3. The Aztec Occupation at Chiconautla, Mexico
Authors: Ananda Cohen, Christina M. Elson
Dr. George Vaillant, the fourth curator of Mexican and Central American Archaeology at the AMNH, was one of the great American archaeologists of his time. Vaillant was born in 1901 in Boston, Massachusets and educated at Harvard University, where he developed a strong interest in Mesoamerican archaeology. He received his PhD in 1927 and immediately thereafter became an assistant curator at the American Museum of Natural History (AMNH). In 1930 he became an associate curator, a position he held for ten years. In 1941 he was appointed the Director of the University of Pennsylvania Museum. Vaillant also served in several government posts.
George and Susannah Vaillant
Between 1926-1936, Vaillant (later accompanied by his wife Susannah Clapp Vaillant) excavated extensively throughout Mexico, completing nearly nine years of fieldwork at sites ranging from Formative to Postclassic in time period. Although many aspects of archaeological theory and field methods have changed over the past century, Vaillant stands out as a pioneer. His fieldwork was comprehensive and systematic, making it possible to conduct new research on data from archaeological excavations he headed over seventy years ago. He took detailed notes, documented architectural structures with numerous drawings and photographs, and coded artifacts by provenience. In Central Mexico, Vaillant excavated at Zacatenco, Ticomán, El Arbolillo, San Juan Teotihuacan, Gualupita, San Francisco Mazapan, Santiago Ahuitzotla, Chiconautla, Nonoalco, and Los Melones. He was a responsible scholar and before his untimely death in 1945 he published informative monographs on his work at Zacatenco, Ticoman, El Arbolillo, and (with Susannah Vaillant) Gualupita. His book The Aztecs of Mexico was a major synthesis of archaeological and ethnohistorical information describing cultural evolution in Central Mexico.
Vaillant conducted significant excavations at the Postclassic site of Chiconautla. In recent years Dr. Christina Elson has directed a project to analyze and publish Vaillant's work at this site (see Elson 1999; Elson and Smith 2001). Christine Chen (Columbia University), Ananda Cohen (University of Michigan), Heather Heineman (New York University), Katie McGurl (Boston University), Nina Neivens (Barnard), and Amy Nyack (Fordham University) all have worked with her to analyze different aspects of the collection, contributing to a better understanding of Chiconautla in a political, cultural, and economic context. This article summarizes some of the major finding thus far of that research, focusing on the ceramic objects from Chiconautla.
Chiconautla is located in the northeastern part of the Basin of Mexico. The pre-Hispanic town was on the shore of Lake Texcoco. After the fall of Teotihuacan around A.D. 650, the Basin of Mexico became a politically fragmented region. As populations shifted, new centers emerged to head competing polities. The first substantial occupation at Chiconautla took place during the Early Postclassic Period (A.D. 950-1150) when it grew to become one of many large villages in the Basin of Mexico. [Chronology Table]. Chiconautla's location placed it at an important crossroads for maritime and overland trade among Basin of Mexico polities, the Teotihuacan Valley, and the Gulf Coast lowlands. [Map of Central Mexico]
In the Middle Postclassic Period (A.D. 1150 - 1350), Chiconaulta emerged as the head town of a polity called an altepetl in Nahuatl. The hereditary noble ruler of an altepetl was called the tlatoani. By the Late Postclassic Period (A.D. 1350-1520) Chiconautla was one of the 50 altepetl that existed in the Basin of Mexico. After the formation of the Triple Alliance in 1428, Chiconautla's tlatoani was under the political control of Texcoco, a city ruled by the famous Nezahualcoyotl. Chiconautla paid tribute to Texcoco in the form of labor and building materials.
Fieldwork at Chiconautla
Vaillant first visited Chiconautla in October 1934 to explore the site and determine the boundaries of the occupation. According to the Relación Geográfica for Chiconautla written in 1557, the town had four neighborhoods. One of these was called Calpulpan Yacanqui in Nahuatl (or "casería principal" in Spanish) which means "noble residence." Vaillant asked local people where this neighborhood was and they pointed him to an area they called "Casas Reales" or "royal houses."
