Archaeological Deposits from the Aztec New Fire Ceremony

Abstract

Abstract The New Fire Ceremony,is one of the few Aztec rituals documented,in both the archaeological and historical records. The Spanish chroniclers described the New Fire Ceremony,as an imperial celebration of the renewal of cosmic time that was observed on the local level by the renewal of household,goods. George C. Vaillant first proposed the identification of artifact dumps,at Aztec sites with descriptions of these local celebrations. We describe unpublished artifact dumps,excavated by Vaillant at Chiconautla and Nonoalco in the Basin of Mexico and by Smith at Cuexcomate,in Morelos and show,that their context and content support Vaillant’s hypothesis. Our data suggest that the New Fire Ceremony,was an ancient and widespread ritual in Postclassic central Mexico that was appropriated by the Aztec empire as part of its program,of ideological legitimization and control. Our current knowledge,of Aztec religion and ideology is strongly

Full text

ARCHAEOLOGICAL DEPOSITS FROM THE AZTEC
NEW FIRE CEREMONY
Christina M. Elson
a
and Michael E. Smith
b
aDepartment of Anthropology, American Museum of Natural History, Central Park West at 79th Street,
New York, NY 10024-5192, USA
bDepartment of Anthropology, State University of New York at Albany, Albany, NY 12222, USA
Abstract
The New Fire Ceremony is one of the few Aztec rituals documented in both the archaeological and historical records. The Spanish
chroniclers described the New Fire Ceremony as an imperial celebration of the renewal of cosmic time that was observed on the
local level by the renewal of household goods. George C. Vaillant first proposed the identification of artifact dumps at Aztec sites
with descriptions of these local celebrations. We describe unpublished artifact dumps excavated by Vaillant at Chiconautla and
Nonoalco in the Basin of Mexico and by Smith at Cuexcomate in Morelos and show that their context and content support
Vaillant’s hypothesis. Our data suggest that the New Fire Ceremony was an ancient and widespread ritual in Postclassic central
Mexico that was appropriated by the Aztec empire as part of its program of ideological legitimization and control.
Our current knowledge of Aztec religion and ideology is strongly
skewed toward the official state religion of Tenochtitlan. There
are very few ethnohistoric descriptions of domestic or community
religion, in contrast to the abundant sixteenth-century descriptions
of Aztec state religion, with its complex mythology, its elaborate
cosmology and iconography, and its flamboyant rituals involving
human sacrifice, public processions, music, and dance. Modern
scholarship has followed this trend in the available data by focus-
ing almost exclusively on Aztec state religion (e.g., Brundage 1985;
D. Carrasco 1991, 1999; Graulich 1999; Nicholson 1971b). There
is little information on the extent to which the well-described
state-sponsored myths, cosmology, and rituals were shared or im-
itated by commoners or people outside Tenochtitlan.
The lack of excavations at Aztec sites makes it difficult to
establish points of contact between state and household religion.
The extensive excavations at the Templo Mayor of Tenochtitlan,
which provide important archaeological data on key Aztec impe-
rial rituals (Boone 1987; Broda et al. 1987; Guilliem Arroyo et al.
1998; López Luján 1994; Matos Moctezuma 1988; Olmos Frese
1999) are not matched by excavations of residential contexts in
the imperial capital. Excavations and surveys of residential sites
outside Tenochtitlan, on the other hand, have yielded physical
material remains of household-level rituals involving ceramic fig-
urines, incense burners, and other objects (Brumfiel 1996; Par-
sons 1972; Smith 2001b), but the nature of the rituals that employed
these items—and their relationship to Aztec state religion—are
far from clear.
At the time of conquest, Tenochtitlan was the capital of an
empire with several million subjects, and although its influence
was felt throughout much of Mesoamerica, most city-states re-
tained a great deal of local autonomy (Berdan et al. 1996; P.
Carrasco 1999). Did the imperial government at Tenochtitlan con-
trol the timing, practice, and content of local religious practices?
Was household religion just a smaller-scale version of the well-
described state religion taking place under the direction of local or
imperial elite? Or was domestic ritual an autonomous domain only
distantly related to the state and imperial religion? Scholars need
to develop models explaining how state and household religion
interacted if we are to address Aztec religion and ideology in a
culturally dynamic way.
An examination of the ethnohistoric and archaeological evi-
dence for rituals conducted as part of the Aztec New Fire Cer-
emony, a ritual of renewal conducted every 52 years, provides
important new insights into the relationship between household
and state religion. This is one of the few known Aztec rituals
whose celebration can be documented in both state and domestic
contexts, and with both ethnohistoric and archaeological data. In
this article we present the ethnohistorical descriptions of state and
household rituals conducted during the New Fire Ceremony and
evaluate the archaeological evidence for those rituals that took
place at the household level.
THE XIUHTLALPILLI (NEW FIRE) CEREMONY
The Aztecs conceived of time as cyclical, full of perpetual motion,
and of themselves as a “chosen people” who collaborated with the
gods to maintain the cosmic order (Leon-Portilla 1963:36–46).
The Aztec calendar consisted of a 365-day secular calendar and a
260-day ritual calendar. The two calendars combined every 52
years to form a cycle similar to our century, which the Aztecs
called a “bundle” of 52 years. The Aztecs believed that the world
had been created and destroyed four times, and that the current
age (the “fifth sun”) would come to a violent end at the termina-
tion of a 52-year calendrical cycle (Taube 1993). It was not known
Ancient Mesoamerica,12 (2001), 157–174
Copyright © 2001 Cambridge University Press. Printed in the U.S.A.
157

which cycle would mark this destruction, and the purpose of the
New Fire Ceremony was to celebrate the start of another 52-year
cycle and the renewal of the world.
In addition to its role in Aztec cyclical celebrations, the New
Fire Ceremony was important in Aztec conceptions of linear time.
The continuous year-count annal was an Aztec historical genre
that kept track of the key events in the histories of individual
city-states. As analyzed by Elizabeth Boone (2000:197–237), these
painted histories stressed continuity, and time was expressed in
the linear arrangement of year glyphs across the pages. The origin
and expansion of the Mexica empire were major themes in these
histories, and the form may have originated in Tenochtitlan (Boone
2000:198). Boone notes that “[t]he most basic elements [in the
Aztec annals], after the year count, are the sequent rulers and the
binding of the years or the New Fire ceremony at the turn of
the fifty-two year cycle” (Boone 2000:223). In these histories, the
New Fire Ceremony played a role in tracking the histories of
city-states and, most prominently, the history of the Mexica empire.
Although the native histories of the Aztecs contain numerous
references to New Fire celebrations (Tena 1987), the actual cer-
emony is described in detail in only one source (Sahagún 1950–
1982:Book 7:25–32), with brief mentions in several others (Códice
Tudela 1980:Folio 83v–84r; Durán 1967:Book II:453– 454; Gómez
de Orozco 1945; Motolinía 1951). Scholarly descriptions and analy-
ses can be found in a number of sources (Broda 1982; Brundage
1985; D. Carrasco 1999; Moedano 1951; Nicholson 1971a, 1971b;
Sáenz 1967).
The State Level
On the evening of the last day of the 52-year cycle, priests climbed
a hill called Huixachtlan near Culhuacan in the Basin of Mexico,
where they watched the stars. When the Pleiades constellation
crossed the zenith, this signaled that the sun would rise the next
day to start another 52-year calendar round. Using a fire drill, the
priests then started a new fire on the chest of a sacrificial victim.
When the flames grew into a full fire, the victim was sacrificed,
and his heart was thrown into the fire. Over the course of the day,
numerous victims were offered as sacrifices (Motolinía [1951:113]
gives the total as 400).
All fires in Aztec houses and temples had been extinguished
five days previously so that people could wait for the possible end
of the world. Once the priests had started the new fire on Mount
Huixachtlan, signaling the renewal of the world, this flame was
used to light torches that were carried by runners to all the neigh-
borhoods and towns to relight the domestic and public fires. The
temple gods were refurbished. Several scholars (Brundage 1985:9;
Caso 1967:129–140; Taube 2000:315) have pointed out a sym-
bolic association between the lighting of the new fire and the
creation of the fifth sun at Teotihuacan, in which a god leapt into
a bonfire to become the sun. Davíd Carrasco (1999:96–114) adds
a political dimension by focusing on the prominent role played by
the Mexica king Motecuhzoma in some parts of the ceremony,
and on its connections to the Mexica state and empire. For exam-
ple, Motolinía (1951:112) states that people had extinguished their
fires “by command of the ministers of the temples.” The hierarchy
in the distribution of the New Fire followed the imperial order:
The twin temples at the Templo Mayor received the first flames,
then Huitzilopochtli’s calmecac; then other temples and calme-
cac; then various telpochcalli; and finally individual households.
The following passage from Motolinía (1951:113) supports
Carrasco’s political interpretation of the distribution of the new
fire:
Thereupon, as the fire was being blessed, waiting Indians from
many towns carried new fire to their temples. They did this
after asking permission from the great chief of Mexico [Mo-
tecuhzoma], the pontiff who was, as it were, their pope. They
performed this with great zeal and haste. Although the place
might be many leagues away, they went in such haste that in a
short time they placed the fire there. In the provinces distant
from Mexico they performed the same ceremony and it was
done everywhere with much feasting and rejoicing.
The text of the Codex Tudela (Códice Tudela 1980) also em-
phasizes that the new fire was carried from the palace of Motecu-
hzoma to a wide area, stating twice that it was distributed all over
the land (después llevan el fuego a casa de Motençuma o del
s[eño]r q[ue] governava y de allí se reparíia por toda la tierra,
“then the fire was carried to the house [palace] of Motecuhzoma,
the lord who ruled, and from there was distributed all over the
land” [Códice Tudela 1980:Folio 84r; see also Gómez de Orozco
1945:62]). In Carrasco’s words, “[T]his fiery display ignited the
imperial landscape as well as the new time period” (D. Carrasco
1999:125). Motolinía (1951:112) mentions new fire celebrations
in “Texcoco and Its Provinces,” but it is not clear whether these
ceremonies were an ancient Acolhua ritual or whether he was
describing the celebration of the Mexica ritual in Texcoco just as
it was done “all over the land.”
The Local Level
The imperial celebrations of the New Fire Ceremony were matched
by observances in local communities throughout the empire. On
the last night of the five-day period during which the Aztecs had
lived without fire and in fear of the end of the world, the people
held an all-night vigil directing their unwavering attention toward
Mount Huixachtlan. At the site of the distant new fire, they re-
joiced. Once the new fire was lit and distributed all around, “ev-
eryone renewed his clothing and all the household goods....Thus
it was said that truly the year newly started” (Sahagún 1950–
1982:Book 7:31). Sahagún mentions that adults (and children)
performed autosacrifice. Some commoners threw themselves into
temple fires to blister themselves. At home, incense was offered.
They grasped the incense ladle, and raised it in dedication to
the four directions in the courtyard. Then they cast it into the
hearth. Thus incense was offered [Sahagún 1950– 82:Book 7:31].
In another passage, Sahagún describes in greater detail the
household rituals that took place:
First they put out fires everywhere in the country around. And
the statues, hewn in either wood or stone, kept in each man’s
home and regarded as gods, were all cast into the water. Also
(were) these (cast away)—the pestles and the three hear th stones
(upon which the cooking pots rested); and everywhere there
was much sweeping—there was sweeping very clean. Rubbish
was thrown out; none lay in any of the houses [Sahagún 1950–
1982:Book 7:25].
158 Elson and Smith

