Calendar
Traditional High Cultures
Oral and Written Traditions and Rock Art as the Histories of Both Ancient and Living Peoples, Especially of the Americas before the Arrival of Europeans
 

Written History and Geography in Central Mexico --

Codices, Lienzos, and Mapas Linked to the Ground

 

Inter-Regional Scribal Discourse between the Maya and the Mixteca-Puebla: Evidence from the Madrid Codex

by Bryan R. Just, Tulane University Email Bryan Just

Abstract: This paper addresses questions of inter-regional scribal interaction through a consideration of four almanacs in the Maya Madrid codex that share structural affinities with the in extenso format known from Mixteca-Puebla, or Borgia Group, codices. Various aspects of these almanacs reveal that the integration of this non-Maya format was a novel and challenging project for the Madrid’s scribes. Further, structural features of these almanacs suggest a specific function that was likely the rationale for their inclusion in the manuscript. The discussion of these almanacs will elucidate two ways in which we can productively interpret notational ‘errors’ evident in them: (1) as symptoms of the novel integration of a non-Maya notational system; and (2) as deviations from standard conventions meant to signal correlations to other sections of the manuscript. This second class of ‘error,’ in conjunction with the structural and positional relations among almanacs and the physical character of the Madrid codex itself, will be shown to facilitate the coordination of information on the opposing sides of the manuscript. Some broader socio-cultural implications of this inter-regional esoteric interaction will be considered in concluding.

Text: (preliminary draft only -- absolutely no quotation -- and with apologies to the author, lacking the footnotes and figures) Copyright © 2003 Bryan Just. All rights reserved

CHAPTER NINE
IN EXTENSO ALMANACS IN THE MADRID CODEX Bryan R. Just

The Madrid Codex is known to be an eclectic compilation of astronomical, seasonal, ritual and calendrical information drawn from various sources across time and space. Recent research has specified temporally the relevance of portions of the manuscript, revealing that the Madrid’s almanacs reference at least five hundred years of astronomical data (cf. Bricker et al. 1997; Bricker 1997b). Additionally, Alfonso Lacadena (1997) has demonstrated that the Madrid’s hieroglyphic texts incorporate both Ch’olan and Yucatecan lexemes and morphology, suggesting a process of translation from earlier sources and a diglossic context of reception and use. Further, current scholarship is revealing an ever-growing number of thematic and structural links between sections of the Madrid and Mixteca-Puebla, or Borgia group, codices (Bricker and Hernandez, this volume; Hernandez, this volume; Just 2000).
Although the Madrid Codex’s synthetic complexity is widely recognized, little consideration has been given to understanding why select portions of these diverse sources were chosen to be presented together in a single manuscript. Is the Madrid Codex merely a compendium of temporally, thematically, and culturally diverse information, or is there some functional rationale behind the selective inclusion of its constituent sections? And, the related methodological question—can such an overarching cohesion be gleaned from evidence within the manuscript?
This chapter addresses these questions for a related set of four almanacs in the Madrid Codex, each of which shares structural affinities with the in extenso format known from Mixteca-Puebla codices. Various aspects of these almanacs reveal that the integration of this non-Maya format was a novel and challenging project for the Madrid’s scribes. Further, structural features of these almanacs suggest a specific function that was likely the rationale for their inclusion in the manuscript. The first section briefly reviews the structural characteristics of both standard Maya and Mixteca-Puebla in extenso almanacs, focusing on similarities and differences between the two formats. The subsequent section considers in detail the four almanacs in the Madrid Codex that evidence Maya borrowing of the in extenso format. The discussion of these almanacs will elucidate two ways in which we can meaningfully interpret notational "errors" evident in them: (1) as symptoms of the novel integration of a non-Maya notational system; and (2) as deviations from standard conventions meant to signal correlations to other sections of the manuscript. The third section demonstrates that this second class of "error," in conjunction with structural and positional relations among almanacs and the physical character of the Madrid Codex itself, facilitates the coordination of data on the opposing sides of the manuscript. The resultant implications with respect to the use and composition of the Madrid Codex will be noted in the concluding section.
STRUCTURAL OVERVIEW OF STANDARD MAYA AND MIXTECA-PUEBLA IN EXTENSO ALMANACS
Standard Maya Almanacs
Typically, Maya codical almanacs present seasonal, astronomical, and ritual information within a graphically abbreviated "template" of 260 days or a multiple thereof (Figure 1a). Although this period is quantitatively equivalent to the Maya ritual tzolkin calendar, almanacs rarely begin with its first day, 1 Imix. This base template thus serves to provide structural consistency as opposed to describing the tzolkin per se. That is, the mathematical structure of the tzolkin rather than the ritual calendar itself underlies Maya almanac form.