He returned again in February 1935 and conducted his first excavations on the mounds of "Casas Reales." Excavations into the northern mound, which he called Area A, produced no architecture. Excavations into the southern mound, which he called Area B, produced the remains of an Aztec palace (tecpan in Nahuatl). Characteristic of archaeological field methods of his time, Vaillant dug a series of large trenches at the site and then subsequently divided these into smaller excavation units as he uncovered the remains of architecture. Although Vaillant never published a monograph on the site, he left behind extensive documentation of his findings in the form of photographs, field notes, maps of the site, and a very large collection of artifacts including ceramics, obsidian, bone, metal, stone, and ethnobotanical remains. The collection required many years of organization and archival work before the materials could be analyzed.
Vaillant left an architectural drawing of the Aztec palace recording the location of the provenience units he excavated. For each one of the 184 provenience units, he coded and tabulated all the excavated ceramics. After computerizing the ceramic data base, Elson was able to use these data and Vaillant's analysis of the construction stages of the palace to construct a comprehensive view of the Postclassic occupation of Chiconautla. Her examination supported Vaillant's original claims that he excavated the palace of the ruler of Chiconautla. With funds from the Foundation for the Advancement of Mesoamerican Studies, Inc. a project directed by Elson and Dr. Deborah Nichols of Dartmouth College is underway to conduct Instrumental Neutron Activation Analysis on sherds from the site. The results will shed light on Chiconautla's role within a broader economic and regional context.
Palace Architecture at Chiconautla
The architectural form, building materials, and decorative features distinguish the structure Vaillant excavated at Chiconautla as the remains of a tlatoani's palace. The overall form of rooms built on platforms around enclosed patios is typical of elite residences. Volcanic bedrock (tepetate) and lava boulders finished at the corners with cut stone were used to build the platforms. The rooms of the palace had stone foundations and adobe brick walls plastered on both sides. Floors were plastered. Wooden beams may have been used as door jambs and to support the roof.
Vaillant found the remains of large pumice and ceramic cones that probably were attached to platforms and placed in ornamental positions above doorways. Susan Evans has analyzed pictorial depictions of palaces and suggests that these elements are associated with elite residences. Other elements associated with elite residences found in the Chiconautla palace are stone fireplaces (tlacuilli), sweatbaths (temazcalli), and mud or plaster covered granaries (cuexcomatl).
The construction of Chiconautla's palace can be divided into a distinct early phase and late phase. In the early phase, the palace consisted of a several sets of rooms placed on platforms. The eastern rooms were situated around an enclosed patio. The western rooms were attached to a large, open platform. The slightly detached southern construction consisted of another set of rooms around an enclosed patio. During the Late Aztec Period (A.D. 1450-1520) the palace was expanded. The late phase construction takes the shape of a set of enclosed rooms around a sunken patio flanked on the west and south by large platforms. The western platform clearly was the focus of ceremonial activities.
When he excavated the zone around the northwest corner of the platform, Vaillant found an abundance of charcoal, obsidian blades, and incense burners-probably because these objects had been dumped there after they were used in ceremonies atop the platform. Vaillant also found a number of fragments from large ceramic tiles. These are the remains of crenelations, which are known to have adorned the tops of temples. Finally, this area also contained many elongated cones of stone and baked clay which originally had been affixed to the platform's exterior.
Ceramic Vessels and Chronology
In Mesoamerica, ceramics were formed by hand through a coil technique. At Chiconautla many ceramics (particularly utilitarian wares like jars, comals, and bowls) were produced locally but other vessels (often fancy serving vessels) were obtained through trade. Ceramics were of critical importance in the domestic sphere for preparing, cooking, serving, and storing food and drink. Ceramic analyses allow archaeologists to make hypotheses about (1) the length of a site's occupation and (2) the socio-economic status of its inhabitants.
In the field, Vaillant analyzed over 100,000 sherds and coded each as pertaining to one of nearly 200 ceramic types. He defined types based primarily on form and decoration. Although he did not bring back every sherd to the AMNH, he did bring back a representative sample of each type. Some types are represented by many sherds, some by only a few.
Ceramics helped Vaillant determine the occupational sequence of the site. In his initial excavations in Area A (which produced 14,696 sherds), Vaillant uncovered a high proportion of Mazapan ceramics, indicating that the site was first extensively occupied during the Early Postclassic Period (A.D. 950-1150). The most common Mazapan wares are Red on Buff wares that are characterized by thick or thin red lines painted in straight or wavy repeating patterns on the interior or exterior of a vessel. Most Red on Buff ceramics in the AMNH type collection are bowls, but there are a few plates. Red on Buff ceramics account for 21% of the ceramics in Area A excavations. Undecorated jar rims and bodies account for another 21% of the ceramic assemblage. Another 7% of the sherds were coded as Coyotlatelco types (a ceramic type dated to circa A.D. 650-950), indicating some settlement here during the Epiclassic Period. No structural remains were found in this area, so possibly these early houses were built of more perishable materials like wattle and daub.
