A Hypothesis on the start times of the 260 day tzolk’in, the 365 day haab
and the Long Count Calendars
At what point during the day did Ancient Maya scribes commence the calendar count, at sunrise, noon, sunset, or midnight? The question becomes even more tangled with the realization that to signify a day there is not just one calendar in play but three: the 260 day tzolk’in, the 365 day haab and the Long Count. Did all three calendars start at the same initial hour or did each have their own distinct starting point (e.g. dawn, noon, sunset or midnight)?
How Maya scribes refer to a Calendar Round holds clues as to the respective starting point for each of the three calendars. Clues to a Calendar Round’s inner workings are found in a spurious set of inscriptions which Proskouriakoff and Thompson (1947) named “Puuc-Style Dates” where the numerical coefficient for the recorded month (the haab) was out of synchronization with the tzolk’in coefficient by one day. These researchers also noted that “Puuc-Style” dating occurred widely during the Late Classic in northern Yucatan and occasionally in lowland areas as well. In general, they saw the “Puuc-Style” dating as a local variant of a non-conformist calendar system that differed with the lowland calendar system by one day (Stuart 2004). Mathews (2001) addressed the question of “Puuc-Style” dating in his examination of the Dos Pilas Stela 8 text. The inscription recorded a Calendar Round date with a month coefficient that was obviously off by one day. Mathews did not attribute the miscalculation to scribal error or to the aberrant “Puuc-Style” counting system. He proposed that the tzolk’in and the haab calendars began at different times during the day. This idea assumes that the tzolk’in commences earlier than the haab or the Long Count cycles (Mathews 2001:406). Specifically, Mathews states:
Moreover, it is possible that the tzolk’in day and the haab day began at different times in the 24 hour day; if so, we could expect a minority of the dates to not be in the “normal” form. In other words: if, for example, the tzolk’in day began at 6:00 P.M., and the haab day began at 6:00 A.M., and some event took place at midnight, then the tzolkin date would be advanced one position over the haab date. Thus. . . the date 220.127.116.11.3 2 Akbal 1 Kankin would after 6 P.M., be 18.104.22.168.3 3 Kan 1 Kankin―as recorded at H13-I13-and not until 6.00 A.M. the following day would the next “normal” date begin, viz., 22.214.171.124.4 3 Kan 2 Kankin (Mathews 2001:406).
What Mathews posited was that the “Puuc-Style” dating was not an aberrant counting system, but rather the “error” of minus-one-day, revealed by the inner mechanics of a Calendar Round date. Mathews also noted that on nine examples of aberrant dates a “half-darkend k’in” sign preceded the errant date and posited that this glyph signals a nighttime event (Mathews 2001:406) . In his final analysis he postulated a likely scenario for start times: the tzolk’in commenced at a prior sunset while the haab commenced at the following dawn.
David Stuart later championed Mathews’ insights with a paper titled The Entering of the Day: An Unusual date from Northern Campeche (Stuart 2004) where he examined the inscription carved on a door lintel from the Hecelchakan Museum reading 4 Muluk K’IN o-chi-ya tu-16 MAK. As Stuart noted:
The remarkable feature of the date record is the sign grouping o-chi-ya located between the day and the month glyphs. This can only be the verb ochiiy, ‘it entered’ . . . The mention of the day ‘entering’ within the haab suggests that we have been misled in thinking that northen Puuc-style dates simply reflect a localized structural change in the reckoning of time. Could it be that many of the ritual events commemorated in Puuc inscriptions—the vast majority of them are dedication rites—actually took place in the window of time between the turn of the haab and the arrival of the tzolkin—perhaps between midnight and dawn? . . . If these were nighttime rituals, scribes of the Puuc region may have been especially diligent in utilizing the subtle mechanisms of the Calendar Round to specify just when certain events took place within our own conception of a 24-hour day . . . (Stuart 2004:1-2).
Stuart continues to say that it is possible that the “Puuc-Style” dates are not a separate system after all but a calendar containing “nighttime indicators” recording night rituals. Yet, he differs with Mathews in theorizing the turning point between the two calendars and favors a separation by six hours rather than twelve, with the haab starting at midnight and the tzolk’in at dawn.