The normal convention of abbreviation in Maya codices divides the 260-day template into a number of equal sets of consecutive days that is divisible by thirteen—typically of 26, 52, or 65 days, producing ten, five, and four sets respectively. In a single column at the far left of an almanac, each set is represented by the glyph for its first day. Because there are thirteen repeating numerical coefficients applied to the day signs in the tzolkin, a single red number above this column specifies the tzolkin position for each of the depicted days.
The initial column of day signs is followed by a series of columns that refer arithmetically to subsections of days within each set marked by the initial day signs. Each column normally consists of one black and one red bar-and-dot number, placed between short hieroglyphic texts (above) and imagery (below). The black number signals the quantity of days required to progress through the section of the template referenced by the column, and the red number records the coefficient of the day arrived at by adding this quantity to the last day of the previous column. At the end of one pass through each column, the count arrives at the day coefficient of the initial column, signaling the reader to return to the beginning of the almanac, shift down one row, and repeat the process. Almanacs continue in this fashion through the entire 260-day sequence. Although it is not made explicit, the imagery and texts associated with these columns of days usually record information pertinent to only one of the many days referenced by the arithmetic (Bricker 1997a:10-12). Sometimes these templates must be cycled through multiple times to account for all of the events recorded in the imagery and text (Bricker 1997a:15-17).
This method of almanac abbreviation, while efficiently condensing large sequences of days and mathematically standardizing the presentation of diverse codical information, inherently entails some degree of ambiguity with respect to precisely locating recorded phenomena within the arithmetic intervals. Further, the standard Maya almanac form minimizes the possibility of correlating multiple events that occurred on various days referenced by a single column, as little space is available for text and/or imagery to mark such commensuration.
Mixteca-Puebla In Extenso Almanacs
Compared to this Maya codical convention, the Mixteca-Puebla in extenso format offers a contrastive suite of organizational advantages and disadvantages while maintaining the basic 260-day structure (Figure 1b). Karl Nowotny (n.d.:354) first coined the term in extenso to describe the introductory eight-page almanacs of the Borgia, Cospi, and Vaticanus B codices, which list in sequence all of the day signs in the 260-day Mixteca-Puebla analog to the Maya tzolkin. While the overall proportions of in extenso almanacs are identical to the standard Maya template, they present the tonalpohualli itself, consistently beginning with the ritual calendar’s first day, 1 Crocodile (cognate with 1 Imix in the Maya system).
Like many of the standard Maya almanacs, in extenso almanacs divide their 260-day sequence into five horizontal rows of 52 days. The Borgia and the Cospi in extenso almanacs read from right to left and from bottom to top, while the Vaticanus B version reads from left to right and bottom to top. Each two-page spread of an in extenso almanac consists of thirteen columns of days, referencing implicitly the thirteen assigned coefficients, as schematized in Figure 1b. This alignment of pages and thirteen-day sets (known to Central Mexican manuscript scholars as trecenas) serves to divide the tonalpohualli into smaller sections analogous to those produced arithmetically in Maya almanacs. Recall, however, that the correlate subsections of Maya almanacs consist of highly varied quantities (see Aveni, this volume).
The division of the tonalpohualli into four groups of trecenas (i.e. one per two-page set) further organizes the ritual calendar, implying a quadripartite structure traditionally associated by scholars with cardinal directionality. The convenient mathematical relation between four and thirteen (i.e. 13/4 gives a remainder of one) allows the trecena pattern to accommodate a shorter four-part cycling, in which the first day shares its association with the trecena and subsequent days cycle through the quadripartite pattern, arriving at the appropriate next position in the sequence on the first day of the subsequent trecena.
The days in each column of an in extenso almanac are also associated with imagery above and below the glyphic register. The images accompanying day-sign columns in the in extenso almanacs are not fully understood but may carry ritual information related to whole columns of days or, as discussed for standard Maya almanacs, to select days within a given column.
In addition to these shared structural characteristics of Mixteca-Puebla in extenso almanacs, the Cospi and the Borgia versions include an enigmatic sequence of footprints that mark nine- and seven-day intervals. Additionally, the Cospi presents the repeating sequence of the nine "Lords of the Night." Each of these patterns are noted within the in extenso grid system, overlapping without interrupting the regular sequence and structure of the tonalpohualli.
