The History of Zero

Nicholas Baronowsky

2021

1
Introduction

Zero is an elusive concept. While every early culture developed some basic written math-
ematics for practical purposes like accounting or construction, and some even began to
consider what would now be called “pure” math, the modern conception of zero took much
longer to manifest. Part of this is likely due to the numerous roles zero plays. First and
foremost, it can be considered as a reference point from which to start counting, or as an
absence to be filled in later. It is also used a placeholder in positional numeral notation, both
to indicate that a place value should be skipped and also (if used at the end of a number)
to indicate an order of magnitude. Most cultures doing math for practical purposes had
these ideas in some form or another, but they may not have connected them with the same
symbol. The most elusive role of zero, however, is that of a number itself, with which any
operation can be done. This concept was much more rare and took hundreds of years after
the previous two to develop. Understanding zero as a number that is subject to the same
operations as more obvious numbers like integers is a sign that a culture is truly studying
mathematics for its own sake, and not for any practical purpose. The steps to get to this
point take thousands of years and can be mapped through many different cultures.

Early Ideas

Egypt

Figure 1: The Egyptian symbol for zero.

The first recorded use of a zero symbol is from ancient Egyptian sources. A balance sheet
from around 1770 BCE uses the hieroglyph nfr (shown above), which meant “good” or
“beautiful”, to refer to values of zero balance. Later on, some architectural drawings use
the same symbol as a reference point for ground level [2]. There is no evidence showing that

2
this symbol was included in the writing of other numbers, meaning the Egyptians thought
of it as totally separate from true numbers and didn’t consider its use as a placeholder to
indicate orders of magnitude. However, Egyptian numerals were not positional, so there
was no need for them to have such a symbol.

Babylon

Figure 2: The Babylonian symbol for zero [3].

Some of the first examples of a placeholder symbol come from between the 6th and 3rd
centuries BCE, in ancient Babylon. Clay tablets from that period begin to include a number
of different symbols distinct from the already existing numerals for 1 and 10, one of which
is shown above. While the Babylonian notation was positional, previously the only way
to distinguish the position of a numeral was an ambiguous space between it and the next.
The addition of a placeholder made it clear that a position was to be skipped, allowing an
obvious distinction between numbers like 22 and 202, for example. However, this symbol
was still never used at the end of numbers, meaning that the actual scale still had to be
inferred from context [3].

3
Greece

Figure 3: A table of angles from Ptolemy’s Almagest [6].

This table from Ptolemy’s Almagest (c. 150 CE) is his calculation of inclination angles for
eclipses. The Almagest is largely an astronomical text, not a mathematical one, so one
would not expect to find calculations involving zero. However, as this table shows, Ptolemy
did at least use a zero symbol as a starting point in angle measurements. This is similar
to the way the Egyptians used zero, and it was notated in various ways as an “o” symbol
with a bar. The exact shape is not standardized, suggesting that it was only thought of
as a placeholder and not a number in its own right. However, the zero symbol’s extensive
inclusion in Ptolemy’s angle tables marks the limit to which Mediterranean scholars were
willing to treat it [6, 319].

4
India

Figure 4: An excerpt from the Bakhshali manuscript [4].

In this excerpt from the Bakhshali manuscript, dating from at earliest the 3rd century CE,
multiple ways of calculating a sum to 377 are shown. Each different way of summing contains
90
one of the previous numbers divided by an increasing quantity: the first is 120 + 3 + 80 +
75+72, then 120+90+ 80
4 +75+72, 120+90+80+ 75
5 +72, and finally 120+90+80+75+ 72
6 .

In this case, zero is being used to denote a number’s scale: the difference between 12 and
120 (this is the role the Babylonian placeholder zero was missing). The appearance of both
numbers in this excerpt shows that by this point a zero symbol was in use in India, and
indeed the Bakhshali manuscript is the earliest known source of one [4].

5
Further Indian Efforts - Brahmagupta and Bhaskara

Brahmagupta

The first recorded evidence of zero’s treatment as an actual number comes from the Brah-
masphutasiddhanta by the Indian mathematician Brahmagupta, dating to the 7th century
CE. In Chapter 18, Section 2, he lays out the rules for basic arithmetic operations involving
positive (“affirmative”) and negative numbers, and a quantity translated by Colebrooke as
“cipher” or “nought”, which is the number zero. The following excerpt contains only the
lines pertaining to this quantity.

31. The sum of cipher and negative is negative; of affirmative and nought is
positive; of two ciphers is cipher.

32-33. Negative, taken from cipher, becomes positive; and affirmative, becomes
negative. Negative, less cipher, is negative; positive, is positive; cipher, nought.

34. The product of cipher and negative, or of cipher and affirmative, is nought;
of two ciphers, is cipher.

35-36. Cipher, divided by cipher, is nought. Positive, or negative, divided by

cipher, is a fraction with that for denominator: or cipher divided by negative or
affirmative.