Black on Orange Wares
A few contexts in Area B also provided data on the Mazapan and Early Aztec Periods of occupation. In the West House rooms, an early structure whose remains were sealed below later phases of construction, Mazapan Period ceramics were found mixed with Early Aztec Period ceramics like Black on Orange II and Black-and-White/Red I bowls. Many of these Black-and-White/Red I bowls are decorated with large triangular areas delimited by wide or narrow diagonal lines. In contexts dating to the Early Aztec Period occupation, between 42-52% of the sherds are from utilitarian vessels (jar rims and handles, plain bowls, and comals) while 19-26% of the sherds are decorated serving wares.
In the Early Aztec (A.D. 1150-1350) and Late Aztec (A.D. 1350-1520) Periods, Black on Orange ceramics dominate as the primary decorated type. Vaillant's excavations at several Aztec sites led him to define four phases (Phases I-IV) of Black on Orange ceramics based on decorative motifs. He thought these decorative types were chronological markers, but recently Jeffrey Parsons his colleagues have shown that Black on Orange I and II coexist and date primarily to the Early Aztec Period and that Black on Orange III and IV coexist and date primarily to the Late Aztec Period. Vaillant coded very few Black on Orange I at Chiconautla, which is not surprising since Leah Minc, Mary Hodge and their colleagues have found that this type occurs primarily in the southern part of the basin.
Ceramic Types and Motifs, Production, and Elite Feasting Activities
In the Late Aztec Period structures that Vaillant excavated, he found utilitarian wares, ceremonial vessels, and fancy decorated ceramics. Here we want to present information on some of the types Vaillant found and what they can tell us about food production and feasting. Texcoco Fabric Marked sherds are one type present in great quantity-Vaillant coded about 3,000 of these in Late Aztec Period deposits. Jeffrey Parsons has documented that these vessels probably were used to produce and transport salt throughout the Valley of Mexico. Salt was a highly valued commodity in ancient Mexico: it provides a critical source of iodide for the human diet, can be used to preserve food, can be used as an agent for making soap, and is important as a textile mordant. The Relación Geográfica for Chiconautla states that the town's residents collected salt from the northeastern shore of Lake Texcoco. The high proportion of Texcoco Fabric Marked sherds suggests that salt played a prominent role in the local economy.
Black on Orange Motifs
Common decorative types found in the palace include Black on Orange III/IV, Black-and-White/Red II, and Black/Red II. Black on Orange III/IV wares occur as plates, bowls, molcajetes (tripod grinding bowls), basins, and bottles/pitchers. These decorated wares are the most common types of Aztec serving vessels and would have been used by rich and poor. Many of the examples in the type collection have a line emanating from the rim of the vessel to the beginning of the interior design.
Sometimes the line takes the form of a zig-zag motif, less frequently it appears as a neat triangle. Mary Hodge and Leah Minc have associated these motifs with the Tenochtitlan zone of production. Another motif that appears on some sherds is the spiral with petals on the vessel near rim, which has been associated with the southern production zone (Chalco-Xochimilco) of the Basin of Mexico. It is likely that Chiconautla's location on the lakeshore allowed its residents to acquire some of their decorated serving wares from other areas of the Basin of Mexico, even though politically the site would have been most closely tied to Texcoco.
Black-and-White/Red II and Black/Red II primarily occur as bowls. Most of these serving bowls are thin walled (less than 5mm thick) and only about 5-8 cm tall. Black-and-White/Red II and Black/Red II vessels would be found in greater frequencies in wealthier households. For example, in the rooms 6, 7, 9, and 10 of the North House, Vaillant coded a total of 3,695 sherds. Of these, 10% are Black on Orange, 13% are Black-and-White/Red, and 7% and Black/Red. We interpret these data as suggesting that the palace's inhabitants used a relatively high frequency of Aztec Redwares. Black-and-White/Red II sherds have decorations including braids, flowers, shield designs, scrolls and other motifs. Black/Red II sherds have several design sets. Some vessels have straight or slanting parallel lines on the exterior often with large flowers or petal-spirals. On others, the main decorative element is spiral petals and wind motifs or "swoosh" and "knotted bundle" designs.