The in extenso format, through its comprehensive, grid-like structure, coordinates all of these subgroupings of the tonalpohualli, facilitating the synthetic analysis of multiple mathematical patterns as inherent qualities of the ritual calendar. Additionally, these groupings of days often structure subsequent almanacs in Mixteca-Puebla codices, allowing this more specific information to be coordinated easily through the in extenso almanac. In the Codex Borgia, for example, the almanacs on pages 9-13 and 22-24 relate to the twenty day names, the almanacs on pages 18-21, 49-52 and 53b-c, 57, 72 and 75-76 are associated with the five-day sets found in the in extenso columns, and the almanac on pages 61-70 records more detailed information about the trecenas. That is, the Borgia in extenso almanac relates structurally to at least 31 of the manuscript’s remaining 68 pages. The introductory in extenso almanacs thus allowed their Mixteca-Puebla users not only to read the auguries suggested by the images above and below a given calendrical date, but to see a variety of cycles within a single diagram. They could note how these cycles relate to one another, synthesizing the more detailed information that follows for a more complete and nuanced interpretation. While standard Maya almanacs condense the 260-day template, focusing on a few select events and a single mathematical pattern, in extenso almanacs diagram the multi-layered mathematical complexity of the 260-day ritual calendar, facilitating the analytic synthesis of various cyclical patterns.
IN EXTENSO COGNATES IN THE MADRID CODEX
Madrid Pages 12b-18b
The almanac on pages 12b-18b of the Madrid Codex shares a number of structural characteristics with Mixteca-Puebla in extenso almanacs: (1) it lists the days of the 260-day tzolkin in sequence; (2) these days are organized into horizontal rows that span several pages; (3) imagery is presented both above and below the calendrical register; and (4) the cycle begins on the first day of the Maya ritual calendar, 1 Imix (Figure 2). The conjunction of this medley of characteristics poses a compelling case for a cognate relation between the Maya almanac and its Mixteca-Puebla counterparts. Unlike the Mixteca-Puebla examples, however, (1) the imagery above and below the day signs in the Madrid version is not clearly associated with single columns of days, (2) the page breaks do not imply the sequence of thirteen calendrical coefficients, and (3) the Madrid almanac follows a 4 x 65-day structure instead of a 5 x 52-day one (although there are only 52 day signs painted in each row). This shift to a 4 x 65-day structure, however, may in fact further link these cognate almanacs functionally, a possibility considered further in the concluding section.
Bricker et al. (1997) demonstrated that this almanac and the immediately preceding almanacs on the lower two registers of pages 10-11 record overlapping astronomical and seasonal phenomena for a specific period of time, from A.D. 924 to 925 (Figure 2b; Table 1). While the slight spatial misalignment of some of these events and their corresponding dates is not atypical of Maya codical calculations, it is notable that the first eclipse is depicted on a different page from the day on which it was predicted to occur. This potential discrepancy dissolves, however, if one approaches the almanac vis-à-vis the trecena aspect of the Mixteca-Puebla in extenso structure. Although this trecena pacing is not referenced explicitly by page breaks in the Madrid almanac, it is implied by mathematical and pictorial clues—clues intrinsically tied to the novel borrowing of this format.
The beginning and the end of the in extenso sequence on Madrid 12b-18b include evidence of the process of translating Maya codical information into the Mixteca-Puebla in extenso format. First, page 18b incompletely represents the final day signs of each of the calendrical rows (Figure 2a). It seems as if the scribe ran out of space for the last portion of the almanac and/or chose an alternate method of recording this section. Page 12b references the missing days through an intriguing set of three numbers: a black ‘13’, a red ‘13’, and a black ‘1’, labeled A, B, and C respectively in Figure 3. In standard Maya calendrical computation, a single red and black number pair would be satisfactory to account for the glyphically absent interval, bringing the count back into the in extenso portion of the almanac. The presence of three numbers is thus curious. The anomalous mathematics here provide an excellent example of the first class of so-called errors considered herein, and are directly related to the attempted synthesis of two notational systems. Recall that in Mixteca-Puebla codical notation, each day is explicitly referenced graphically, either by a day-sign or a "spacer," a simple dot denoting de-focused days within an interval. The Maya, on the other hand, typically record quantities to be added to a glyphically recorded day to reach the next glyphic day, one arithmetic interval later. That is, Mixteca-Puebla notation records the units between two points with spacers, while Maya notation records the arithmetic required to move from one point to the next with a quantitative distance number.