The square of negative or affirmative is positive; of cipher, is cipher [1, 339-340].

This section is remarkably thorough in its examination of zero as a number. Brahmagupta

correctly defines the properties, for a a real number:

0 + (−a) = −a a+0=a 0+0=0 0 − (−a) = a 0 − a = −a

−a − 0 = −a a−0=a 0−0=0 (−a) · 0 = a · 0 = 0 0·0=0

These are expressed simply today by saying that zero is the additive identity, and that
0
any number multiplied with zero is itself zero. Interestingly, he also defines 0 = 0, which
now would be considered invalid. Furthermore, Brahmagupta allows for division of regular
numbers by zero, but only defines them as a fraction with zero in the denominator. He

6
leaves zero divided by any nonzero number defined in the same way, as a fraction with zero
in the numerator. Finally, he again correctly defines 02 = 0.
Brahmagupta’s text represents an important milestone in the understanding of zero. For
the first time in history, it is treated in the same way as integers, with specific rules about
how to add, subtract, multiply, and divide with it. Perhaps even more impressive is that
almost all the definitions Brahmagupta writes are correct in the modern view. The only false
claims he makes are those involving division of and by zero, operations which confounded
mathematicians until well into recent history.

Bhaskara

Writing some 500 years later, another Indian mathematician named Bhaskara would clarify
and extend Brahmagupta’s treatment of zero. In Chapter 2, Section 4 of his arithmetic text
Lilavati, he sets out the same rules as his predecessor in a more concise way.

In addition, cipher makes the sum equal to the additive. In involution and
[evolution] the result is cipher. A definite quantity, divided by cipher, is the
submultiple of nought. The product of cipher is nought: but it must be retained
as a multiple of cipher, if any further operation impend. Cipher having become
a multiplier, should nought afterwards become a divisor, the definite quantity
must be understood to be unchanged. So likewise any quantity, to which cipher
is added, or from which it is subtracted, [is unaltered.] [1, 19]

In modern notation, these rules would be written as

√ a a
0x = 0
x
a+0=a 0=0 =
0 0
(a · 0)
a · 0 = 0a =a a±0=a
0
Bhaskara’s addition here is that “involution and [evolution]” - exponents and roots in modern
terms - leave zero unchanged. This extends Brahmagupta’s work, which only mentioned that
the square of zero is zero. Interestingly, products of zero are defined correctly, but Bhaskara
cautions the reader to carry zero as a multiple through any calculation, since the zero may
later be divided out. Again, the rules regarding division by zero are unclear, but Bhaskara
returns to this issue later.

7
 The rest of the section is a collection of example problems dealing with zero.

Example. Tell me how much is cipher added to five? and the square of cipher?
and its square root? its cube? and cube-root? and five multiplied by cipher? and
how much is ten, subtracting cipher? and what number is it, which multiplied by
cipher, and added to half itself, and multiplied by three, and divided by cipher,
amounts to the given number sixty-three? [1, 19-20]

In modern notation, the first few of these would be written with their answers as 5 + 0 = 0,
√ √
02 = 0, 0 = 0, 03 = 0, 3 0 = 0, 5 · 0 = 0, and 10 − 0 = 0. The last question is a longer
algebraic equation involving the aforementioned multiplication and later division by zero.
In modern notation, it would be written as:
 
x·0 x
+ · 3 = 63
0 2
The answer is given correctly as 14.
Bhaskara reiterates these rules more carefully in Chapter 1, Section 3 of Vija-Ganita,
the volume after Lilavati which pertains more to algebraic problems. In this section, the
additive/subtractive and multiplicative/divisive rules regarding zero are each given specific
example problems, and it is here where division by zero is defined to be infinite.

In this quantity consisting of that which has cipher for its divisor, there is no
alteration, though many be inserted or extracted; as no change takes place in
the infinite and immutable God, at the period of the destruction or creation of
worlds, though numerous orders of beings are absorbed or put forth. [1, 138]

This definition significantly expands that given by Brahmagupta. By modern standards,

it is still incorrect, but it is close to something that works. This definition seems natural
when considering that fractions get larger without limit as their denominator decreases,
which Bhaskara and his contemporaries were no doubt aware of. He acknowledges in the
conclusion to Vija-Ganita that his text is a distillation of works by earlier authors, including
Brahmagupta - thus, it is possible that his extended definitions regarding zero are due to
an earlier author whose text has not survived. Nevertheless, the works of Bhaskara and
Brahmagupta provide the earliest considerations of zero as a number with which calculations
could be done, an idea that likely made its way back to the West along with Indian numerals.

8
Back to Europe - Fibonacci

The current common usage of Hindu-Arabic numerals in the West can be traced to a monu-
mental book called Liber Abaci by the Pisan mathematician Fibonacci, dating to the early
13th century. This is not so much a mathematical treatise as it is a manual on how to use
and calculate with the ”Indian figures” as he calls them. In the first chapter, Fibonacci
explains what they are and how to write numbers with them.

