: 3002 – 1875

3002 – 1875
= (3000 + 2) – (1000 + 800 + 70 + 5)
= ( 3000 + 1000 +100 + 10 + 2) – (1000 +1000 + 800 + 100 + 70 + 10 + 5)
= (3000 + 1000 + 100 + 12) – ( 2000 + 900 + 80 + 5)
= (3000 – 2000) + (1000 – 900) + (100 – 80) + (12 – 5)
= 1000 + 100 + 20 + 7
= 1127
Texas State University Joyce Fischer & Jean Davis

Method of Complements for Subtraction

This method for subtraction is used in Europe. With this algorithm, we start by adding
some power of the base in the middle of the problem then we subtract it out at the end of
the problem. Beginning with the middle number, subtract the digit below this number and
then add the digit above this number. If you need to carry, add the carried number to the
top number. The leftmost digit of the answer (difference) will always be a one. Cross it
out and you have the difference. Here are several examples.

Original Problem New Problem

two digits with more space
1
5 3 5 3 ones column 10 – 7 + 3
1
- 2 7 9 0 tens column 9 – 2 + 5
- 2 7 hundreds column bring down
1 2 6 the carried one
Cross out the one in the leftmost column. The final answer is 26.

Original Problem New Problem

four digits with more space
1
3 0 0 2 3 0 0 2 ones column 10 – 5 + 2
1
- 1 8 7 5 9 9 9 0 tens column 9 – 7 + 0
- 1 8 7 5 hundreds column 9 – 8 + 0
1 1 1 2 7 thousands column 9 – 1 + 3
ten thousands column bring
down the carried one
Cross out the one in the leftmost column. The final answer is 1127.

Original Problem New Problem

multi-digit with more space
1
1 5 3 1 1 5 3 1 ones column 10 – 5 + 1
1
- 1 4 5 9 9 9 0 tens column 9 – 4 + 3
- 1 4 5 hundreds column 9 – 1 + 5
1 1 3 8 6 thousands column 9 – 0 + 2
ten thousands column bring
down the carried one
Cross out the one in the leftmost column. The final answer is 1386.
Texas State University Joyce Fischer & Jean Davis

Multiplication

Egyptian Multiplication
Our insight into the arithmetic of the ancient Egyptians comes primarily from the Rhind
Papyrus. (Burton, p. 32) Written about 1650 B.C., it contains numerous multiplication and
division problems. A cumbersome numeration system necessitated performing
multiplication through repeated additions. This method for multiplication reduces the
process to doubling and adding. It is a type of Duplation. Beginning with 1; make a chart,
like the one below, by doubling the number in the left column each time. When two times
the number in the left column is greater than the first factor, STOP!! Find the numbers in
the left column that add up to the first factor. Add the corresponding numbers in the right
column. This method does not translate well to other bases for beginning students.

Original Problem: 26 X 42

1 X 42 = 42

2 X 42 = 84

4 X 42 = 168

8 X 42 = 336

16 X 42 = 672

Since 2 x 16 = 32 which is greater than 26, STOP!!

26 = 16 + 8 + 2 so add 672
336
+ 64_
1092

Therefore, the product of 26 and 42 is 1092, by Egyptian Multiplication.

Texas State University Joyce Fischer & Jean Davis

Lattice Multiplication

Lattice multiplication was introduced into Europe in the 13th century by the mathematician
Leonardo of Pisa, more commonly known as Fibonacci. The lattice multiplication
algorithm appeared in the very first printed arithmetic book, in Treviso, Italy in 1478.
(Sonnabend, p. 170)

Let’s look at the procedure using 172 × 93 as an example.

1) Start with a rectangular grid having the same number of columns (3) as the number of
digits in one factor (172), and the same number of rows (2) as the number of digits in
the other factor (93). Place the digits of one factor across the top and the digits from
the other factor down the right side. Then draw diagonals from upper right to lower left
to create the “lattice” design.

2) Multiply the corresponding digits and enter their products into the lattice as shown
below, the tens digit going in the upper triangular region and the ones digit in the lower.
Texas State University Joyce Fischer & Jean Davis

3) Finally, add along the diagonals, starting in the lower right corner, moving left, and
carrying to the next diagonal when necessary.

The answer is read, following the arrow, starting

at the upper left.
172 × 93 = 15,996

Two other examples:

The advantage of this method is that all the multiplying is done first, then the adding. It is
an excellent way to organize the multiplication of numbers with 3 or more digits each. It
also translates wonderfully to non-decimal bases.
Texas State University Joyce Fischer & Jean Davis

Russian Peasant Multiplication

The so-called “Russian Peasant” method, frequently seen in textbooks for pre-service
teachers (Long, p. 208), is the ancient Egyptian “duplation and mediation” procedure. It
has been renamed because of its use by Russian peasants, even into the 20th century.
(Burton, p. 36) The advantage of this method is that the only multiplication facts we need
to know are how to double and how to take half of a number.

