Stewart Smith

A cleaner grid of time

I declare this day—Wednesday, 12 December 2012—Pythagoras day. As the calendar marks 12-12-12 we ought to reflect upon the majesty of Base 12 as Pythagoreans are wont to do. (Ridiculous / lofty / over-reaching tone is entirely intentional.) To that effect I present a modest manifesto (unfinished and ill-advised) for calendar revision inspired by the beauty of twelve’s divisibility.

<h4>Redefining Time</h4>

The measurement of time is a human right. Measure itself shapes the experience of that which is measured. It is with these two convictions that we engage our current popular system for measuring time.

It is not the goal of this proposal to be absurdly radical—to propose non-linear measurements of time or uselessly abstract definitions, however poetic. Rather, it is the goal of this proposal to be modest and sympathetic to the current Gregorian system by insisting on simple improvements and even reaffirming aspects of the current system after investigating alternatives.

This proposal is divided into installments of investigation. This current installment in particular investigates the division of a year into both months and weeks. It does not investigate the location and measurement of an epoch nor does it propose a fixed location for the year (for example beginning a year in mid-winter as the current system does versus beginning on the first day of spring.) It does not propose a new system for dividing days into hours, minutes, and so on.


1.The earth's relationship to the sun is our primary notion of time, both for earth's rotations and revolutions.

  1. The earth's relationship to the moon is our secondary notion of time. Unfortunately this relationship conflicts with the sun. The sun must take priority. For example, dividing the year into 12 months will be better in the long run than dividing it into 13 (a prime number) to accommodate for the moon's 28 day cycle, privileging a unit of 30 to 31 days over a unit of 28 days. This is unfortunate, but it also mirrors our current notion of time: Weeks do not fit in evenly to months. The moon's cycle does not fit evenly into months.

  2. In our everyday experience local observations outweigh global observations, therefore it is unsuitable to have a single time for all of Earth. We must have timezones.

  3. Irregularities and remainders are inevitable and it is tidiest to place these at the end of a regular pattern. For example, why have a leap day at the close of the second month of the year when it would be more appropriate placed at the close of the entire year?


    Sun relative, moon sympathetic.

Months are an attempt to divide the year (Sun-relative) into units that are sympathetic to the Moon’s period. Unfortunately these units are not truly harmonious, yielding dissonance. (There are slightly more than 13 average moon cycles per year.) Twelve is an acceptable substitute as it is close to 13 but divides the year evenly by 2, 3, 4, and 6.

The number of days in a month is either 30 or 31 with a relationship to the month's zero-indexed number such that:

zero-indexed month number % 2 = number of days % 2

The exception is the final month which will have 30 days with exception for leap years during which it will have 31. This system intentionally places the exception at the end of a patterned stack, rather than the beginning or elsewhere.

In addition to units of measure this proposal includes new names for the divided units themselves. Presently names of the month have no relation to their order. For example the prefix "oct" in "October" of course means "eight" as October was originally the 8th month of the year prior to the insertion of July and August. September, November, and December suffer the same discontinuity. Their names are decoupled from their original numeric meaning and now at best serve as false cognates. Additionally the alphabetical sorting of month names as we know them also has no relationship to their sequential order. This is of importance particularly when listing the contents of a digital folder.

To resolve these issues we may draw upon the Greek alphabet and other sources of inspiration. These humble proposals for new month names simultaneously sort themselves chronologically and alphabetically. Names beginning with vowels evenly divide the year into thirds and have six letters, while names beginning with consonants have seven letters.

No Name Abr Days

01 Alamon Am 30 02 Bedamon Bm 31 03 Cadamon Cm 30 04 Delamon Dm 31 05 Epamon Em 30 06 Fedamon Fm 31 07 Gedamon Gm 30 08 Hedamon Hm 31 09 Idamon Im 30 10 Jakamon Jm 31 11 Kadamon Km 30 12 Lulamon Lm 31 or 30

= 366 or 365 days = 1 year

<em>Moon relative.</em>

Each week is composed of 7 days, allowing 4 weeks to the moon's 28 day cycle. This unfortunately does not fit evenly into months, but in this way it is no worse than the current system.

No Name Abr

01 Aladin Ad 02 Bedadin Bd 03 Cadadin Cd 04 Deladin Dd 05 Epadin Ed 06 Fedadin Fd 07 Gedadin Gd

4 weeks = 28 days = 1 moon cycle = 0.93 months (30) = 0.90 months (31)

Although it was highly tempting to create a regular system where the first day of each month was always the first day of the week, the need for an absolutely regular system (the ability to measure time in weeks) outweighs the impulse to refine the relationship of weeks to months in this way. (For example, the measurement of fetus and newborn development is in weeks.) It seems the dissonance between weeks and months is a painful necessity and will always be with us.

This dissonance is a serious blow to most invariable calendar models. The Hanke-Henry model, which takes the unusual tack of adding an extra week every 5 or 6 years, breaks our assertion that the length of months should always be as equal as possible. That models reshuffle would have created only 240 weekdays of 366 (66%) versus the current calendar which yields 261 (71%) thus decreasing regulated human productivity severely.

To be continued some day . . .