In a Mixteca-Puebla in extenso almanac, the transition from one row of calendrical glyphs to the next is conventional; no units exist between the two points in time, and thus no spacers are required. The standard Maya almanac form, on the other hand, arrives at the starting date of a row at the close of the previous row. However, when the count moves from Madrid page 18b back to 12b, it is still on the last day of the previous row, fourteen days before the re-entry date. The arithmetic follows a clockwise pattern, moving back to page 12b (from 18b) in the lower right, adding a quantity of thirteen days to arrive at another ‘13’ coefficient day (the last of the row), and then adding one more day to cycle back into the in extenso portion of the almanac (Figure 3). The arrangement of these numbers and the ‘1’ coefficient on page 13b suggests that the count actually begins on page 13b, and that page 12b mathematically facilitates re-entry into the in extenso portion of the almanac. The red ‘1’ marks the first column of days on page 13b as the analog to the initial glyphic column of standard abbreviated Maya almanacs. The position of the calculations on page 12b before the in extenso portion of the almanac foregrounds this mathematical pattern, stressing distances of one (i.e. the distance moved from day sign to day sign), and thirteen (i.e. the trecena—a fundamental structural unit in Mixteca-Puebla in extenso almanacs).
This reference to the thirteen-day interval signals that this trecena pacing continues though the entire tzolkin count. Such a trecena rhythm is supported by the non-calendrical contents of the following pages; if one overlays a thirteen-day structure on the rows of day signs, as illustrated in Figure 2c, the astronomical phenomena consistently appear in the trecena of their occurrence; the slight discrepancies of placement do not exist vis-à-vis these proposed trecena intervals. Further, the serpents that wind through the calendrical register (1) appear at least once in each trecena; (2) never cross the trecena boundaries; and (3) delineate the first two divisions. The graphic composition of these pages thus implicitly corroborates the trecena pacing suggested by the arithmetic on page 12b.
Whereas the standard Maya abbreviated almanac structure focuses on a single mathematical pattern and provides limited space for recording associated phenomena textually and visually, the in extenso format both presents multiple overlapping sub-groupings of the tzolkin and expands the pictorial space available for correlating pictorially and textually recorded information. The trecena sets, for example, group numerous images above and below them, facilitating the sort of synthetic analysis that seems to functionally drive Mixteca-Puebla codical structure. In this Maya in extenso almanac, Mixteca-Puebla and Maya notational conventions are brought together to facilitate the correlation of multiple events within various overlapping pacings of the tzolkin. As discussed for the Borgia Codex, this Madrid in extenso almanac further allows more specific information presented in adjacent almanacs, such as the seasonal and astronomical information recorded on pages 10-11, to be coordinated within the tzolkin cycle. The Madrid in extenso almanac can thus be seen to "concordance" various data.
Madrid Pages 65-72, 73b
Victoria Bricker (1997b:169; personal communication, 1998) noted that the "Calendar Round" almanac on pages 65-72, 73b shares with 12b-18b an in extenso listing of all 260 tzolkin days and a 1 Imix starting point (Chapter 7, Figure X). The almanac is read along the upper register through the first 16 day signs before shifting to the lower register. At the end of a row in the lower register (on page 72), the count returns to the upper register on the first page and continues in this fashion through the first 256 days of the tzolkin. This tzolkin count moves from day to day without the use of distance numbers, as in Mixteca-Puebla almanacs yet rarely in Maya ones. Unlike both Maya and Mixteca-Puebla canonical almanacs, the coefficients of days in each column are not consistent—each day-sign is accompanied by its own red coefficient. Still, the use of 16 columns per row (a factor of four) groups the day signs in each column into the same quadripartite sets produced by Mixteca-Puebla in extenso almanacs, a feature lacking in the in extenso almanac on Madrid pages 12b-18b because of its 4 x 65-day structure. No cardinal directionality or other four-part designation is recorded glyphically on these pages, yet such a correlation would be recognized readily by scribes familiar with the groupings of five day signs normally produced through a 5 x 52-day format. This grouping thus potentially aligns mathematical tzolkin information, such as quadripartite associations, with seasonal information presented pictorially and textually.
The final four days of this in extenso tzolkin are depicted in the four corners of the final image, on page 73b. This final page also includes, on the last day of the tzolkin, an explicit Calendar Round date of 13 Ahau 13 Cumku, giving it its "Calendar Round" almanac designation (Bricker 1997b:169). By linking seven of its images directly to adjacent almanacs, Victoria Bricker (1997b) was able to date the almanac’s relevance to a time span from about A.D. 1350-1450, some four to five centuries after the events referenced on Madrid pages 12b-18b. Further, Bricker (1997b:180) noted that the shared iconography of this almanac and others adjacent to it suggests that it acted as an index or table of contents for this section of the manuscript, a reading quite similar to the "concordancing" designation for Madrid 12b-18b presented above. It is notable that the central location Bricker noticed for this almanac, among the almanacs it cross-references, is almost identical to the physical placement of 12b-18b on the obverse, appearing on the opposite side of pages 9-17, further suggesting that these two in extenso almanacs may have served similar concordancing roles (see Table 3). This possibility is considered in more detail below.