This passage was depicted by Sahagún’s artists in the illustra-
tion shown in Figure 1. A similar brief description in the Codex
Tudela emphasized the destruction of cooking pots:
y quebravan todas las ollas y cantaros q(ue) avian servido y los
comals y basijas q(ue) tenían, todo lo quebravan [Códice Tudela
1980:Folio 83v; see also Gómez de Orozco 1945:62].
and they broke all of the jars and pitchers that had been used,
and the cooking griddles and [other] vessels that they had, all
were broken [authors’ translation].
Why would people destroy household goods such as “statues
regarded as gods,” clothes, pots, pestles, incense burners, and hearth
stones? We believe that these actions illustrate Aztec religious
ideology. Often inAztec religious ceremonies, a person was cho-
sen to play the role of a divinity for a time before being sacrificed
to that divinity. The impersonator, or ixiptla, was thought to be-
come infused with the divinity’s “essence.” In one example from
Durán discussed by López Austin (1997:33), after the imperson-
ator of the goddess Atlan Tonan (patroness of lepers) was sacri-
ficed, not only the body, but also any object touched by the
impersonator, was destroyed because these objects were infused
with the divine essence of the goddess. In Aztec ideology people,
inanimate things (such as fire), and objects manufactured by hu-
mans could become infused with divine essence. This belief has
been discussed regarding to the ubiquitous Aztec “idol,” but other
examples of objects becoming infused with divine essence in-
clude musical instruments, objects used in games, and hunting
implements (Brundage 1985:67–71, López Austin 1997:28–35).
The Aztecs rejoiced at the beginning of the new century be-
cause, after passing though the dangerous year-end period, they
believed that “sickness and famine have left us” (Sahagún 1950–
1982:Book 7:31). We suggest that the Aztecs replaced some kinds
of household objects because, as potential receptacles of “essenc-
es,” these things no longer held desirable properties.
ARCHAEOLOGY AND NEW FIRE RITUALS
The descriptions of the New Fire Ceremony clearly document
both imperial and domestic behavior. The events recorded at Mount
Huixachtlan were imperial-level activities that took place only
once every 52 years under the auspices of the emperor. In the
household, events such as offering incense and cleaning and sweep-
ing the hearth took place. The culmination of the 52-year cycle—
which for most people in Aztec society was a once-in-a-lifetime
event—was the destruction and replacement of idols, clothes, and
household utensils such as jars, pitchers, and comales.
The suggestion that certain artifact dumps reflect New Fire
rituals was first proposed by George C. Vaillant in the 1930s.
Unfortunately, it was presented as an integral part of a now-
discredited model linking artifact dumps to specific historical events
and ethnic groups. Vaillant (1937, 1938) suggested that the ethno-
historically documented destruction of household items produced
archaeologically identifiable deposits of “simultaneously de-
stroyed groups of vessels, as opposed to the more usual gradual
accumulations in refuse heaps” (Vaillant 1938:552); he called these
deposits “cyclical dumps.” We can add some precision to Vail-
lant’s expectations for dumps. First, New Fire artifact dumps should
contain ceramic vessels that are either whole or reconstructable.
There should be very few fragmentary vessels or single sherds, in
contrast to middens, which consist primarily of many sherds and
highly fragmented vessels. Second, the kinds of vessels found in
artifact dumps should reflect a typical household assemblage, and
the proportions of vessels in artifact dumps should be similar to
that of middens and other domestic deposits. Third, the deposits
should be unstratified and consist of a single zone of artifacts.
Fourth, the deposits should be located in the vicinity of houses or
house groups.
At least three sets of deposits have been interpreted as archae-
ological remains of New Fire celebrations. First, Vaillant (1937,
1938) identified artifact dumps at Chiconautla and Nonoalco as
New Fire deposits. Unfortunately, he never published these exca-
vations adequately. We present data later suggesting that Vaillant’s
interpretation of the deposits as New Fire dumps was correct,
even if his larger model was not. Second, Smith (1992) excavated
a number of “rock piles” at the site of Cuexcomate in Morelos,
which he has interpreted as New Fire dumps. Third, Eduardo Nogu-
era (1968) interpreted the so-called Volador offering from Tenoch-
titlan (also excavated in the 1930s) as a New Fire deposit, an
identification that many scholars have accepted. We question this
view later by showing that the Volador deposit bears little relation
to either the ethnohistoric descriptions or the other archaeological
examples of such deposits.
Vaillant’s Model
Vaillant played a particularly active and significant role in work-
ing out the basic archaeological chronology for central Mexico. In
the late 1920s, Vaillant excavated several Formative sites in the
Basin of Mexico and Morelos and quickly published the site re-
ports (Vaillant 1930, 1931, 1935a; Vaillant and Vaillant 1934),
resulting in the refinement of that period. He then turned his at-
tention to the Classic and Postclassic periods (with a particular
Figure 1. The dumping of domestic goods for the New Fire Ceremony
(Sahagún 1950–1982:Book 7, Figure 19).
Archaeological deposits from the Aztec New Fire Ceremony 159

interest in refining early Aztec history) and, in the mid-1930s,
conducted a series of excavations with Suzannah Vaillant. Unfor-
tunately, they never published the results of this fieldwork, except
for a few photos and general remarks in journal articles and in
George Vaillant’s textbook, Aztecs of Mexico (Vaillant 1937, 1938,
1941).
1
Nevertheless, Vaillant’s data and interpretations influ-
enced scholars working in central Mexico.
2
In the summer and fall of 1935, while conducting fieldwork at
Chiconautla and Nonoalco, the Vaillants took a course with Al-
fonso Caso on Mesoamerican codices and calendar systems. Sev-
eral years later, Vaillant (1937, 1938) published a correlation of
the archaeological and historical sequences in the Valley of Mex-
ico. Using Caso’s (1939, 1967) correlation of the Aztec and Chris-
tian calendars, Vaillant assigned absolute dates to artifact dumps
and suggested that the ceramic styles in the dumps reflected the
material culture of particular ethnic groups (e.g., Aztec I Black-
on-Orange pottery was made and used by the “Chichimecs”).
3
In the mid-1950s, the ethnohistorian H. B. Nicholson (1955:599–
601) critiqued Vaillant’s attempt to link material culture to spe-
cific ethnic groups as too simplistic (see also Nicholson 1971a).
Research on the periodization of Aztec ceramics showed that Vail-
lant’s correlation of Black-on-Orange types with 52-year-long ce-
ramic phases also was too simplistic (Table 1). Independent research
by José Luis Franco (1949:185–187) and Vaillant on the motifs of
Black-on-Orange ceramics established the basic chronology still
used today (Hodge 1998; Hodge and Minc 1990; Parsons 1966);
however, we now know that key types such asAztec I and Aztec II
Black-on-Orange are not chronologically distinct (see also Griffin
and Espejo 1947, 1950) but, rather, coeval (Hodge 1998; Whalen
and Parsons 1982).
4
Aztec I and Aztec II types date to the mid-
1Vaillant’s unpublished excavations include San Fransisco Mazapan,
Chiconautla, Nonoalco, “Los Melones” at Texcoco, and “El Corral” at
Santiago Ahuitzotla. His failure to publish his Classic and Postclassic
excavations is particularly puzzling after his diligence and speed in pro-
ducing high-quality site reports on the Formative sites. His notes and
major artifact collections are available for consultation at the American
Museum of Natural History (AMNH) in New York, and many scholars
have examined these over the years. Elson has been working on the ma-
terials from Chiconautla and Nonoalco (Elson 1999).
2Vaillant counted Alfonso Caso, Manuel Gamio, Sigvald Linné, Igna-
cio Marquina, and Eduardo Noguera among his friends and colleagues.
His notes and correspondence at the AMNH show that over dinner, visits
to one another’s excavations, and conferences they exchanged information
and ideas.
3Vaillant also suggested that the cyclical reconstructions of some reli-
gious structures were manifestations of New Fire activities. At Tenayuca
(excavated in the 1930s by Alfonso Caso, Eduardo Noguera, and Ignacio
Marquina),Marquina(1951:164–180) documentedsixsuccessiveconstruc-
tion episodes. Based on the ceramic types in the fill between each layer,
Vaillant (1941:81–82) suggested that the temple was refurbished every 52
years in conjunction with the celebration of the New Fire Ceremony.
4Both Vaillant and Franco suggested that Aztec I began in the mid-
thirteenth century, and that the beginning of the fifteenth century marked
the dividing line between Aztec II and III. But Franco suggested that
Vaillant’s 52-year ceramic periods were problematic because stylistic
changes were more gradual, especially from Aztec I to II, which Franco
thought coexisted from 1299 to 1351, and Aztec II to III, which he sug-
gested coexisted from 1403 to 1455.
160 Elson and Smith

