Re: [HM] Further Fruitfulness of Fractions

Milo Gardner (milo.gardner@24stex.com)
Fri, 18 Sep 98 11:58:46 -0800

John Conway's response to Sam's ratio
question is logically correct. That is,
Britain's 'Golden Rule' is only an arithmetic
idea, one that connects to geometry, as John
clearly cites; however does not connect to number
theory nor algebra. To connect to number theory
and to algebra ancient Egyptian arithmetic can be
cited by reviewing two ancient documents.

Both documents add a historical level of understanding
that kids and adults might enjoy. I would like to
suggest a close reading of the RMP 2/nth table
and the EMLR, both hieratic documents written
before 1500 BC. Proto-number theory is easily
seen, as well as algebric relationships that
are summarized as follows:

1. The RMP 2/nth table wrote all 2/pq series,
except for 2/95, as exact unit fraction series
by an ancient Egyptian rule, that is equivalent to:

a. 2/pq = 1/a x a/pq

where a = (p + 1) and (p + q).

Interestingly, the selection of a = (p + q) is
a special case that does produced a more concise
series than (p + 1) for the 2/35 and 2/91 cases.
In addition, two other forms may have been
historical known to the ancient scribes, like
Ahmes, as noted by:

b. 2/pq = (1/p + 1/q)2/(p + q)

as seen an an inverse Golden Proportion, 2/p = A x H
the product the arithmetic mean (A) and harmonic mean (H)
in the form of:

b1: 2/pq = 1/A x 1/H

and/or

c. n/pq = 1/pr + 1/qr

where n = 2, and r = (p + q)/n

as the Akhmim P. is cited by Howard Eves (AN INTRODUCTION
TO THE HISTORY OF MATHEMATICS) as equation c. being an
algebraic identity.

Taking the modern British form of the Golden Rule of
fractions, and looking for a confirmation of an ancient
Egyptian context, consider rule a., stated above, from the
RMP and rule d, from the EMLR, stated below:

d. 1/pq = 1/a x a/pq

can be seen as a 'false postion' -like pick a number
type of mental process. Line 1 of the EMLR lists

1/8 = 1/10 + 1/40

There are several available methods, some historical and
others not historical, that can achieve the desired
exact conversion of 1/8th to 1/10 1/40. I suggest that
rule d is appropriate by letting a = 5, is given by:

1/8 = 1/5 x 5/8 = 1/5 x (1/2 + 1/8) = 1/40 + 1/40.

One way to confirm that the EMLR student scribe selected
various values for a paritioning value a, is shows by

1/8 = 1/25 + 1/15 + 1/75 + 1/200

and,

1/16 = 1/50 + 1/30 + 1/150 + 1/400

two 'out of order' EMLR series.

as reversed engineered, by setting a = 25, as
shown by:

1/8 = 1/5 x (1/5 + 1/3 + 1/15 + 1/40)

= 1/5 x (1/5 + 2/5 + 1/40), since 2/5 = 1/3 + 1/15

= 1/5 x (3/5 + 1/40)

= 1/5 x (25/40)

= 1/25 x 25/8

or,

1/pq = 1/a x a/pq

with a = 25.

Thank you all for considering this ancient form
of exact Egyptian fraction arithmetic as a very
old example of the modern Golden Rule of fractions.

Regards,

Milo Gardner
Sacramento, Calif.

Postscript: The 2/95 case simply is 2/19 times 1/5,
where 2/19 was computed by:

e. 2/19 - 1/a = (2a -p)/ap

with a = 12, and the aliquot parts of 12 (6, 4, 3, 2, 1)
used to additively compute 2a - p, or 24 - 19 = 5
mentally by 3 + 2 rather than 4 + 1, or

2/19 = 1/12 + (3 + 2)/(12*19)

= 1/12 + 1/76 + 1/114

as all 2/p RMP series < 2/101 were computed.

Finally, calculating 2/95 by the multiple 1/5
reveals:

2/95 = 1/60 + 1/380 + 1/570

as Wilbur Knorr partially explained in 1982 (Historia
Mathematica) in his article on Egyptian and Greek
unit fraction series.

The RMP 2/101 series in the RMP followed a simple
EMLR example of 1/p = 1/p(1/2 + 1/3 + 1/6), given by:

2/p = 1/p(1/1 + 1/2 + 1/3 + 1/6)

or,

2/101 = 1/101 + 1/202 + 1/303 + 1/606