Madrid pages 75-76
The Formée cross tzolkin on pages 75-76 is the best known example of a Mixteca-Puebla borrowing in the Madrid Codex, bearing a remarkable resemblance to the first page of the codex Fejérváry-Mayer (Plates X and X). While the overall composition of these almanacs is distinct from the in extenso almanacs considered so far, it is arguably an alternative format for recording the same information in that (1) all 260 days of the ritual calendar are listed in sequence; (2) these days are grouped into trecena sets; and (3) in the Fejérváry-Mayer, this almanac acts as the introduction to the manuscript, holding the same initial position in the codex as the in extenso almanacs in the Borgia, Cospi, and Vaticanus B codices. Further, the Formée cross composition groups the 260-day calendar into four sets of 65 days (or five trecenas), each of which are associated with a cardinal direction. Thus, this Formée cross almanac also shares with the Madrid in extenso almanac a 4 x 65-day basic structure.
Like the in extenso almanac on Madrid 12b-18b, the Madrid Formée cross almanac indexes the novelty of its borrowing in its mixing of Mixteca-Puebla and Maya notational systems. The Fejérváry-Mayer version consistently represents the first day-sign of each trecena, followed by twelve spacers representing the remaining twelve days in that trecena. In contrast, the Madrid cognate glyphically records the day sign and the numerical coefficient for both the first and the last day of each trecena. The number of spacers that intervene varies from trecena to trecena, traditionally considered evidence of scribal sloppiness or haste. However, the quantity of spacers per trecena in the Madrid version total eleven, twelve, or thirteen exclusively (Table 2). If spacers represent the interval arithmetically (Maya style), one would need thirteen to progress from one trecena to the next; in contrast, if one follows the format exemplified by the Fejérváry-Mayer, each trecena should include twelve spacers; finally, if the first and last days of each trecena are present glyphically (as they are on Madrid pages 75-76), only eleven spacers are required in the Mixteca-Puebla notational system to account for the eleven intermediary days. Although twelve spacers is the most commonly used variant in the Madrid Formée cross almanac, it seems as if the designer of this almanac was not sure which of these choices was preferable. Thus, we may see this variation not simply as scribal sloppiness, but as an index of a scribe struggling to present a non-Maya notational system. The limited variability reflects the novelty of this borrowing, as it references the logical options available to a Maya scribe newly acquainted with Mixteca-Puebla calendrical notation.
Another peculiar aspect of the Madrid Formée cross almanac reflects a different sort of structural deviation from Mixteca-Puebla convention. While the Fejérváry-Mayer version places the eastern quadrant of the quincunx at the top of the image—the standard orientational convention for Mesoamerica—the Madrid version glyphically marks the eastern region at the bottom of the page (Plates X and X). Still, the correlation of trecenas with directions remains consistent in that the Madrid version also places the starting date of the tzolkin, 1 Imix, in this lower eastern quadrant. Further, the orientation of the imagery and the glyphs on Madrid pages 75-76 favors the standard orientation of the reverse of the manuscript, making the placement of the eastern quadrant at the bottom of the page the only internal suggestion of a possible mixed orientation for this almanac. Its relation to the following almanac on pages 77-78, however, provides some incentive for considering this unexpected cardinal orientation a significant deviation from Mixteca-Puebla norms.
Madrid pages 77-78
Merideth Paxton (1997) recently proposed that the potentially mixed orientation of the Formée cross almanac relates it to the subsequent almanac on Madrid pages 77-78 (Figure 4). The almanac on these two pages is unique in that it is the only almanac on the reverse of the Madrid oriented in the same way as the obverse. When viewed in the opposite orientation of its neighboring almanacs, Madrid 77-78 lists, across the top row, the first thirteen day signs of the tzolkin—that is, the first trecena (Figure 5). The coefficients, recorded immediately below the day signs, are the only elements of the almanac that correspond to the common orientation of this side of the manuscript.
Each day of this first trecena is associated with a series of glyphic and numerical information organized into columns, including directional glyphs, quantities of offerings, and deities. As is common in Mixteca-Puebla divinatory codices, this almanac thus aligns days in the ritual calendar with other ritual information. While it does not present the 260-day calendar in extenso, it shares with the in extenso format a listing of consecutive days presented in sequence without arithmetic abbreviation and with a beginning day of 1 Imix. For these reasons, as well as aspects of its orientation and placement in the manuscript (to be discussed below), this trecena almanac warrants consideration as cognate to the Mixteca-Puebla in extenso format.