twelfth century, and the transition from the Early Aztec period
(with the Aztec I and II types) to the Late Aztec period (with
Aztec III and IV types) occurred in the mid-fourteenth century.
5
A third difficulty with Vaillant’s model is that he seemed to
assume that only one archaeological dump would be created for
each New Fire Ceremony. This is not explicitly stated, but he
made a major effort to find one archaeological deposit for every
New Fire date listed in the historical sources. Although research-
ers have discarded Vaillant’s chronology of 52-year ceramic phases,
archaeologists and historians (Franco 1949:184; Nicholson
1955:600) have suggested that documenting contemporaneous
archaeological examples of dumps from different parts of the
valley could substantiate Vaillant’s hypothesis suggesting that
artifact dumps were a manifestation of the New Fire Ceremony.
Excavation and Content of Ritual Dumps
Two sites with proposed New Fire dumps are located in the Basin
of Mexico on the eastern (Nonoalco) and western (Chiconautla)
edge of Lake Texcoco. The third site, Cuexcomate, is in western
Morelos (Figure 2). The so-called Volador deposit excavated in
Mexico City by Noguera has been interpreted by some as a New
Fire dump. We provide a brief ethnohistoric description and an
overview of the archaeological work conducted at these sites, fo-
cusing on the excavations of the dumps. Then we describe the
contents of these features.
6
We employ a form-based functional classification of ceramics
with six major categories, each of which has numerous constituent
functional types. The six major categories are: serving vessels
(bowls, plates, cups, pitchers, and miniature vessels); kitchen ves-
sels (jars, griddles, basins, grinding bowls, and ladles); ritual ob-
jects (censers, figurines, sculptures, and a variety of small ceramic
objects such as bells, whistles, and pipes); production tools (spin-
dle whorls, spinning bowls, molds, and sherd scrapers); and other
domestic items (a catch-all category). This classification, which is
presented in more detail in Smith (2001a), is derived from Smith’s
form-based classification of Postclassic ceramics from Morelos
(Smith 2002).
Table 2 presents the ceramic inventories of the ritual dumps at
Chiconautla, Nonoalco, and Cuexcomate, organized by functional
types and categories. These inventories are fundamentally domes-
tic in character: Serving and kitchen vessels predominate, with low
frequencies of vessels and objects used in domestic ritual, textile
production, and other activities. They are nothing like the invento-
ries of ceramic vessels in Aztec public ceremonial contexts such as
the Templo Mayor offerings, the Volador deposit of Tenochtitlan,
or various burials and caches in public contexts at sites in Morelos
(Smith 2002). The ceramics in such public contexts are generally
wholevesselsinstead of sherds, and theyemphasizeeither drinking
vesselsor special-offering vessels (see Smith et al. [2002] for com-
parisons of vessel inventories from these public contexts).
The similarities in ceramic inventories between ritual dumps
and domestic ceramics are explored in Table 3. For Cuexcomate,
the dumps are compared with the houses from their patio group.
For Chiconautla, the dump is compared with two other units in the
elite residence, but for Nonoalco there is no good comparative
deposit. These tables also include the ceramic vessels from the
Volador deposit.
Chiconautla
Chiconautla is located about 17 km northwest of the Acolhua
capital, Texcoco. At the time of the Spanish Conquest, Chiconau-
tla was a town with its own tlatoani (Hodge 1984:Table 2–1).
Vaillant (1935b) excavated a structure at the site that he inter-
preted as an elite residence. Elson’s (1999) recent analysis of the
size and layout of the building and the kind of artifacts associated
with the structure supports Vaillant’s interpretation and suggests
that the building could have been the palace of the local tlatoani.
As such, it would have been the focus of administrative, religious,
and domestic activities.
Vaillant began excavations in the South House with a large
trench that cut into the dump. The focal point of the ritual dump
was the patio of the South House, but the dump spread through
much of the house (Figures 3 and 4). Some rooms were filled with
heavy concentrations of ash and large sherds. The dump extended
slightly north and east outside the South House Rooms 10, 12, and
5Aztec I might go back to the ninth or tenth century a.d. and in some
parts of the valley pre-date Aztec II (Parsons et al. 1996).
6It should be noted that only ceramic data were included in this analy-
sis. Vaillant did not record good provenience information for all obsidian
and stone at Chiconautla or Nonoalco, and the shell, bone, and plant re-
mains have not been studied in detail.
Figure 2. Map of central Mexico showing the location of sites mentioned
in the text.
Archaeological deposits from the Aztec New Fire Ceremony 161

13, where the refuse was intermingled with wall fall and debris
from the destruction of the South House.
In the field, Vaillant (1935b:27–29) noted that most of the Black-
on-Orange sherds in the dump were Period III–IV types, and analy-
sis of the ceramics supports his suggestion (Elson 1999:Table 1).
The density of the ceramic deposit and the condition of the mate-
rial in the South House distinguishes this part of the elite resi-
dence from the rest of the palace (see Vaillant 1937, 1941:Plate
32, second row, for illustrations of complete vessels found in the
South House).
7
It appears that the artifact dump was created dur-
ing a one-time event, then sealed under the demolished South
House; a platform subsequently was constructed over the house
remains.
There are more miscellaneous artifacts such as hollow and solid
female figurines and spindle whorls in the dump than in the Late
East House or the North House (Elson 1999:Tables 3 and 4).
Aztec female figurines were most likely used in household rituals
(Smith 2001b), and we have noted the destruction of these “idols”
as part of the New Fire celebrations.
Nonoalco
Vaillant’s notes provide much less information on his excavations
at Nonoalco, and there is little historical information about the
site. Calnek (1976:Map 20) represents the site as almost a neigh-
borhood of Tlatelolco. In a letter to Clarence Hay dated January 5,
1935, Vaillant stated that his excavations were located one hun-
dred yards east of the church of San Miguel Nonoalco (which
probably was constructed above a pre-Hispanic temple) and were
bordered on the north by the Calzada de Nonoalco (which in
pre-Hispanic times was the causeway to Azcapotzalco). Given
Vaillant’s description of the site, the location he dug roughly cor-
responds with a site that Sanders et al. (1979:Map 18) place on the
lake shore between Tlatelolco and Tlacopan, where the causeway
from Azcapotzalco joined the causeway connecting Tlatelolco and
Tlacopan.
By Vaillant’s day, this area was urbanized, and Vaillant took ad-
vantageof the little open space that wasleft by excavating trenches
under electrical towers that ran across the site (Vaillant 1935b:
Plate1). His firsttrenches were inthe lakebed andproduced almost
nomaterial.Aftermoving farthernorth,he opened the EastTrench,
whichmeasured 20 mlong on thenorth–southaxisand 3.5 m wide.
He did not locate the remains of a structure, but he did find the re-
mainsofa“canal” sealed under 1.2 m of stratigraphicdeposits(Fig-
ures 5 and 6) that was filled with black clay and contained whole
vessels and sherds that could be reassembled into whole or nearly
whole vessels (Vaillant 1941:Plate 32).
8
Vaillant states that most of the Black-on-Orange ceramics in
the Nonoalco artifact dump corresponded to his Period Late III.
Stratigraphically, above this deposit and slightly to the north he
found a midden resulting from gradual accumulation and architec-
tural remains, suggesting that he was digging near an area of
residences. The midden contained a high frequency of Black-
on-Orange pottery with naturalistic decorations, some of which
were clearly European-influenced. Thus, Vaillant suggested that
the ritual dump dated to the a.d. 1507 New Fire Ceremony—the
last one celebrated before the arrival of the Spaniards (letter to
Ignacio Marquina, January 24, 1936, Archives,Anthropology De-
partment, American Museum of Natural History).
We lack context for the artifact dump at Nonoalco, but the
kinds of ceramics in the dump and the frequency with which they
occur is similar to the other artifact dumps. The majority of the
artifacts are plain and decorated serving wares and kitchen wares
(bowls, jars, basins, and comals) with low frequencies of artifacts
used for ritual or production. Nonoalco and Cuexcomate Units
245 and 222 contain a higher frequency of censers than the other
7It is not possible to quantify the number of sherds per cubic meter
from the three structures; however, Vaillant’s field notes, drawings, scale
sections, and photographs make it clear that the volume of the excavated
area did not vary enough in the three locations to be significant (see also
Elson [1999:Table 2] for the size of the three structures).
8Using Vaillant’s notes and ceramic counts in the field, we have iden-
tified discrete proveniences associated with the dump. Very probably this
is only a portion of the material that could be classified as part of the
artifact dump, but Vaillant’s field notes are not always clear.
Ta b l e 2 . Ceramic inventories of ritual dumps (%total sherds)
Cuexcomate
Ceramic Category Chiconautla Nonoalco 245 222 224
Serving
Plain bowl 22.3 11.1 19.5 24.6 26.2
Decorated bowl 15.3 8.3 7.3 5.2 13.5
Tripod plate 19.5 12.4 0 0 0
Pitcher 1.1 1.4 0 .3 .4
Copa 1.7 2.9 .4 .3 .4
Eroded bowl 0 0 20.6 18.1 4.4
Kitchen
Jar 5.1 7.3 2.3 2.5 5.1
Basin 6.9 11.8 0 .6 .5
Comal 16.3 29 33.6 34.3 41
Salt vessel 2.8 2.9 3.8 5 2.2
Ladle .1 0 0 0 0
Ritual
Censer 1.3 8.7 9.2 6.9 1.9
Figurine 3.1 2 .4 0 .3
Temple model .2 .1 0 0 0
Brazier .1 0 .4 0 .6
Massive bowl .1 .1 0 0 0
Stamp .1 .1 0 0 .1
Whistle .3 0 0 0 .1
Rattle .2 .5 0 0 0
Pipe .2 .2 0 0 0
Bell .1 0 0 0 0
Flute .1 .1 0 0 0
Production
Spinning bowl .8 1 1.1 1.7 1.7
Spindle whorl 1.5 0 .8 .6 .9
Worked sherd .1 .2 0 0 .1
Other
Miniature vessel .8 0 0 0 .1
Tiny vessel 0 0 .4 0 0
Ball 0 .1 0 0 0
Bead 0 0 .4 0 0
Fragments .1 .1 0 0 .3
Total sherds 5,252 1,328 262 362 2,878
Note: All figures are counts of sherds except for the Jar category, which includes
only rim sherds. This unusual form of quantification is required due to the nature
of the ceramic data recorded by Vaillant.
162 Elson and Smith

