Abstract
In Mayan mathematics, zero is supposed to be, in some sense, equal
to infinity. At first glance, while this statement may have a deep philosophical
meaning, it does not seem to make much mathematical sense.
In this paper, we show, that this statement may be made mathematically
reasonable. Specifically, on a real line, it is often useful to consider
both −∞ and +∞ as a single infinity. When we deal with very small
and very large numbers, it makes sense to use floating point representation,
i.e., in effect, consider logarithms of the original values. In terms of
logarithms, the original value 0 corresponds to −∞, while the original
infinite value corresponds to +∞. When we treat both possible values
−∞ and +∞ as a single infinity, we thus treat the original values 0 and
∞ as similar.
Mathematics Subject Classification: 01A12
Keywords: Mayan mathematics, infinity, zero
6194 O. Kosheleva and V. Kreinovich
1 Formulation of the Problem
Zero is infinity in Mayan math. According to the traditional Mayan
teachings, zero and infinite are one. These teaching are described in [4, 5, 6]
and summarized in [1].
There are similar philosophical statements in other religious teachings.
Similar statements about the similarity between nothing (zero) and
everything (infinity) abound in many philosophical and religious teachings.
For example, many Jewish thinkers cite a 19 century Rabbi Simhah Bunim
who said that “A person should have two pieces of paper, one in each pocket,
to be used as necessary. On one of them [is written] ‘The world was created for
me,’ and on the other, ‘I am dust and ashes’.” [7]; according to Rabbi Bunim,
the key to living a successful life is to be guided by both of those statements
and keep those two opposing truths in balance.
While philosophically profound, the Mayan statement about zero
and infinity does not seem to make much mathematical sense. While
the above Mayan statement – that zero and infinity is one – may have deep
philosophical roots, from the purely mathematical viewpoint, it does not seem
to make sense: a number zero is clearly different from infinity.
What we do in this paper. In this paper, we provide a mathematical
explanation in which the “equality” between zero and infinity starts making
sense.
2 Our Idea
Infinity in mathematical description of numbers: a brief reminder.
The above Mayan statement equates zero and infinity. So, to analyze this
statement, let us briefly recall the use of infinity in describing real numbers.
A consistent use of infinity in describing real numbers emerged with the
development of calculus, where infinity appears as a limit. Some sequences of
real numbers, such as xn = 1
n
, have finite limits: e.g., for the above sequence,
we have xn → 0. Other sequences, such as yn = n, increase indefinitely, they
do not have a finite limit, so we say that yn → +∞. Similarly, for a sequence
zn = −n, we have zn → −∞.
For such sequences, we have two different infinities: +∞ and −∞, located
at two different sides of the real line.
Why in Mayan math, 0 and ∞ are the same 6195
Sometimes, we have a singly infinity. If we know the limit x of a sequence
xn, i.e., if we know that xn → x, then for x = 0, we can conclude that the
sequence
1
xn
tends to 1
x
.
What if x = 0?
• For a sequence xn = 1
n → 0, we have 1
xn
= n → +∞.
• For a sequence xn = −1
n → 0, we have 1
xn
= −n → −∞.
• For an oscillating sequence xn = (−1)n
n → 0, we have 1
xn
= (−1)n · n,
and this sequence converges neither to +∞ nor to −∞.
To describe the behavior of such oscillating sequences, we can “merge” the two
previously separate infinities −∞ and +∞ into a single infinity, and say that
these sequence “converge to ∞”.
This merging enables us to extend the above rule about the limit of the
sequence
1
xn
to the case when xn → x = 0: in this case, we have 1
xn
→ 1
x
,
where we defined 1
0
def
= ∞.
Comment. It should be mentioned that while for real numbers, having a
single infinity is a (reasonably minor) convenience, for complex numbers, the
existence of a single infinite point is a must for many methods and results; see,
e.g., [2].
Unsigned infinities are also useful for computations. Infinities are
useful no only in pure math, they are also unseful for computations. For
example, infinities are often useful to make sure that seemingly equivalent
algebraic transformations of an expression do not change its computed value.
For example, an expression a
a + b can be represented in the equivalent form
1
1 + b
a
. A possible problem with this transformation occurs when a = 0 and
b = 0; in this case:
• the original expression is simply equal to 0, while
• the second expression requires division by zero and thus, does not have
a direct mathematical sense (at least if we only consider usual (finite)
numbers).
6196 O. Kosheleva and V. Kreinovich
To avoid this problem, most computers assume that b
0 = ∞ for b = 0. In this
case, 1 + b
a
=1+ ∞ = ∞, and 1
1 + b
a
= 1
∞
= 0.
How to represent numbers ranging from very small to very large:
floating point representation is needed. To represent usual-size numbers,
we can use a usual fixed point format, and represent, e.g., 1
4 as 0.25.
However, if we want to represent all the numbers describing the Universe,
from very small numbers describing the size of elementary particles to very
large numbers describing the size of galaxies and of the Universe itself, then we
cannot avoid using floating point numbers, i.e., numbers of the type 1.2·10−23.
This is such numbers are described and processed in physics (see, e.g., [3]), this
is how computers represent such numbers.
Resulting explanation of Mayan identification of zero and infinity.
In mathematical terms, when we represent a real number x in the form a · 10b
with a ≈ 1, then b ≈ log10(x). From this viewpoint:
• values x ≈ 0 correspond to e ≈ −∞, while
• values x ≈ +∞ correspond to e ≈ +∞.
When we apply the usual mathematical idea of treating both limit values
e = −∞ and e = +∞ as a single infinity, we thus treat the original values
x = 0 and x = ∞ as similar.
This provides an explanation for Mayan identification of zero and infinity.
Acknowledgments. This work was supported in part by the National Science
Foundation grants HRD-0734825 and HRD-1242122 (Cyber-ShARE Center
of Excellence) and DUE-0926721, by Grants 1 T36 GM078000-01 and
1R43TR000173-01 from the National Institutes of Health, and by a grant
N62909-12-1-7039 from the Office of Naval Research.