Paxton (1997) suggested that the Formée cross almanac on pages 75-76 relates compositionally to this trecena almanac in that each references two opposing orientations and uses the trecena as a basic structural unit (Figure 4). As a pair, the Formée cross almanac and the subsequent trecena almanac allow a reader to coordinate two sets of information, following the tzolkin count on both pages through the first trecena and subsequently recycling the trecena almanac to continue through the remainder of the tzolkin recorded in extenso in the Formée cross. Paxton (1997) noted that when the manuscript is reoriented to read the trecena pages, the eastern region and the beginning 1 Imix day sign on pages 75-76 appear in their expected position at the top of the page, providing a possible rationale for the placement of the eastern quadrant and starting date in a different position from its Fejérváry-Mayer counterpart. While this functional pairing offers at least some justification for classifying the Formée cross almanac as having a mixed orientation, it does not explain why these two almanacs would thus contradict the otherwise consistent orientation of their side of the codex.
The anomalous orientation of Madrid pages 77-78 has long been recognized, extending back to Léon de Rosny’s (1881) observations linking the Troano and the Cortesianus fragments through this almanac. Subsequently, Ernst Förstemann (1902:138) suggested that this was meant to show that this trecena almanac was in the wrong place in the manuscript. While there is a thread of truth to this proposition, the placement and orientation of this almanac are not arbitrary—they in fact link pages 75-78 structurally and physically to the in extenso almanac on pages 12b-18b of the obverse.
IN EXTENSO ALMANACS, ORIENTATION, AND THE COORDINATION OF THE OPPOSING SIDES OF THE MADRID CODEX
The four almanacs in the Madrid Codex discussed above evince various affinities with the in extenso formats of Mixteca-Puebla introductory almanacs. These four almanacs are the only ones in the Madrid Codex to exhibit these structural similarities and thus merit consideration as facets of a singular act of borrowing. The grouping of these almanacs is not only supported by structural features and patterns of mathematical "errors," but by the their physical distribution in the Madrid Codex, which offers a means of using them conjunctively.
As mentioned for the "Calendar Round" almanac on Madrid pages 65-72, 73b, the pair of almanacs on pages 75-76 and 77-78 have a proximal physical relation to the in extenso almanac on pages 12b-18b (Table 3). Note that the "Calendar Round" almanac on the reverse begins slightly before the in extenso pages of the obverse. Similarly, the Formée cross and trecena almanacs fall at the close of the in extenso almanac. This physical bracketing provides an ideal relative placement for viewing various combinations of these almanacs simultaneously. While the divergent orientation of the two sides of the Madrid Codex has traditionally dissuaded consideration of their possible integration, the positioning of the only cases of mixed orientation on the reverse within this two-sided clustering of the Maya analogs to the Mixteca-Puebla formats facilitates exactly this sort of coordination. Figure 6 presents a series of possible manipulations of the Madrid manuscript that allow simultaneous viewing of the four structurally related almanacs that I have considered, as well as the thematically related almanacs on pages 10-11 (Bricker et al. 1997). Each of these possible articulations will be discussed in turn.
When the manuscript is opened to the in extenso almanac, the trecena almanac can be viewed simultaneously by folding the manuscript along the natural, inward-turning fold between pages 20 and 21 (Figure 6a). Thus folded, pages 77-78 lay on top of pages 19 and 20, immediately after page 18. Presented in this manner, both the in extenso almanac and the trecena almanac are readable in the same orientation in their entirety. The reader can easily correlate the contents of the trecena almanac with the corresponding first trecena of the in extenso almanac. Such a cross-referencing between the two offers a logical explanation for the otherwise anomalous orientation and placement of the trecena almanac; the two are not only structurally related, but convey complimentary information about the tzolkin sequence. The orientation of the coefficients on pages 77-78, then, can be seen as a sign to the reader to continue reading onto the reverse of the manuscript. When continuing in this way, the following pages comprise the Formée cross almanac of pages 75-76.
This almanac too is positioned conveniently to be viewed in conjunction with the in extenso almanac. Bringing these pages around for viewing, one can either fold under pages 77-78 or keep them visible as well (Figure 6b, c). Either of these articulations covers the final pages of the in extenso listing, yet the first trecena in the sequence is still completely visible. Presenting this combination of pages facilitates correlation of all three tzolkin sequences. The cardinal orientation of the Formée cross almanac may allude to its relation to pages 12b-18b, in that the eastern quadrant and the initial 1 Imix day appear in the expected upper region when viewed in conjunction with the in extenso almanac. Further, the Formée cross almanac may clue the reader to continue reading onto the back side of the manuscript through the orientation of the figures in the central column, the orientation of the glyphs, and the circular reading of its tzolkin sequence, each of which suggests the necessary reorientation.