Ta b l e 3 . Comparisons of ritual dumps and middens (%total sherds)
Plain
Serving Decorated
Serving Eroded
Bowls Kitchen Ritual Production Other Total
Sherds
Cuexcomate
Patio Group 3
House 1 (Unit 214) 46.3 5.9 8.2 34.3 3.5 .9 .9 341
House 2 (Unit 215) 26.1 8.6 19.7 42.1 1.4 1.9 .3 701
House 3 (Unit 270) 26.3 8 25.0 35.1 1.6 1.4 2.5 1,046
Dump (Unit 245) 19.5 7.6 20.6 39.7 9.9 1.9 .8 262
Patio Group 9
House (Unit 258) 30 10 8.1 46.2 1.7 2.2 1.7 689
Dump (Unit 222) 24.5 5.8 18.1 42.3 6.9 2.5 0 364
Patio Group 10
House 1 (Unit 201) 23.9 12.6 6.3 40.8 3 4.1 9.3 6,344
House 2 (Unit 202) 25 11.3 19.2 39 .6 3.1 1.8 639
House 3 (Unit 203) 20.3 17.1 17.7 41.6 2 .8 .4 502
Dump (Unit 224) 26.2 14.4 4.4 48.9 3 2.7 .5 2,878
Chiconautla
East 16.3 43.6 37.8 .6 .6 1.2 1,587
North 15 45.4 37.4 1.9 .3 0 2,075
South 25.4 29 40 3 1.4 1.2 21,807
Dump 24 40.5 32.3 1.5 .9 .9 4,882
Nonoalco
Dump 11.5 22.9 52.6 12 1 0 1,280
Volador .2 95.9 2.3 1.1 0 .4 897
Note: See note to Table 2 on the quantification of jars.
Figure 3. Plan of the elite residence at
Chiconautla.
Archaeological deposits from the Aztec New Fire Ceremony 163

dump locations. At Nonoalco, these long-handled censers were
found in a very good state of preservation. None is whole, but the
collection Vaillant bought back includes handles measuring 15 to
38 cm long, bowls that are one-third complete, and whole serpent
heads and bowtie appliqués commonly found at the ends of the
handles of censers. The role of incense burners in the New Fire
celebrations was discussed earlier.
Cuexcomate
Cuexcomate,asmallAztectownwithabout900inhabitantsinwest-
ernMorelos,wasoccupiedduringtheLatePostclassic period (Smith
1992;Smith and Heath-Smith 1994).Atthecenterof the settlement
was a public plaza surrounded by two elite residences, a small tem-
ple,anda special residential compound.Residentialareas were dis-
tributed along a sloping ridgetop on either side of the site center.
House walls are visible on the ground surface, and it was possible
tomaptheentirecommunitypriortoexcavation.Mosthousesformed
partsofpatiogroupsconsistingoftwotofourhousesarrangedaround
a patio. The ritual dumps were visible as circular scatters of rocks
locatedin the centersor at the edges of patio groups. These features
were called rock piles during excavation and analysis, and sub-
sequent artifact analyses and ethnohistoric comparisons suggested
their functions as ritual dumps (Smith 1992:369–373).
Ritual dumps were identified at five of the 25 patio groups at
Cuexcomate;threegroupshadonedump, one group had two dumps,
and one group had five dumps. “Site 3,” an isolated patio group in
a zone of agricultural terraces 1 km from Cuexcomate (probably a
farmstead), also had a ritualdump.Thevillage site of Capilco,3km
away, did not have any ritual dumps (Smith 1992). Figure 7 shows
two patio groups at Cuexcomate with ritual dumps that were exca-
vatedin1986.Group3hadoneritualdump(Unit245),andGroup10
had five dumps, one of which (Unit 224) was fully excavated. One
other dump, Unit 222 in Patio Group 9, was also excavated exten-
sively, and four other dumps were tested with small excavations.
Figure 8 (Unit 245) gives an idea of the appearance of most
ritual dumps at Cuexcomate prior to excavation; Figure 9 shows
the stratigraphy of Units 224 and 245. These features consist of a
layer of stones covering an unstratified deposit with very dense
broken domestic objects. The artifact deposits lie in shallow pits
excavated into the sandy clay B-horizon soils at Cuexcomate; in
some cases the pits extended into the C-horizon hardpan (known
as tepetate in central Mexico). Of the seven tested dumps, one had
a burial of a juvenile under the artifact deposit (Figure 9), and four
had caches or offerings of ceramic serving bowls located either
under the artifact deposit or adjacent to the dump.
Three characteristics of the artifact deposits in the ritual dumps
stand out in comparison to the middens adjacent to residential
Figure 4. Photo of South House excava-
tions at Chiconautla. (a) Rooms 3, 6, and
10, looking north; (b) Rooms 6, 10, 12,
and 13, looking northwest.
164 Elson and Smith

structures at Cuexcomate. First, the density of artifacts is much
higher than in middens. The mean ceramic density of Late Cuau-
hnahuac phase middens is 1,100 sherds/m
3
, compared with a mean
of 3,100 sherds/m
3
in the ritual dumps; for obsidian, the corre-
sponding figures are 43 pieces/m
3
in middens compared with 100
pieces/m
3
in the dumps. For a limited sample of excavated lots
(selected to illustrate sherd size; see later), the corresponding den-
sities are 3,130 sherds/m
3
in the ritual dump, Unit 224, and 1,700
sherds/m
3
in a sample of 52 midden lots (Table 4). The second
distinctive characteristic is that the sherds in the ritual dumps are
larger than those in middens (Table 4).
Third, it was possible to reassemble numerous whole or partial
ceramic vessels from the ritual dumps, and sherd mends often cut
across the dump deposit both vertically and horizontally. This
situation was not found in the middens, whose ceramic remains
were highly fragmented and could never be reassembled into whole,
or even partial, vessels. Limited quantitative support for this sug-
gestion is provided by the data in Table 5, which compares ratios
of sherd counts to minimal vessel counts in the Unit 224 ritual
dump and the nearest extensively excavated house, Unit 201. Min-
imum numbers of vessels were estimated by a procedure based on
rim counts (described in Smith 1992). The lower ratios for the
ritual dump indicate that there were fewer sherds per vessel in the
ritual dump—that is, the broken vessels are less highly frag-
mented than in middens.
9
A final observation about the ceramics of the ritual dumps is
that they always dated to the same ceramic phase as the middens
of the houses in their patio group. These characteristics of the
Cuexcomate ritual dumps and their contents suggest that they were
9The ratios in Table 5 substantially underestimate the differences be-
tween the two deposits due to data-recording decisions in the laboratory.
When sherds were glued together to form complete or partial vessel, they
were counted as a single sherd in the sherd counts. If the original counts of
unmended sherds were available, the ratios for Unit 224 would be far
lower than indicated in Table 5.
Figure 5. Photo and stratigraphy of the East trench at Nonoalco.
Archaeological deposits from the Aztec New Fire Ceremony 165

formed in single episodes in which large numbers of domestic
items were discarded into special pits, then covered with a layer of
rocks.
The ceramic inventories of the dumps are quite similar to do-
mestic inventories (Table 3). All of the major domestic functional
and form classes are present in the ritual dumps, and most of the
specific types of local and imported painted serving ware charac-
teristic of domestic inventories (Smith 2002) are also present in
the ritual dumps. Most of the quantitative differences between the
dumps and their nearby domestic deposits are relatively minor,
and where significant differences exist, they are not consistent in
all three cases. For example, the dumps in Patio Groups 3 and 9
Figure 6. Photo of excavations at Nonoalco (note the church cupola at far left).
Figure 7. Plans of Patio Groups 3 and
10 at Cuexcomate showing the loca-
tions of ritual dumps and the exca-
vated areas of two dumps, Units 245
and 224.
166 Elson and Smith

have more censers and other ritual items than the houses do, but in
Patio Group 10, the frequencies of ritual items are not very differ-
ent between the dump and the houses.
One difference between the ceramics of the Unit 224 dump and
domestic ceramics at Cuexcomate is the presence of several unique
vessels in the ritual dump. These include a globular jar with a
pedestal base and an oval opening in the side of the jar (perhaps an
unusual censer?) and a vessel with a flat, roughened base similar
to a comal, but with tall sloping walls like a basin. Either these
unusual vessels were deliberately deposited only in ritual dumps,
Figure 8. Photo of ritual dump at Cuex-
comate (Unit 245) prior to excavation.
Figure 9. Plan and profile of ritual
dumps at Cuexcomate (Units 224 and
225).
Archaeological deposits from the Aztec New Fire Ceremony 167

or such vessel forms may have been more widely distributed at
Cuexcomate but have not been identified due to the fragmentary
and eroded condition of the sherds in the middens.
The Volador Deposit from Tenochtitlan
In 1936–1937, Eduardo Noguera excavated an offering of approx-
imately 1,000 ceramic vessels in Mexico City that has been inter-
preted by some as a New Fire deposit. The vessels had been placed
inside a large stone platform located at the southeastern corner of
the Zócalo of Mexico City in what had been known as the Plaza
del Volador, presumably the pre-Hispanic location of the Aztec
volador ceremony. In the volador ceremony, several individuals,
whose feet were tied to long ropes wrapped around a tall pole,
leapt off the pole to “fly” in circles as the ropes unwound. The
ceramic vessels and a stone sculpture were found in a stone-lined
chamber. The excavation was never published in detail. In his
report to the Mexican government (reproduced in Solís Olguín
and Morales Gómez 1991:79–80), Noguera suggested that the of-
ferings related to the volador ceremony. He later published a brief
article (Noguera 1968) attributing the offering to the New Fire
Ceremony. The excavation was carried out soon afterVaillant had
excavated at Chiconautla and Nonoalco, and Vaillant and his Mex-
ican colleagues (including Noguera) had been discussing the iden-
tification of the deposits at those (and perhaps other) sites as New
Fire remains.
Noguera’s interpretation of the volador deposit as the remains
of a New Fire Ceremony has been accepted by most authors
(e.g., López Luján 1994:12; Solís Olguín and Morales Gómez
1991; Vega Sosa 1975:8), but there is little support for this spec-
ulation. We will show later that the ceramics from this deposit do
not at all resemble the remains from the ritual dumps described
earlier. In a recent paper, Smith et al. (2002) suggest that, in-
stead, the volador deposit contains the remains of elite feasting
activities, with no necessary connection to the New Fire Cer-
emony. Most of the volador ceramics are now stored in the Mu-
seo Nacional de Antropología e Historia; photos of each of 897
vessels are published by Solís and Morales (1991). Additional
vessels from the volador deposit, now in the collections of the
Field Museum of Natural History in Chicago, are illustrated in
McVicker (1992).
It is clear from Table 3 that the inventory of ceramic vessels in
the Volador deposit is quite different from domestic deposits and
from the ritual dumps described earlier. Although most of the
vessels in that deposit are forms that also occur in domestic con-
texts, the frequencies are radically different. Sixty percent of the
nearly 900 vessels illustrated in Solís and Morales (1991) are red
biconical drinking cups, a form that occurs in domestic deposits
and ritual dumps at a frequency of 1–2% (Table 2); other abundant
forms in the Volador collection are bi-level Black-on-Orange tri-
pod dishes (16%) and serving bowls painted with skulls and cross-
bones (9%). The few kitchen vessels in that deposit are mostly
molcajetes, or grater bowls.
Discussion
The ritual dumps described here contain the kinds of vessels ex-
pected from ethnohistoric descriptions of ritual dumps—basic do-
mestic wares in similar proportions to household middens. The
contents of the ritual dumps stand out in important ways, how-
ever. The condition of the ceramic assemblages in the dumps (large
sherds, some whole vessels, and vessels that can be reassembled)
and the manner in which the deposits were created (one-time events
that produced dense and unstratified contexts) suggest that they
are not middens. The locations of the Cuexcomate dumps within
patio groups suggest that the activities that created these features
involved the inhabitants of those groups. The fact that only five of
25 patio groups at the site had ritual dumps might be explained
either by variation in ritual practices within the community or by
specialization in ritual activities by the inhabitants of some patio
groups.
The dump at Chiconautla’s South House was a much larger
deposit (in terms of numbers of artifacts), suggesting that this
deposit may pertain to a larger group than the 15–20 members of
the Cuexcomate patio groups. This feature may have been created
as part of a public event—or, at least, an event in which members
of the community other than the immediate residents of the house
participated. The abundance of ash Vaillant found in the deposit
suggests another parallel between public state rituals and house-
hold rituals: the importance of sweeping, and of keeping and re-
newing fires. Possibly, the ash from many household hearths was
ritually swept and dumped into the deposit.
Can the Ritual Dumps Be Used for Dating?
Although the majority of Aztec Black-on-Orange ceramics in the
dumps date to the Late Aztec period (after a.d. 1350), this period
includes four New Fire dates: a.d. 1351, 1403, 1455, 1507. In
western Morelos this interval has been divided into two phases—
Early Cuauhnahuac and Late Cuauhnahuac (Smith and Doershuk
1991)—each of which includes two New Fire dates. One ritual
dump at Cuexcomate produced a radiocarbon sample whose
2-sigma range (a.d. 1397–1486) includes the New Fire dates a.d.
1403 and 1455 (sample no. SMU-2363, carbonized wood; see
Ta b l e 4 . Ceramic density and sherd size in ritual dumps and middens
at Cuexcomate
Ritual Dump: Middens:
Measure Mean SD Mean SD
Ceramic density 3,129.8 948.1 1,702.4 1,367.1
Bowl sherd size 6.4 2.1 4 1.1
Jar sherd size 14.4 6.3 7.6 2.5
Notes: The ritual dump sample consists of all 13 excavated levels in unit 224. The
midden sample consists of 52 Late Cuauhnahuac excavated levels with more than
100 sherds. SD, standard deviation.
Ta b l e 5 . Ratios of sherds to minimal vessel counts at Cuexcomate
Category Ritual Dump
(Unit 224) Midden
(Unit 201)
Bowls 7 10.8
Jars 34.4 48.4
Comals 18.7 30.7
Total ceramics 13.3 17.3
168 Elson and Smith