What is the proper turning point for each of the three respective calendars and how many hours they are out of sync from each another? The question will be resolved in two parts: (1) in a translation for “the half-darkened k’in” sign and (2) by looking at how a count of days (k’ins) is related to the haab and the Long Count. Recently, MacLeod and Schele (2005) jointly investigated and updated a catalogue of “Puuc-Style” dates within Maya inscriptions. They compiled additional evidence that the haab and the Long Count do indeed begin at sunrise and that it is the tzolk’in that is out of step by twelve hours. The first line of evidence concerns a reading for the “half darkened k’in” sign that often accompanies aberrant dates, as previously noted by Mathews. MacLeod (p.c. 2008) noted she proposed a reading in 1991 for this “half-darkend k’in” sign; the collocation is sometimes spelled yi-K’IN-ni and suggested the readings of chah-k’in or yi’h-k’in (for the variant with T135 /cha/ superfixed “darkened sun” and “aged sun” . A darkened, aged, or black sun lends itself to the idea of “sunset” rather than just “night” and MacLeod posited that this glyph reflects a sunset position (MacLeod and Schele 2005). Additionally the logograph PAS for “dawn” is further evidence that Maya scribes recognized the horizon position of a dawning sun (p.c. MacLeod 2010). In her analysis and comparison of the errant dates on Dos Pilas Stela 8 and Yaxchilan Stela 18, MacLeod (p.c. 2005) saw a similar error pattern emerging. She noted:
[The Yaxchilan date of] 3 Eb given on the monument represents a tzolk’in that has advanced by one, ahead of the haab, just as 3 K’an at Dos Pilas represents a move forward by one in the tzolk’in. These two dates represent the same pattern. Both of these have moved ahead not only of the haab but also of the Long Count. I really think this is the key. The Long Count counts days―that’s what the ‘ones’ unit is . . . k’ins! . . . The monuments record not only a shift forward in the tzolk’in but also a half-darkened k’in sign―further evidence that the out-of-whack Calendar Round refers to a night event which immediately follows the correct Type III [normal] date in each case . . . It is the tzolk’in which is out of step. Furthermore, it just makes good conceptual sense that a system that counts days as the Long Count should start those days when the day begins― at sunrise. Therefore, the only way the tzolk’in can get out of step is to change at sunset. I truly believe that this plus the specific mention of the half-darkened k’in (a perfect image of sunset) in these two critical cases is all the proof we need (p.c. MacLeod 2005).
For the sake of building an event-line horizon and plotting the starting points of all three calendars, the current study agrees with a combined Mathews and MacLeod/Schele hypothesis that: (1) the haab is in-step with the Long Count, (2) the tzolk’in is out-of-step with the haab by 12 hours, and (3) the tzolk’in begins at sunset 12 hours prior to haab at dawn.
This combined hypothesis now allows the proposal of a timeline corresponding to the solar trek. First, the two parts of the Calendar Round, the tzolk’in and the haab have separate starting points with the haab commencing at dawn and the tzolk’in starting at sunset 12 hours earlier. The Long Count (a count of suns) is synchronized with the haab solar calendar and therefore begins at dawn as well. A charted time-line (with a twelve hour shift between the tzolk’in and the haab) for the base date 4 Ajaw 8 Kumk’u is as follows:
4 AJAW ― DUSK (tzolk’in cycle starts) [+ 12 hours] DAWN (haab day and Long Count cycle start) [+12 hours] DUSK (tzolk’in cycle completes) [+ 12 hours] DAWN (haab day and Long Count day cycle complete the first day) — 8 KUMK’U
Only when the tzolk’in and haab have both completed their respective 24 hour cycles (totaling a span of 36 hours due to the 12 hour shift between both calendars) can the day be recorded as a complete elapsed day 4 Ajaw 8 Kumk’u. If the day records an action after the tzolk’in has finished its 24 hour cycle but prior to the haab completion, then the day and month coefficients of the Calendar Round are recorded out-of-sync by one day and therefore record a tzolk’in coefficient advanced by one day ahead of the haab date .
More grist for the mill.
 See Yaxchilan Stela 18 (A1-A3) for just such a date.
 A reading of yi-K’IN-ni as “black of the day” has also been recently proposed (Houston, Stuart and Chinchilla 2001:395).
 the yi-K’IN-ni sign also interchanges with glyph G9 of the Lords of the Night)
A calendar with embedded with “nighttime indicators” (Stuart 2004) indicating current time within a 36 hour time span is disturbing to say the least. Past researchers have agreed in principle that the Maya never designated a present day or as an unfinished unit, but always treated the day as elapsed time; as Morley noted “the day recorded is yesterday because to-day can not be considered an entity until, like an hour of astronomical time, it completes itself and becomes a unit, that is yesterday” (Morely 1975:470). The Puuc dates are heretical indeed!
Houston, Stephen D. Stuart, and O. Chinchilla M. (eds.)
2011 The Decipherment of Maya Hieroglyphic Writing. Norman, University of Oklahoma Press.
MacLeod, Barbara and Elaine Schele
2005 The Puuc Heresy. Paper written for a seminar on Maya Hieroglyphic Writing, Department of Art History, University of Texas at Austin. Paper in possession of the authors.
2001(1977) The Inscription on the Back of Stela 8, Dos Pilas, Guatemala. In:
TheDecipherment of Maya Hieroglyphic Writing, edited by S. D. Houston, D. Stuart, and
O.Chinchilla M., pp. 394-415. Norman,University of Oklahoma Press.
Morley, Sylvanus Griswold
Morley, Sylvanus Griswold
1975 An Introduction To The Study Of The Maya Hieroglyphs. New York: Dover
Proskouriakoff, Tatiana, and J. Eric S. Thompson
1947 Maya Calendar Round Dates Such as 9 Ahau 17 Mol. In: Notes on Middle American Archaeology and Ethnology, no. 79. Carnegie Institution of Washington, Division of Historical Research, Washington, D.C.
2004 The Entering of the Day: An Unusual Date from Northern
Campeche. On-line at Mesoweb:www.mesoweb.com/stuart/notes/EnteringDay.pdf.