## The Correlation Problem

Because various versions of the Mesoamerican calendar have been recorded in regular use from the time of the first arrival of the Spaniards to the present time, it would seem to be a simple matter to determine what day of our calendar corresponds to a specified day of the Mesoamerican calendar. Indeed, the agreement in our sources for the colonial and modern period is such that we can say with assurance that JDN 2447893, 1 Jan 1990 (Gregorian), corresponds to a Mayan date: 11 Muluc 7 Kankin and to the equivalent Aztec date: 11 Water of the year 4 Rabbit, with a possible error of not more than ±1 day with respect to the colonial calendar. Unfortunately, the era base had ceased to be used before the Spaniards arrived; so the Long Count position is not known, and we do not know if any shift in the CR occurred. Determining the equivalence between our calendar and the classic Maya dates is the correlation problem. The solution to the problem is defined as the number of days that must be added to a particular Mayan Long Count date to equal the Julian Day Number (cf. §4.1.5). This interval is normally called the correlation constant and was occasionally referred to in the older literature as the Ahau equation. Over 30 correlations have been proposed in the past century, differing by over 1200 years, and many others have been considered. Most Mesoamericanists at the present time favor one or the other of the two solutions proposed by J. Eric Thompson. His first solution was 584,285, which seems to agree with data on eclipses and new moons as usually interpreted; later, he revised this by two days to 584,283, which agrees better with modern calendri-cal evidence but less well with astronomy. If there never has been a major calendar revision (and we have no direct evidence for such a major revision), the true correlation should be one of the two Thompson correlations or one deviating from 584,284 (±1 day) by some multiple of 18,980. These are called continuity correlations. In the Mesoamerican calendars still in use, 12 Kan 2 Pop (shifted in some cases to 1 Pop) was the equivalent of JDN 2440 328 ± 1 Apr. 16,1969. In continuity correlations, this is a repetition at an undetermined number of CRs of 10.2.8.9.4 12 Kan 2 Pop, the date of the Initial Series Lintel at Chichen Itza, still in the Maya Late Classic Period. The archeological evidence and the historic evidence make it virtually impossible for this date to be later than 1200 a.d. Even the Diettrich correlation, which would put the date in 1189, or the Vaillant correlation No. 1, which would put it in 1187, seem unacceptably late to DHK. The correlation 660,205, now preferred by DHK, would put the Lintel in 1085. The widely accepted Thompson correlations would put the Lintel 208 years earlier. A considerable number of Mesoamerican lists and statements of the correlation of native day names with the Julian or Gregorian date may be found in Caso (1967). It is disconcerting to find highly competent scholars accepting both a pre-Columbian leap year and a Thompson calendar correlation for the Maya. Those two correlations are entirely based on the premise that there was no leap year and that the same system was in step from Michoacan to Yucatan and Guatemala and from the Classic Maya to the 20th century. For a full discussion of the correlation problem, see Kelley (1983,1989), where it is suggested that there is evidence for a calendar reform of some sort in the 10th century and that there are at least 8-10 reasonably probable criteria that the Thompson correlations do not meet. They include a considerable number of astronomical interpretations that appear later in this book. Most of these are met by the correlation 663,310, proposed by Kelley (1983), but evidence on the Uaxactun alignment and its relationship to the inscriptions (cf. §12.21) suggests that this too is incorrect. A solution to the problem should bring the astronomical evidence into much clearer focus. A list of a number of correlations that have been proposed is found in Table 12.6. Bryan Wells and Adreas Fuls independently developed the correlation 660,208. This is a modification of the continuity correlation 660,205. Strong historical evidence indicates that these correlations (three days apart) give dates falling in the correct year.

0 0