Smith and Doershuk [1991]). Based on the presence of some Early
Cuauhnahuac ceramics in this Late Cuauhnahuac deposit, Smith
suggested that the New Fire date of a.d. 1455 (shortly after the
transition from Early to Late Cuauhanhuac) is more likely than the
next such date, a.d. 1507. Vaillant suggested a date of a.d. 1507
for the dump at Nonoalco. Without an alternative method of chro-
nological refinement, ritual dumps cannot be used for dating; how-
ever, the seven examples we have discussed all fall within the
Late Aztec Period, suggesting that this kind of celebration took
place simultaneously in different parts of the Aztec empire.
AZTEC RELIGION, IDEOLOGY, AND CULTURAL
EVOLUTION
The Aztec New Fire Ceremony provides a rare opportunity to
integrate our knowledge of Aztec state and domestic ritual, as well
as to compare ethnohistoric and archaeological data. This is one of
the few examples outside of the Templo Mayor where specific
archaeological deposits can be related with confidence to ritual
activities described in the ethnohistoric sources.
History of the Aztec New Fire Ceremony
The iconographic symbol of the New Fire Ceremony was a fire
drill with flames. The earliest clear-cut example of this occurs on
a stone relief from Xochicalco, Morelos (Figure 10A), in which
the symbol is accompanied by calendrical glyphs 1 Rabbit (a year
designation) and 2 Snake (a probable day name). This stone is
carved in the Xochicalco sculptural style of a.d. 650–900 (Hirth
2000), and César Sáenz (1967) considers it a commemoration of
the first New Fire Ceremony. Emily Umberger (1987a, 1996) has
shown that Mexica artists deliberately imitated the sculptural styles
of Xochicalco, Tula, and Teotihuacan. She notes that in Tenochti-
tlan, “another sculpture with a date surrounded by a Xochicalco-
style frame was carved in connection with the New Fire Ceremony
of 1507, and the archaizing style recalls the site where the Tenochca
considered that their calendar was initiated and the first new fire
was lighted” (Umberger 1996:94).
The method of historical transmission of the New Fire Cer-
emony from Xochicalco to Tenochtitlan is not clear. Nine New
Fire Ceremonies are noted in the Aztec historical sources; the
Mapa Sigüenza, for example, mentions eight New Fires (Boone
2000:166–173). These are listed in Table 6, which we take from
Figure 10. Iconographic depictions of
the New Fire Ceremony. (A) The ear-
liest known New Fire, from Xochi-
calco (after Sáenz 1967:12). This piece is
designated No. B1 in the catalog of
Smith and Hirth (2000). (B) The final
Mexica New Fire Ceremony in the year
2 Reed (A.D. 1507) , during the reign of
Motecuhzoma Xocoyotzin (from the
Codex Mendoza, after Berdan and
Anawalt 1992:Volume 4, Folio 15v). (C)
Stone sculptural representations of the
52-year bundles, Xiuhmolpilli, exca-
vated at the Templo Mayor (after Batres
1979 [1905]:153).
Ta b l e 6 . New Fire Ceremonies in Mexica historical accounts
Number Year Year Location
1 1 Rabbit 1090 Teocolhuacan
2 1 Rabbit 1142 Coatepec (Tollan)
3 1 Rabbit 1194 Huitzcol Apazco
4 1 Rabbit 1246 Tecpayocan
5 1 Rabbit 1298 Chapultepec
6 2 Reed 1351 Tenochtitlan
7 2 Reed 1403 Tenochtitlan
8 2 Reed 1455 Huixachtecatl
9 2 Reed 1507 Huixachtecatl
Source: Tena (1987:98)
Archaeological deposits from the Aztec New Fire Ceremony 169

Tena (1987:98), who provides citations to the primary sources that
mention each event (see also Boone 2000:223–224). At least two
centuries separate the Xochicalco New Fire and the earliest Mex-
ica example. The interval between these dates, the Early Postclas-
sic period, was the time of Toltec civilization. Sáenz (1967:13)
and Tena (1987:92) suggest that the Mexica must have adopted
the New Fire Ceremony from the Toltecs. In spite of some spec-
ulation about the possible presence of the New Fire Ceremony at
Toltec-period Tula (Coggins 1987, 1989), we know of no direct
iconographic or other evidence for the ceremony at Tula (de la
Fuente et al. 1988).
10
Instead of assuming that the Mexica of Tenochtitlan adopted
the New Fire Ceremony from a specific ancient city (Xochicalco
or Tula), it may be more fruitful to suggest that the practice was
widespread in Postclassic northern Mesoamerica, perhaps em-
ployed by most peoples. The Mexica adopted this practice and
used it for their own imperial ends.
11
There are many examples of
new fire being drilled in the Mixtec codices (Boone 2000:94–160)
and depictions of the drilling of New Fire in the Historia Tolteca-
Chichimeca (e.g., Boone 2000:180–181; Kirchhoff et al. 1976:
folios 32v–33r) shows the presence of the practice among the
eastern Nahua peoples of southern Puebla. To the Mixtecs, the
New Fire Ceremony had a meaning very different from that of the
Aztecs. As depicted in the Mixtec codices and lienzos, the new
fire was drilled by priests or nobles as one of the rituals of the
foundation of a new town (Boone 2000:94–160; Furst 1990). Like
the Aztec version, the Mixtec New Fire Ceremony was a rite of
renewal and beginning, but the focus was on the start of a new
town and polity, not a new cosmic era.
According to the reconstructions by Tena (1987) and Umberger
(1987b), the Mexica moved the year of celebration from 1 Rabbit
(1350) to 2 Reed (1351) for their sixth New Fire Ceremony, and 2
Reed became the normal date for these events through the Spanish
Conquest.
12
The Mexica historical sources indicate that some New
Fire Celebrations were carried out as many as several years after
the target date, due to the unstable situation of the Mexica in their
early centuries in the Basin of Mexico (see Tena 1987:91–93).
According to the Anales de Tlatelolco (1948:18), once the Mexica
had founded Tenochtitlan in a.d. 1325 (2 House), they celebrated
an impressive New Fire Ceremony in Chapultepec. This may have
been done to erase the memory of the two former ceremonies,
each of which had to be delayed because of humiliating military
defeats in the years 1 Rabbit (1246 and 1298). At this point in their
history, the Mexica—like the Mixtecs or eastern Nahua peoples—
may have been using the New Fire Ceremony as a ritual of foun-
dation. By the mid-fifteenth century, however, the New Fire
Ceremony at Tenochtitlan had taken on the imperial associations
noted earlier.
Evenaftera.d. 1428 and the formationoftheTripleAlliance,the
date used by the Mexica of Tenochtitlan to mark the New Fire Cer-
emonywasnotnecessarilythesamedateusedbyotherethnicgroups.
The Chichimec used 9 Tecpatl; the Acolhua used 1 Tecpatl; the To-
tomihuaque used 7 Acatl; and the Tepaneca, Culhuaque, and Mex-
icaused2Acatl(LópezAustin1973:99; Marcus 1992:117–118). In
fact, Brundage (1985:36) suggests that in 1507, the imperial lead-
ers decided to celebrate the New Fire in the month of Panquetzal-
iztli(late November–earlyDecember),associatedwithQuetzalcoatl
insteadofIzcalli(late January–mid-February)becauseofthatgod’s
association with the origin of the Fifth Sun (Figure 10b).
In an alternative to Tena’s and Umberger’s interpretation, Has-
sig (2001:38–47, 114), follows a suggestion in the Codex
Telleriano-Remensis (Quiñones Keber 1995:Folio 41v) that the
change from 1 Rabbit to 2 Reed was done for the New Fire
Ceremony of 1507:
In this year [1 Rabbit, 1506] they were to bind the years ac-
cording to their count, and because it was always a difficult
year for them, Motecuhzoma changed it to two reeds [i.e., the
year 2 Reed, 1507] [Quiñones Keber 1995:274].
The Codex goes on to state that “in this year [2 Reed, 1507] they
finished the church of the new fire” on top of Mount Huixachtlan
(Quiñones Keber 1995:274). Hassig (2001:47) takes this to mean
that the state ritual was moved at that time from the Templo Mayor
to Mount Huixachtlan, although this is not stated in the Codex or
other sources. He suggests that scribes later went back and retro-
spectively changed the dates of earlier New Fire celebrations in
the historical records to 2 Reed in order to give the Mexica cer-
emony greater legitimacy. Regardless of exactly when the New
Fire date was changed, and regardless of when (or whether) the
lighting of the fire was moved from Tenochtitlan to Mount Huix-
achtlan, the sources are clear that by 1507 the ritual had become a
major political event for the Mexica state. Provincial peoples were
well aware of the New Fire Ceremony by 1507; the Xicotepec
Codex from the Sierra Norte de Puebla, for example, mentions the
1507 New Fire celebration but not the 1455 event, even though its
span of historical coverage includes both dates (Stresser-Péan 1995).
Top-Down and Bottom-Up Perspectives
Mexica state ideology, as expressed through the works of the chron-
iclers, clearly emphasized the imperial nature of the New Fire
Ceremony. The new fire, controlled by the Mexica king, was dis-
tributed from Tenochtitlan throughout the empire, symbolizing
the subordination of provincial peoples to the capital (D. Carrasco
1999:96–114). As a way to legitimize their imperial rule, the Mex-
10 Several authors have argued for the presence of the New Fire Cer-
emony at other prominent Mesoamerican sites based on tenuous icono-
graphic evidence. Von Winning (1979), for example, suggests that it was
present at Teotihuacan, and his argument has been seconded by Langley
(1997). Coggins (1987, 1989) argues for the New Fire Ceremony at Chi-
chen Itza as well as at Teotihuacan, Kaminaljuyu, Becan, Zaculeu, and
Tula. Coggins bases her argument on the presence of isolated elements
assumed to be part of the ritual complex of the New Fire Ceremony,
including the bundle of canes, fire drill, the number seven, and the Kan
cross and pyrite mirrors (1987:446).
In contrast to these isolated elements without context, the depictions in
the Aztec and Mixtec codices and lienzos show either the New Fire element
linked spatially to date glyphs or individuals engaged in lighting a new fire.
We do not view the isolated iconographic depictions at other sites as strong
evidence for the practice of the New Fire Ceremony at these sites. Furst
(1992) takes a very broad perspective and compares the Aztec New Fire
Ceremony to numerous renewal rituals involving fire found among native
cultures throughout the New World. These comparisons are interesting, but
we do not consider them useful for understanding the history or context of
the Aztec New Fire Ceremony. There is no evidence for New Fire Ceremo-
nies, or other 52-year celebrations, in Classic Maya inscriptions or in Post-
classic Maya codices (John Justeson, personal communication 2001).
11 Hassig (2001:118) suggests that the New Fire Ceremony was “ob-
served only in political centers which controlled the calendar.”According
to Hassig, Tenochtitlan was the only polity in the Basin of Mexico to
celebrate the New Fire Ceremony, and outside the Basin, Tlaxcalla also
held New Fire celebrations (Hassig 2001:97).
12 This interpretation has been questioned by Edward Calnek, who
suggests that “there is little (or no) acceptable documentary evidence for
the shift” (letter to Michael Smith, October 28, 2000, in the possession of
Michael Smith).
170 Elson and Smith