One other correlation between the in extenso almanac and the reverse of the manuscript exists, although it is not corroborated by orientation. The first two pages of the "Calendar Round" almanac can also be viewed in conjunction with 12b-18b, if the introductory page (12b) is covered (Figure 6c). Since the in extenso count begins on page 13b with 1 Imix, however, this overlap does not hinder cross-referencing between the two. In fact, this articulation positions the starting 1 Imix days of each almanac adjacently, an ideal articulation for such comparison. Alternatively, the manuscript can simply be flipped over to switch between these two almanacs. The potential coordination of the in extenso almanac and the "Calendar Round" almanac is of particular interest, as it would provide a means of correlating the various almanacs indexed or concordanced by each of them, producing an expansive network of integrated almanacs bridging the opposing sides of the manuscript.
A suite of evidence thus suggests that the in extenso almanac on Madrid 12b-18b played a central role in cross-referencing the two sides of the Madrid Codex. The physical and structural parallels among this in extenso almanac and the paired trecena and Formée cross almanacs provide the best evidence for such integration of the two sides. The relationship between pages 12b-18b and the "Calendar Round" almanac, while not implied by orientation, carries more profound implications for our understanding of the use of the Madrid Codex. It suggests a means of integrating a whole network of almanacs, as was noted as a basic advantage of the Mixteca-Puebla in extenso format.
The historically specific commensuration of events referenced in each of these almanacs place them in distant periods of absolute time: pages 12b-18b record a sequence of events in the tenth century (Bricker et al. 1997), and pages 65-72, 73b reference events in the fifteenth century (Bricker 1997b). In light of this temporal separation, the cross-referencing potential outlined herein poses an important and challenging question; how might this disjunct information have been used in an integrative manner? Perhaps the Madrid scribes were attempting to use Mixteca-Puebla in extenso conventions of concordancing calendrical cycles as a means of revealing calendrical relationships among temporally diverse astronomical and seasonal phenomena. That is, the convenient articulation of these four almanacs facilitates the interpretive discovery of calendrical agency in observable phenomena.
The fact that the Madrid in extenso almanac is centrally located with respect to these structurally related pages marks an interesting contrast to the positioning of their Mixteca-Puebla analogs. Mixteca-Puebla codices consistently position their in extenso almanacs at the beginning, akin to the standard Western system of presenting a "table of contents" at the beginning of a text. The makers of the Madrid, on the other hand, placed their "indexing" or "concordancing" almanacs among related information. This placement suggests that a Maya reader would open the book to a centrally located, concordancing almanac, and could then flip the pages on either side, where additional relevant information would be presented. The Maya placement of their coordinating almanacs centrally thus constitutes an ingenious, efficient and effective alternate to Mixteca-Puebla (and Western) organizational conventions.
SUMMARY AND DISCUSSION
To review, the four almanacs considered herein, namely the in extenso almanac on pages 12b-18b, the "Calendar Round" almanac on pages 65-72, 73b, the Formée cross almanac on pages 75-76, and the trecena almanac on pages 77-78, are related both in terms of their affinities to the Mixteca-Puebla in extenso structure and their physical articulation in the Madrid Codex. Some of the "errors" in these almanacs can be understood as attempts to reconcile distinct notational systems, and as such seem to index the novel (i.e. original) adoption of Mixteca-Puebla conventions. Other anomalous aspects of these almanacs, particularly the abnormal orientation of pages 77-78, suggest a method of integrating information on the opposing sides of the manuscript.
The relation to Mixteca-Puebla conventions and the coordination of both sides of the manuscript are likely mutually inclusive features. The great utility of the in extenso format, especially when compared to the standard abbreviated format of the Maya, derives from its coordination of multiple mathematical patterns inherent to the 260-day ritual calendar. The in extenso structure is ideal for synthesizing calendrical information. As such, I have postulated that the Mixteca-Puebla structures may have been adopted to help reveal calendrical parallels among temporally disjunct seasonal and astronomical phenomena.
It is notable that none of the four in extenso cognates in the Madrid share the basic 5 x 52-day facet of their Mixteca-Puebla analogs. This deviation from the Mixteca-Puebla structural source may in fact serve to maintain an aspect of their original function. Gabrielle Vail (this volume) has recently argued that the 5 x 52-day almanac structure, used so often in the Madrid Codex, provides a means of calculating seasonally timed rituals for specific haab dates within a 52-year Calendar Round. If the 5 x 52-day template was canonically associated with haab dating as Vail proposes, the modification of the standard Mixteca-Puebla 5x52-day structure in the Madrid in extenso cognate may imply that the 4 x 65-day sequence served as the Maya mathematical template for tzolkin-focused calculations. In this respect, it is noteworthy that the sole in extenso cognate in the Madrid Codex that includes a haab date, namely the "Calendar Round" almanac on pages 65-72, 73b, does not use the 4x65-day structure found in the in extenso and Formée cross almanacs. Continued research on the function of 4 x 65-day almanacs in the Madrid is required to verify this possibility.