The nine Indian figures are:

9 8 7 6 5 4 3 2 1.

With these nine figures, and with the sign 0 which the Arabs call zephir any
number whatsoever is written, as is demonstrated below [5, 17].

He then explains the decimal positional system, grouping numbers from left to right with
their position relative to the leftmost denoting the power of ten by which they are multiplied.
He then gives some examples of numbers written this way, and clearly states the use of zero
as a placeholder.

Truly if so much as five hundred you will wish to write, in the first and in the
second place you will put the zephir, and in the third the figure five, in this way,
500; and you will be able to write any number of hundreds with two zephir. And
if you will wish to write hundreds with tens or units, then you put in the first
place the zephir, in the second tens, and in the third the hundreds that you will
wish. For example, if in the first place is the zephir, and in the second the figure
nine, and in the third the figure two, then 290 will be denoted. If indeed you
will wish to write hundreds with units and without tens, you put in the second
place, namely in the place of the tens, the zephir, and in the first the number of
units that you will wish, and in the third, the figure two, 209; [5, 18]

Fibonacci’s quest to replace Roman numerals with Hindu-Arabic ones was successful, largely
thanks to his demonstrations of calculation with them which are found later in Liber Abaci.
However, as mentioned before, the book is not focused on the math but rather the utility

9
of calculations. As such, Fibonacci includes no specific consideration of zero as a number.
However, he does seem to understand at least some of the rules established by the Indians
from which the numerals came.

Also if one will wish to subtract 80 from 392, then he puts the 80 beneath the
392, and takes the 0 from the 2; there remains 2 which he puts...[5, 46]

This description of subtraction implies that Fibonacci understands zero’s role as an addi-
tive/subtractive identity, since he instructs that 2 − 0 = 2. Similar examples can be found
regarding addition and multiplication with zero. He forbids subtraction of greater numbers
from lesser in his steps, but since the book is meant to teach practical calculation to mer-
chants the omission of negative numbers is understandable. He never considers division by
zero, suggesting that this question was still unresolved in the mathematical community at
the time. Nevertheless, it appears that after this point, the zero symbol still used today was
finally established as a placeholder and a number.

Conclusion

The path to the modern idea of zero is long and winds itself through the earliest civilizations,
to India and back to Europe. At each step, a little more understanding is gained. Sources
from ancient Egypt, Babylon, Greece, and India show ideas of zero as a reference point
and a placeholder, but no considerations of it as a number in its own right. For this, the
works of Brahmagupta and Bhaskara provide the earliest definitions and rules to perform
elementary operations with zero. Eventually, along with the standard numerals used today,
these ideas made their way through the Islamic empire back to Europe through the efforts
of Fibonacci. From there, both the modern conceptions and symbol for zero were set. It is
to these great mathematicians and every other scholar in ancient civilizations that credit is
due for defining this unique number.

10
References

[1] H. T. Colebrooke, editor. Algebra, with Arithmetic and mensuration, from the Sanskrit
of Brahmegupta and Bhascara. J. Murray, London, 1817. URL [Link]
[Link]/discovery/openurl?institution=01ROCH INST&r
fr id=info:sid%2Fsummon&rft dat=ie%3D21161721930005216,language%3DEN&svc
dat=CTO&[Link] date coverage=true&vid=01ROCH INST:Services&Force direc
t=false.

[2] G. G. Joseph. The Crest of the Peacock: non-European Roots of Mathematics. Princeton
University Press, 2011.

[3] R. Kaplan. The Nothing That Is: A natural history of zero. The Penguin Press, 1999.

[4] G. R. Kaye, editor. The Bakhshālı̄ Manuscript. a Study in Mediæval Mathematics.

Government of India, Central Publication Branch, 1927. URL [Link]
details/[Link].2015.23309/page/n15/mode/2up.

[5] L. E. Sigler, editor. Fibonacci’s Liber abaci : a translation into modern English of
Leonardo Pisano’s Book of calculation. Springer, New York, 2002. URL https://
[Link]/discovery/openurl?instituti
on=01ROCH INST&rfr id=info:sid%2Fsummon&rft dat=ie%3D2119352989000521
6,language%3DEN&svc dat=CTO&[Link] date coverage=true&vid=01ROCH INST:
Services&Force direct=false.

[6] G. J. Toomer, editor. Ptolemy’s Almagest. Princeton University Press, Princeton, N. J.,
1998. URL [Link]
nurl?institution=01ROCH INST&rfr id=info:sid%2Fsummon&rft dat=ie%3D211966
64710005216,language%3DEN&svc dat=CTO&[Link] date coverage=true&vid=01R
OCH INST:Services&Force direct=false.

11