ica copied specific sculptural pieces, themes, and styles from Te-
otihuacan, Xochicalco, and Tula (Umberger 1987a, 1996). The
fact that Mexica sculptors, as part of this imperial artistic pro-
gram, deliberately made visual reference to the ancient Xochi-
calco New Fire carving (Umberger 1996:94) reinforces the imperial
associations of the ceremony.
13
Carved-stone year bundles were
buried in various deposits in and around the Templo Mayor (Fig-
ure 10c). From this top-down perspective, one might be tempted
to interpret the presence of New Fire ritual dumps at provincial
sites as evidence for the penetration of imperial ideology down to
the provincial household level. As an alternative, however, we
suggest that a bottom-up perspective—that the Mexica appropri-
ated a more widespread popular ritual for imperial purposes—
better fits the archaeological and ethnohistoric data.
One of the characteristics of Aztec imperialism is that the em-
pire made little or no effort to impose Aztec religion on conquered
peoples. Instead, Aztec religion was actively inclusive, incorpo-
rating numerous gods and rituals from conquered peoples. Idols of
foreign gods were brought to Tenochtitlan and displayed in a spe-
cial temple. Even state-linked rituals such as human sacrifice,
central to Aztec imperial ideology (D. Carrasco 1999), were an-
cient Mesoamerican practices going back at least as far as Teoti-
huacan (Cabrera Castro et al. 1991; Taube 1992) and the Classic
Maya (Taube 1988; Welsh 1988). Human sacrifice is documented
for the Middle Postclassic (pre-Mexica) Nahua cultures of More-
los (Lagunas R. and Sánchez 1972), showing that its use among
the Nahua peoples long predated the Aztec empire.
The Xochicalco inscription (Figure 10a) shows the antiquity of
the New Fire Ceremony (or, at least, the iconographic expression
of that ceremony) in central Mexico. Cuexcomate is located only
3 km from Xochicalco, and structures at the Postclassic site incor-
porate carved stones that were almost certainly looted from Xochi-
calco (Smith 1992:177). Thus, it is entirely possible that the New
Fire Ceremony was a local tradition in western Morelos, main-
tained through the four or five centuries that separated the fall of
Xochicalco from the founding of Cuexcomate; unfortunately, the
lack of Early Postclassic excavations in western Morelos, and the
small number of excavated Middle Postclassic contexts, prevent
the evaluation of this suggestion. One of the six dated ritual dumps
at Cuexcomate pertains to the Early Cuauhnahuac period. Be-
cause this period predates the conquest of Morelos by the Aztec
empire, it is difficult to argue that the Late Cuauhnahuac New Fire
celebrations were somehow imposed by the Aztec empire.
Taking a broader perspective, we know that some rituals asso-
ciated with New Year ceremonies had a wide distribution in Me-
soamerica. For example, the general pattern of Postclassic Yucatec
Maya New Year (Uayeb) ceremonies included the manufacture of
pottery idols representing the god of the coming year (and some-
times other deities) that were paraded to different parts of town
and presided over ceremonies at the residence of a lord chosen to
host feasts and ceremonies. At some celebrations, ceramic censers
were discarded in deposits outside of towns, and people started
over with new censers (Landa 1941). On the first day of the New
Year, daily household articles were replaced during a solemn ren-
ovation ceremony, and houses were swept clean; however, there is
no evidence that the Maya assigned any special significance to the
52-year cycle (Sharer 1994:551). Excavations at Santa Rita Corozal,
Belize, have uncovered caches of ceramic figurines in elite resi-
dences or platforms that never appear in other deposits, and the
excavator suggests that the figurines and associated paraphernalia
were made especially to be deposited during New Year ceremo-
nies (Chase 1985, 1986).
The New Fire Ceremony expressed widespread Mesoamerican
beliefs independent of an imperial ideology: the notion of a “liv-
ing” fire that had to be carefully tended; the ability of inanimate
objects to be receptacles for divine essences; and the conception
of time as cyclical, including the belief that periods of renewal are
dangerous. The behavior associated with the creation of artifact
dumps—ritually rekindling the hearth fire, sweeping and remov-
ing ash, throwing out and purposely smashing household goods
and figurines, and communal feasting and drinking—was not dic-
tated by imperial ideology.
It is likely that the timing of New Fire celebrations, or similar
rituals of renewal, was traditionally the prerogative of local rulers.
City-states in the Basin of Mexico, before the institutionalization
of Tenochtitlan hegemony, had their own priests and probably
kept their own calendars (and presumably New Fire dates), which
would have been a critical way of maintaining local history and
identity (Marcus 1992:118; see Hassig 2001:97, 118, for a dis-
senting view). The Aztec imperial order, however, created fantas-
tic New Year ceremonies, like the one described on Mount
Huixachtlan, on their own auspicious dates that far outshone any
other city-state’s New Fire Ceremony.
We suggest that the existence of the New Fire Ceremony at
diverse settlements—Tenochtitlan, Chiconautla, Nonoalco, and
Cuexcomate—should not be seen simply as a case of imperial
domination through the imposition of religious rituals on subject
peoples. When viewed from a bottom-up perspective on Aztec
cultural evolution, many of the underlying behaviors and beliefs
associated with the New Fire Ceremony, and the presence of the
ceremony at four diverse locations, can be seen as a manifestation
of the cultural unity that linked the Aztec peoples of the Basin of
Mexico, Morelos, and probably other nearby areas, as well (Smith
1996). This Aztec cultural unity was the foundation for the Colonial-
period Nahua culture of central Mexico (Lockhart 1992). Local
elites may have actively emphasized diversity in the timing and
expression of religious behaviors and beliefs as a means of main-
taining local political and ethnic identities. By placing household
rituals such as the distribution of fire to houses and temples from
Mount Huixachtlan in a hierarchical political framework, how-
ever, the emperor cross-cut city-states’ boundaries, undermined
the political power of local elites, and co-opted religious ideology
for a program of imperial legitimization and glorification.
The imperial use of the New Fire Ceremony can be summa-
rized as follows: The Mexica, like other Nahuatl and Mixtec peo-
ples, engaged in a ritual of lighting a new fire, perhaps combining
a symbol of foundation with a celebration of the completion of a
52-year cycle. Once they had forged an empire, however, the Mex-
ica rulers gave the ritual the full trappings of a cosmic imperial
celebration, including performance on top of a mountain, human
sacrifice, participation of the emperor, and ceremonial distribu-
tion of the fire throughout the empire, all within a mythological
framework that connected the empire to the calendric cycles of the
cosmos. Exclusive attention to the propagandistic written sources
from Tenochtitlan might suggest that this was an imperial ritual
imposed on the populace of the hinterland, but our consideration
13 Another example of the transformation of existing myth and ritual
by the Mexica to promote their imperial ideology is provided by Graulich
(1997:76–79), who argues that the Mexica added a fifth creation era, or
“sun,” to the older Mesoamerican mythological theme of four previous
suns. He suggests, “When the Mexica took over and reestablished a great
empire in Mexico, they claimed that their arrival was the beginning of the
Fifth Sun” (Graulich 1997:79).
Archaeological deposits from the Aztec New Fire Ceremony 171