Consider the following problem: 28 X 43

Take half of one factor (disregard remainders) and double the other factor.

Take half Double

28 X 43
Halve 28 14 86 Double 43
7 172
3 344
1 688
When we reach 1 in the halves
column, we’re through.

To get the answer, add the numbers in the doubles column that are opposite ODD numbers
in the halves column:

172 + 344 + 688 = 1204

While this might be a tedious method to have as our only way to multiply, there are certain
problems for which it is quite nice.

Take half Double Take half Double

32 X 25 16 X 42
16 50 8 84
8 100 4 168
Even though we’re not 2 336
technically finished, we know the 1 672
answer! Since the only odd number in the halves
column is 1, the last number in the doubles
column is the answer!
 Texas State University Joyce Fischer & Jean Davis

Division

Egyptian Division
This method for division reduces the process to doubling and subtracting. It is the inverse
of Egyptian multiplication. The divisor is doubled until the dividend is reached and the
quotient is the missing factor. Beginning with 1, make a chart, like the one below, of
doubles of the divisor. When the number in the right column exceeds the dividend, STOP!!
Find the numbers in the right column that add up to the dividend. When the remaining
number that you need to add is strictly less than the divisor, record it as the remainder.
Add the corresponding numbers in the left column to produce the quotient. This method
does not translate well to other bases for beginning students.

Original Problem: 352 divided by 9 or 9 352

1 X 9 = 9

2 X 9 = 18

4 X 9 = 36

8 X 9 = 72

16 X 9 = 144

32 X 9 = 288

64 X 9 = 576

Since 64 x 9 = 576, which is greater than 352, STOP!!

351 = 288 + 36 + 18 + 9. Add the number of 9's. 32
4
2
+ 1
39
!
Thus, 39 X 9 = 351 so there is a remainder of 1. Therefore, when 352 is divided by 9, the
quotient is 39 and the remainder is 1, by Egyptian Division.
Texas State University Joyce Fischer & Jean Davis

Checking the Computational Form Yields:

288 = 32 X 9
+ 36 = + 4 X 9
+ 18 = + 2 X 9
+ 9 = + 1 X 9
351 39 Total

But we need 352 so we must add 1 to the 351. The quotient is 39 and the remainder is 1.

Verifying the Results Using a Different Computational Form:

9 352
– 288 32 X 9
64
– 36 4 X 9
28
– 18 2 X 9
10
– 9 1 X 9
1

Since 1 is strictly less than 9 it is the remainder.

Add 32 + 4 + 2 + 1 = 39 which is the quotient and the remainder is 1.

Therefore the quotient of 352 divided by 9 is 39 plus a remainder of 1, determined by

Egyptian Division and verified by computational forms.
Texas State University Joyce Fischer & Jean Davis

Repeated Subtraction (Scaffolding) Division Algorithm

Consider the problem: 136 ÷ 6

We are literally asking ourselves the question

“How many times can 6 be subtracted from 136?

6 136
-60 10 We have subtracted
76 6 ten times
-60 10
16
-6 1
10
-6 1
4 22 We have subtracted
6 a total of 22 times

The quotient is 22 and the remainder is 4. 136 = 6(22) + 4

Another example: 496 ÷ 16

16 496
-160 10
336
-160 10
176
-160 10
16
-16 1_
Remainder 0 31 Quotient

The quotient is 31 and the remainder is 0 or using the standard form:

496 = 31(16) + 0
Texas State University Joyce Fischer & Jean Davis

Many students are clever enough to construct their own algorithms for the

basic operations of addition, subtraction, multiplication, and division. It is, of

course, the responsibility of the teacher to make sure that these algorithms will work

consistently. For those students who have trouble learning the standard algorithms

or who just want to try some different ones for variety, we recommend these as a

way to engage the students in the process of learning mathematics in an active fun

environment. As is always true with mathematical or real life problem solving,

showing a variety of methods yields creative adept learners who are willing to

approach any challenge fearlessly.

References:

Bennett, Albert B. and Nelson, L. Ted (2004). Mathematics for Elementary Teachers: A

Conceptual Approach. New York: McGraw-Hill.

Burton, David M. (2003). The History of Mathematics, 5th Edition. New York:

McGraw-Hill.

Long, Calvin T. and DeTemple, Duane W. (2000). Mathematical Reasoning for

Elementary Teachers, 2nd Edition. Reading, Massachusetts. Addison Wesley Longman.

Sonnabend, Thomas. (2004). Mathematics for Teachers, 3rd Edition. Belmont,

California. Brooks Cole.

Webster's Seventh New Collegiate Dictionary. (1965). Springfield, Massachusetts. G. &

C. Merriam Company.