Finally, it must be acknowledged that the proposed functional coordination of the four almanacs presented in this paper may not in fact reflect an intentional or pervasive interest in articulating the opposing sides of the Madrid Codex. First, it is possible that the physical articulation of Madrid pages 12b-18b and 65-72, 73b is merely a coincidence of function as opposed to an intended articulation for cross-reference. Each of these almanacs may simply be positioned centrally with respect to the adjacent almanacs to which they relate, producing comparable but not intentionally coordinated placements. Additionally, the trecena almanac on 77-78 and the Formée cross almanac on 75-76 may reside on the reverse because of space limitations on the obverse at the end of the in extenso almanac, a possibility potentially supported by the incomplete sketching of the final portion of the in extenso sequence on page 18. This possibility raises its own important questions with respect to the process and sequence of composition of the manuscript, as these space concerns would imply that the subsequent sections were planned and painted prior to the in extenso and/or the related sections on the reverse. These questions, as well as the problem of the shift from the 5x52-day Mixteca-Puebla in extenso structure to the Maya 4 x 65-day structure, call for continued research on the structural and physical relations among almanacs in the Madrid Codex and consideration of the process of composition of the manuscript as a whole.
REFERENCES CITED
Anders, F., M. Jansen, and L. Reyes García
1994 El Libro de Tezcatlipoca, Señor del Tiempo. Libro explicativo del llamado Códice Fejérváry-Mayer. M/12014, Free Public Museum, Liverpool, England. Introduction and commentary by F. Anders, M. Jansen, and A. Pérez Jiménez. Akademische Druck- und Verlagsanstalt, Graz, Austria; Fondo de Cultura Económica, México, D.F.
Bricker, H. M., V. R. Bricker, and B. Wulfing
1997 Determining the Historicity of Three Almanacs in the Madrid Codex. Archaeoastronomy 22(JHA 36):S17-S36.
Bricker, V. R.
1997a The Structure of Almanacs in the Madrid Codex. In Papers on the Madrid Codex, edited by V. R. Bricker and G. Vail, pp. 1-25. Middle American Research Institute, Pub. 64. Tulane University, New Orleans.
1997b The "Calendar-Round" Almanac in the Madrid Codex. In Papers on the Madrid Codex, edited by V. R. Bricker and G. Vail, pp. 169-180. Middle American Research Institute, Pub. 64. Tulane University, New Orleans.
Bricker, V. R., and H. M. Bricker
1992 A Method of Cross-Dating Almanacs with Tables in the Dresden Codex. In The Sky in Mayan Literature, edited by A. F. Aveni, pp. 43-86. Oxford University Press, New York.
Díaz, G., and A. Rodgers
1993 The Codex Borgia: A Full-Color Restoration of the Ancient Mexican Manuscript. Dover, New York.
Förstemann, E.
1902 Commentar zur Madrider Mayahandscrift: Codex Tro-Cortesianus. Verlag von L. Sauniers Buchandlung, Danzig.
Just, B. R.
2000 Concordances of Time: In Extenso Almanacs in the Madrid and Borgia Group Codices. Human Mosaic 33(1):7-16.
Lacadena García-Gallo, A.
1997 Bilingüismo en el Códice de Madrid. In Los Investigadores de la Cultura Maya, pp. 184-204. Universidad Autónoma de Campeche and Secretaría de Educación Pública, Pub. 5. Campeche, Mexico.
2000 Los escribas del Códice de Madrid: Metodología paleográfica. Revista Española de Antropología Americana 30:27-85.
Nowotny, K. A.
n.d. Tlacuilolli: The Mexican Pictorial Manuscripts, Style and Contents, with a catalog of the Borgia Group. Translated in unpublished form by G. A. Everett, Jr. and E. B. Sisson.
Paxton, M.
1997 Códice Madrid: análisis de las páginas 75-76. In Códices y Documentos sobre Mexico, Segundo Simposio, Volumen 1, edited by S. Rueda Smithers, C. Vega Sosa, and R. Martínez Baracs, pp. 63-80. Instituto Nacional de Antropología e Historia and Dirección General de Publicaciones del Consejo Nacional para la Cultura y las Artes, México, D.F.
Rosny, L. de
1881 Les documents écrits de l’antiquité américaine. Compte-rendu d’une mission scientifique en Espagne et Portugal (1880). Mémoires de la Société d’Ethnographie, n.s., 1(3):57-100. Paris.
Villacorta C., J. A., and C. A. Villacorta
1976 Códices Mayas. 2nd edition. Tipografía Nacional,