of the archaeological remains of local New Fire celebrations shows
that such a perspective is misleading. The forms of Aztec ritual
were worked out through a dialectic between traditional local and
household practices, on one hand, and innovative imperial poli-
cies, on the other. The combination of ethnohistoric and archaeo-
logical data on the New Fire Ceremony helps us see this dialectic
more clearly and thus contributes to a more comprehensive model
of Aztec religion and society.
RESUMEN
La ceremonia del año nuevo llamado Xiuhtlalpilli en Nahuatl ocurrió cada
52 años y marcó un nuevo siglo en el calendario azteca. La ceremonia del
Nuevo Fuego es una de las pocas ceremonias aztecas documentadas en la
arqueología y en la historia. Los cronistas españoles describieron dicha
ceremonia como una celebración imperial que servía para renovar el tiempo
cósmico. Al nivel local, los celebrantes demarcaron el evento renovando
sus enseres domésticos. El arqueólogo George C. Vaillant propuso que
ciertos basureros de artefactos en sitios aztecas constituían evidencia de la
renovación de los enseres domésticos documentados en las celebraciones
locales. Describimos dos basureros de artefactos domésticos excavados
por Vaillant, uno en Chiconautla y uno en Nonoalco, hasta ahora no de-
scritos en la literatura publicada y un basurero de artefactos domésticos
excavado por Smith en Cuexcomate, Morelos. Demostramos que el con-
texto y el contenido de los basureros apoyan el hipótesis de Vaillant. Nues-
tros datos sugieren que la ceremonia del Nuevo Fuego fue un ritual de
mucha antigüedad y muy extenso que en el México postclásico fue apropi-
ado por el imperio azteca como parte de su programa de dominación
política y legitimación ideológica.
ACKNOWLEDGMENTS
We thank Elizabeth Boone, Edward Calnek, Joyce Marcus, Donald
McVicker, and two anonymous referees for comments on an earlier draft
of this paper. Elson thanks Charles S. Spencer of the Department of An-
thropology, American Museum of Natural History, for permission and
encouragement in the study of George Vaillant’s archaeological collec-
tions at the museum. Smith’s fieldwork at Cuexcomate was supported by
the National Science Foundation and Loyola University of Chicago.
REFERENCES
Anales de Tlatelolco
1948 Anales de Tlatelolco: Unos Anales Históricos de la Nación Mex-
icana, y Códice de Tlatelolco. Edited by Heinrich Berlin. Antiguo
Librería Robredo, Mexico City.
Batres, Leopoldo
1979 (1905) Exploraciones arqueológicas en la Calle de las Escaleril-
las. In Trabajos Arqueológicos en el Centro de la Ciudad de México,
edited by Eduardo Matos Moctezuma, pp. 111–170. Instituto Nacio-
nal de Antropología e Historia, Mexico City.
Berdan, Frances F., and Patricia R.Anawalt (editors)
1992 The Codex Mendoza. 4 vols. University of California Press,
Berkeley.
Berdan, Frances F., Richard E. Blanton, Elizabeth H. Boone, Mary G.
Hodge, Michael E. Smith, and Emily Umberger
1996 Aztec Imperial Strategies. Dumbarton Oaks, Washington, DC.
Boone, Elizabeth H.
2000 Stories in Red and Black: Pictorial Histories of the Aztecs and
Mixtecs. University of Texas Press, Austin.
Boone, Elizabeth H. (editor)
1987 The Aztec Templo Mayor. Dumbarton Oaks, Washington, DC.
Broda, Johanna
1982 La fiesta azteca del Fuego Nuevo y el culto de las Pléyades. In
Time and Space in Ancient American Cosmovision, edited by Franz
Tichy, pp. 129–157. Lateinamerican Studien Vol. 10. Universitat
Erlangen–Nurnbery, Munich.
Broda, Johanna, Davíd Carrasco, and Eduardo Matos Moctezuma
1987 The Great Temple of Tenochtitlan: Center and Periphery in the
Aztec World. University of California Press, Berkeley.
Brumfiel, Elizabeth M.
1996 Figurines and the Aztec State: Testing the Effectiveness of Ideo-
logical Domination. In Gender and Archaeology, edited by Rita P.
Wright, pp. 143–166. University of Pennsylvania Press, Philadelphia.
Brundage, Burr C.
1985 The Jade Steps: A Ritual Life of the Aztecs. University of Utah
Press, Salt Lake City.
Cabrera Castro, Rubén, Saburo Sugiyama, and George L. Cowgill
1991 The Templo de Quetzalcoatl Project at Teotihuacan: A Prelimi-
nary Report. Ancient Mesoamerica 2:77–92.
Calnek, Edward E.
1976 The Internal Structure of Tenochtitlan. In The Valley of Mexico:
Studies of Pre-Hispanic Ecology and Society, edited by Eric R. Wolf,
pp. 287–302. University of New Mexico Press, Albuquerque.
Carrasco, Davíd
1991 To Change Place:Aztec Ceremonial Landscapes. University Press
of Colorado, Boulder.
1999 City of Sacrifice: The Aztec Empire and the Role of Violence in
Civilization. Beacon Press, Boston.
Carrasco, Pedro
1999 The Tenochca Empire of Ancient Mexico: The Triple Alliance of
Tenochtitlan, Tetzcoco, and Tlacopan. University of Oklahoma Press,
Norman.
Caso, Alfonso
1939 La correlación de los años Azteca y Cristiano. Revista Mexicana
de Estudios Antropológicos 3:11–45.
1967 Los calendarios prehispánicos. Instituto de Investigaciones
Históricas, Universidad Nacional Autónoma de México, Mexico
City.
Chase, Diane Z.
1985 Ganned but Not Forgotten: Late Postclassic Archaeology and
Ritual at Santa Rita Corozal, Belize. In The Lowland Maya Postclas-
sic, edited by Arlen F. Chase and Prudence M. Rice, pp. 104–125.
University of Texas Press, Austin.
1986 Social and Political Organization in the Land of Cacao and Honey:
Correlating the Archaeology and Ethnohistory of the Postclassic Low-
land Maya. In Late Lowland Maya Civilization: Classic to Postclas-
sic, edited by Jeremy A. Sabloff and E. Wyllys Andrews V, pp. 347–
378. University of New Mexico Press, Albuquerque.
Códice Tudela
1980 Códice Tudela, José Tudela de la Orden. 2 vols. Ediciones Cul-
tura Hispanica, Madrid.
Coggins, Clemency Chase
1987 New Fire at Chichen Itza. In Memorias del Primer Coloquio
Internacional de Mayistas, 5–10 de agosto de 1985, pp. 427–482.
Universidad Nacional Autónoma de México, Mexico City.
1989 A New Sun at Chichen Itza. In World Archaeoastronomy: Se-
lected Papers from the 2nd Oxford International Conference on Ar-
chaeoastronomy, Held at Merida, Yucatan, Mexico, 13–17 January
1986, edited by Anthony F. Aveni, pp. 260–275. Cambridge Univer-
sitiy Press, New York.
de la Fuente, Beatriz, Silvia Terjo, and Gutiérrez Solana
1988 Escultura en Piedra de Tula: Catálogo. Instituto de Investiga-
ciones Estéticas, Universidad Nacional Autónoma de México, Mex-
ico City.
172 Elson and Smith

Durán, Fray Diego
1967 Historia de las Indias de Nueva España. Translated byAngel M.
Garibay K. 2 vols. Porrúa, Mexico City.
Elson, Christina M.
1999 An Aztec Palace at Chiconautla, Mexico. Latin American Antiq-
uity 10:151–167.
Franco C., José Luis
1949 Algunos problemas relativos a la cerámica Azteca. El México
Antiguo 7:162–208.
Furst, Jill Leslie McKeever
1990 Rulership and Ritual: Myth and the Origin of Political Authority
in Mixtec Pictorial Manuscripts. In Circumpacifica: Festschrift für
Thomas S. Barthel, vol. 1, edited by Bruno Illius and Matthias Laub-
scher, pp. 123–141. Peter Lang, Frankfurt.
1992 Aztec New Fire Ritual: AWorld Renewal Rite. Journal of Latin
American Lore 18:29–36.
Gómez de Orozco, F.
1945 Costumbres, fiestas, enterramientos y diversas formas de pro-
ceder de los indios de Nueva España. Tlalocan 2(1):38– 63.
Graulich, Michel
1997 Myths of Ancient Mexico. Translated by Bernard R. Ortiz de
Montellano and Thelma Ortiz de Montellano. University of Okla-
homa Press, Norman.
1999 Fiestas de los pueblos indígenas: Ritos aztecas, las fiestas de las
veintanas. Instituto Nacional Indígenista, Mexico City.
Griffin, James B., and Antonieta Espejo
1947 Alfarería correspondiente al último período de ocupación nahua
del Valle de México. Memorias de la Academía Mexicana de la His-
toria 6(2):131–147.
1950 La alfafería del último período de ocupación del Valle de Méx-
ico, II: Culhuacan, Tenayuac, Tenochtitlan, y Tlatelolco. Memorias
de la Academía Mexicana de la Historia 9(1):118–169.
Guilliem Arroyo, Salvador, Saturnino Vallejo Zamora, and Ángeles Me-
dina Pérez
1998 Ofrenda en el Templo Mayor de México–Tlatelolco. Arqueología
19:101–118.
Hassig, Ross
2001 Time, History and Belief in Aztec and Colonial Mexico. Univer-
sity of Texas Press, Austin.
Hirth, Kenneth G. (editor)
2000 Archaeological Research at Xochicalco. Volume 1, Ancient Ur-
banism at Xochicalco: The Evolution and Organization of a Pre-
Hispanic Society. Volume 2, The Xochicalco Mapping Project.2
vols. University of Utah Press, Salt Lake City.
Hodge, Mary G.
1984 Aztec City-States. Memoirs of the Museum of Anthropology Vol.
18. University of Michigan, Ann Arbor.
1998 Archaeological Views ofAztec Culture. Journal of Archaeolog-
ical Research 6:197–238.
Hodge, Mary G., and Leah D. Minc
1990 The Spatial Patterning of Aztec Ceramics: Implications for Pre-
hispanic Exchange Systems in the Valley of Mexico. Journal of Field
Archaeology 17:415–437.
1991 Aztec-Period Ceramic Distribution and Exchange Systems. Re-
port to the National Science Foundation, Washington, DC.
Kirchhoff, Paul, Lina Odena Güemes, and Luis Reyes García (editors)
1976 Historia tolteca-chichimeca. Instituto Nacional de Antropología
e Historia, Mexico City.
Lagunas R., Zaid, and Carlos Serrano Sánchez
1972 Decapitación y desmembramiento corporal en Teopanzolco, Mo-
relos. In Religión en Mesoamérica. XII Mesa Redonda, edited by
Jaime Litvak King and Noemí Castillo Tejero, pp. 429–434. So-
ciedad Mexicana de Antropología, Mexico City.
Landa, Diego de
1941 Landa’s Relación de las Cosas de Yucatan. Translated by Alfred
M. Tozzer. Papers of the Peabody Museum of Archaeology and Eth-
nography Vol. 18. Harvard University, Cambridge.
Langley, James
1997 Teotihuacan Incensarios: The “V” Manta and Its Message. Teo-
tihuacan Notes: Internet Journal for TeotihuacanArchaeoalogy and
Iconography 1(3):1–11.Available from: http://archaeology/la.asu.edu /
teo/notes/JL/notes1_3.htm.
León-Portilla, Miguel
1963 Aztec Thought and Culture: A Study of the Ancient Náhuatl Mind.
University of Oklahoma Press, Norman.
Lockhart, James
1992 The Nahuas After the Conquest: A Social and Cultural History
of the Indians of Central Mexico, Sixteenth Through Eighteenth Cen-
turies. Stanford University Press, Stanford, CA.
López Austin, Alfredo
1973 Hombre-Díos: Religión y política en el mundo Náhuatl. Univer-
sidad Nacional Autónoma de México, Mexico City.
1997 Tamoanchan, Tlalocan: Places of Mist. Translated by Bernard
R. Ortiz de Montellano and Thelma Ortiz de Montellano. University
Press of Colorado, Niwot.
López Luján, Leonardo
1994 The Offerings of the Templo Mayor of Tenochtitlan. Translated
by Bernard R. Ortiz de Montellano and Thelma Ortiz de Montellano.
University Press of Colorado, Niwot.
Marcus, Joyce
1992 Mesoamerican Writing Systems: Propaganda, Myth, and His-
tory in Four Ancient Civilizations. Princeton University Press, Prince-
ton, NJ.
Marquina, Ignacio
1951 Arquitectura Prehispánica. Instituto Nacional deAntropología e
Historia, Mexico City.
Matos Moctezuma, Eduardo
1988 The Great Temple of the Aztecs. Thames and Hudson, New York.
McVicker, Donald E.
1992 México: La Visión del Cosmos: Three Thousand Years of Cre-
ativity. Mexican Fine Arts Center Museum, Chicago.
Moedano, Hugo
1951 Ce Acatl igual Ome Acatl, como fin de Xiuhmolpilli. Revista
Mexicana de Estudios Antropológicos 12:103–131.
Motolinía, Fray Toribio de Benavente
1951 Motolinía’s History of the Indians of New Spain. Translated by
Francis Borgia Steck. Academic of American Franciscan History,
Washington, DC.
Nicholson, H.B.
1955 Native Historical Traditions of Nuclear America and the Prob-
lem of Their Archeological Correlation. American Anthropologist
57:594– 613.
1971a Pre-Hispanic Central Mexican Historiography. In Investiga-
cionesContemporáneasSobreHistoriade México, pp. 38–81.ElCole-
giode México and University of Texas Press, Mexico City and Austin.
1971b Religion in Pre-Hispanic Central Mexico. In Archaeology of
Northern Mesoamerica, Part 1, edited by Gordon F. Ekholm and
Ignacio Bernal, pp. 395–446. Handbook of Middle American Indi-
ans, vol. 10. University of Texas Press, Austin.
Noguera, Eduardo
1968 Ceremonias del Fuego Nuevo. Cuadernos Americanos 158(3):
146–151.
Olmos Frese, Laura del
1999 Análisis de la ofrenda 98 del Templo Mayor de Tenochtitlan.
Colección Científica, Vol. 384. Instituto Nacional de Antropología e
Historia, Mexico City.
Parsons, Jeffrey R.
1966 The Aztec Ceramic Sequence in the Teotihuacan Valley, Mexico.
2 vols. Ph.D. dissertation, Department of Anthropology, University
of Michigan, Ann Arbor.
Parsons, Jeffrey R., Elizabeth Brumfiel, and Mary Hodge
1996 Developmental Implications of Earlier Dates for Early Aztec in
the Basin of Mexico. Ancient Mesoamerica 7:217–230.
Parsons, Mary H.
1972 Aztec Figurines from the Teotihuacán Valley, Mexico. In Mis-
cellaneous Studies in Mexican Prehistory, edited by Michael W.
Spence, Jeffrey R. Parsons, and Mary H. Parsons, pp. 81–164. An-
thropological Papers Vol. 49. Museum of Anthropology, University
of Michigan, Ann Arbor.
Quiñones Keber, Eloise
1995 Codex Telleriano-Remensis: Ritual, Divination, and History in
a Pictorial Aztec Manuscript. University of Texas Press, Austin.
Sáenz, César A.
1967 El Fuego Nuevo. Serie Historia, Vol. 18. Instituto Nacional de
Antropología e Historia, Mexico City.
Sahagún, Fray Bernardino de
1950–1982 Florentine Codex, General History of the Things of New
Spain. 12 books. Translated and edited byArthur J.O. Anderson and
Charles E. Dibble. School of American Research and University of
Utah Press, Santa Fe and Salt Lake City.
Archaeological deposits from the Aztec New Fire Ceremony 173

Sanders, William T., Jeffrey R. Parsons, and Robert S. Santley
1979 The Basin of Mexico: Ecological Processes in the Evolution of a
Civilization. Academic Press, New York.
Sharer, Robert J.
1994 The Ancient Maya. 5th ed. Stanford University Press, Stanford,
CA.
Smith, Michael E.
1992 Archaeological Research at Aztec-Period Rural Sites in More-
los, Mexico. Volume 1, Excavations and Architecture/Investigaciones
Arqueológicas en Sitios Rurales de la Epoca Azteca en Morelos, Tomo
1, Excavaciones y Arquitectura. University of Pittsburgh Memoirs in
Latin American Archaeology, Vol. 4. University of Pittsburgh,
Pittsburgh.
1996 The Aztecs. Blackwell Publishers, Oxford.
2001a La cerámica Postclásica de Morelos. In La Producción Alfar-
era en el México Antiguo, edited by B. Leonor Merino Carrión and
Angel García Cook,. Colección Científica. Instituto Nacional de
Antropología e Historia, Mexico City.
2001b Domestic Ritual at Aztec Provincial Sites in Morelos. In Do-
mestic Ritual in Ancient Mesoamerica, edited by Patricia Plunket.
University of California at Los Angeles Institute ofArchaeology, Los
Angeles.
2002 Tlahuica Ceramics: The Aztec-Period Ceramics of Morelos, Mex-
ico. IMS Monographs, Vol. 13. Institute for Mesoamerican Studies,
State University of New York, Albany.
Smith, Michael E., and John F. Doershuk
1991 Late Postclassic Chronology in Western Morelos, Mexico. Latin
American Antiquity 2:291–310.
Smith, Michael E., and Cynthia Heath-Smith
1994 Rural Economy in Late Postclassic Morelos: AnArchaeological
Study. In Economies and Polities in the Aztec Realm, edited by Mary
G. Hodge and Michael E. Smith, pp. 349–376. Institute for Meso-
american Studies, State University of New York, Albany.
Smith, Michael E., Jennifer Wharton, and Jan Marie Olson
2002 Aztec Feasts, Rituals, and Markets: Political Uses of Ceramic
Vessels in a Commercial Economy. In Pots as Political Tools: The
Culinary Equipment of Early Imperial States in Comparative Per-
spective, edited by Tamara Bray, in press.
Smith, Virginia, and Kenneth G. Hirth
2000 A Catalog of Carved Monuments and a Guide to the Visual
Characteristics of Xochicalco’s Art Style. In The Xochicalco Map-
ping Project, edited by Kenneth G. Hirth, pp. 17–56. Archaeological
Research at Xochicalco, Vol. 2. University of Utah Press, Salt Lake
City.
Solís Olguín, Felipe R., and Davíd A. Morales Gómez
1991 Rescate de un rescate: Colección de objetos arqueológicos de el
Volador, ciudad de México. Catálogo de las colecciones arqueológi-
cas de Museo Nacional de Antropología. Instituto Nacional de
Antropología e Historia, Mexico City.
Stresser-Péan, Guy
1995 El Códice de Xicotepec: estudio e interpretación. Gobierno del
Estado de Puebla, Centro Francés de Estudios Mexicanos y Cen-
troamericanos, Fondo de Cultura Económica, Mexico City.
Taube, Karl A.
1988 A Study of Classic Maya Scaffold Sacrifice. In Maya Iconogra-
phy, edited by Elizabeth P. Benson and Gillett G. Griffin, pp. 331–
351. Princeton University Press, Princeton, NJ.
1992 The Temple of Quetzalcoatl and the Cult of SacredWar at Teo-
tihuacan. Res: Anthropology and Aesthetics 21:53–87.
1993 Aztec and Maya Myths. University of Texas Press, Austin.
2000 The Turquoise Hearth: Fire, Self Sacrifice, and the Central Mex-
ican Cult of War. In Mesoamerica’s Classic Heritage: From Teoti-
huacan to the Aztecs, edited by Davíd Carrasco, Lindsay Jones, and
Scott Sessions, pp. 269–340. University Press of Colorado, Niwot.
Tena, Rafael
1987 El calendario mexica y la cronografía. Instituto Nacional de
Antropología e Historia, Mexico City.
Umberger, Emily
1987a Antiques, Revivals, and References to the Past in Aztec Art.
RES: Anthropology and Aesthetics 13:62–105.
1987b Events Commemorated by Date Plaques at the Templo Mayor:
Further Thoughts on the Solar Metaphor. In The Aztec Tempo Mayor,
edited by Elizabeth H. Boone, pp. 411–450. Dumbarton Oaks, Wash-
ington, DC.
1996 Art and Imperial Strategy in Tenochtitlan. In Aztec Imperial
Strategies, by Frances F. Berdan, Richard E. Blanton, Elizabeth H.
Boone, Mary G. Hodge, Michael E. Smith, and Emily Umberger,
pp. 85–106. Dumbarton Oaks, Washington, DC.
Vaillant, George C.
1930 Excavations at Zacatenco. Anthropological Papers, Vol. 32, No. 1.
American Museum of Natural History, New York.
1931 Excavations at Ticoman.Anthropological Papers, Vol. 32, No. 2.
American Museum of Natural History, New York.
1935a Excavations at El Arbolillo. Anthropological Papers, Vol. 35,
No. 2. American Museum of Natural History, New York.
1935b Untitled manuscript. In the collection of the American Mu-
seum of Natural History, New York.
1937 History and Stratigraphy in the Valley of Mexico. Scientific
Monthly 44:307–324.
1938 A Correlation ofArchaeological and Historical Sequences in the
Valley of Mexico. American Anthropologist 40:535–573.
1941 Aztecs of Mexico: Origin, Rise and Fall of the Aztec Nation.
Doubleday Doran, Garden City.
Vaillant, George C., and Suzannah B. Vaillant
1934 Excavations at Gualupita. Anthropological Papers, Vol. 35, No. 1.
American Museum of Natural History, New York.
Vega Sosa, Constanza
1975 Forma y decoración en vasijas de tradición Azteca. Colección
Científica, Vol. 23. Instituto Nacional de Antropología e Historia,
Mexico City.
Von Winning, Hasso
1979 The Binding of the Years and the “New Fire” at Teotihuacan.
Indiana 5:15–27.
Welsh, W.B.M.
1988 An Analysis of Classic Lowland Maya Burials. BAR Inter-
national Series, Vol. S409. British Archaeological Reports, Oxford.
Whalen, Michael E., and Jeffrey R. Parsons
1982 Ceramic Markers Used for Period Designations. In Prehispanic
Settlement Patterns in the Southern Valley of Mexico: The Chalco–
Xochimilco Region, by Jeffrey R. Parsons, Elizabeth Brumfiel, Mary
H. Parsons, and David J. Wilson, pp. 385–459. Memoirs of the Mu-
seum of Anthropology, Vol. 14. University of Michigan, Ann Arbor.
174 Elson and Smith