Vaillant identified some kinds of vessels that may have been used in ritual activities and elite feasting including incense burners, hourglass-shaped cups, and globular cups with ring stand bases.
Incense burners are bowls with long handles. Illustrations in the Florentine Codex show that incense burners were used to conduct rituals both in homes and in temples. In the section on palace architecture, we described that the zone around the northwest corner of the west platform contained an abundance of fragments of broken incense burners and suggested that they probably had been used in ceremonies taking place on the platform and then discarded.
Pulque is a highly intoxicating drink produced from the sap of the maguey plant. The Aztecs had laws about who could drink pulque and when it could be drunk. Often, it was consumed ritually. Special hourglass-shaped cups may have been fashioned to hold pulque, which could have been served from pitchers. At Chiconautla, most of these hourglass-shaped cups as well as pitchers that could have been used to serve the beverage are Black/Red ceramic types.
Yet another product consumed primarily by elites was cacao. Only wealthy people had access to cacao, which had to be imported from tropical climates. Cacao was used a medium of exchange and also prepared as a drink. Michael Smith has reviewed pictorial documents and has suggested that globular cups were used for the purpose of consuming cacao-based drinks. Examples found at Chiconautla are very finely made polychrome vessels, which could have been imported from the Chalco-Cholula region.
Other kinds of ceramic materials
In addition to ceramic vessels, Vaillant excavated a range of other ceramic artifacts including objects such as spindle whorls, flutes, whistles, rattles, and stamps. Spindle whorls provide insight on the local economy. Jeffrey and Mary Parsons have conducted studies showing that based on weight, whorls weighing under 11 grams are best suited for spinning cotton; whorls weighing between 11 and 30 grams fall into the optimal range for producing fine maguey fiber; and whorls weighing over 30 grams are best-suited for spinning corse maguey fibers. Frederic Hicks has reviewed the role of cloth in the Aztec economy. Generally, commoners wore clothes woven out of maguey while cotton clothing was reserved for nobles. The textiles produced for men and women's clothing include loincloths, capes, shirts, skirts, and the huipil (a wide-sleeved woman's pullover garment). Maguey and cotton were also woven to produce shrouds, bags, armor, and bedding. Ethnohistorical documents tell us that spinning was an activity performed in the household by women of all social positions. At Chiconautla, the majority of whorls from excavated contexts (182 out of 224) are whorls that weigh less than 11 grams. Eight whorls weigh between 11 and 30 grams and 34 whorls weigh over 30 grams. In the Late Aztec Period, cotton clothing was in great demand. Garments were produced to fulfill tribute demands, to exchange for exotic goods, and to use in gift-exchange. Besides spindle whorls, cloth production could have incorporated artifacts like prismatic obsidian blades for making spinning implements and cutting cloth and salt, which was used as a mordant in dying cloth.
Sound devices made of ceramic including rattles, flutes, whistles, rasps and bells also were found at Chiconautla. A study by Robert Stevenson describes the role music played in Aztec society. It was an important accompaniment to rituals like prayers and sacrifices. Smoking pipes and figurines also commonly are classified as ritual items. Aztec pipes, used to smoke tobacco, are known to have been made of ceramic and bone.
Figurines found at Chiconautla represent humans (both males and females), animals, and objects like models of temples. Vaillant suggested that most figurines represent rain and water deities or female deities associated with fertility and probably were used within the domestic sphere of ritual. Christine Chen recently has reanalyzed the Chiconautla figurines and her results support Vaillant's original assertion. Out of 509 figurine fragments, 91% (462) are fragments of human figures. While only 209 of the 462 human fragments are identifiable by gender, 90% (188) of these are female figures.
Nina Neivens recently has studied many small objects from the site including the ceramic stamps. These mold made objects have geometric, zoomorphic, and naturalistic motifs. One interesting animalistic form that is found on five stamps is the 'Fire Serpent' or xihuhcoatl form which consists of a serpent with arrows motifs below it and emanating from its mouth. The arrows may represent rays of sunlight or fire. Seven other stamps show highly intricate patterns that may imitate the patterns in woven textiles. Since stamps are fairly common at archaeological sites, and there are few examples of pottery impressed with stamps, Neivens suggests that stamps probably were used primarily with paint as body decoration or to decorate clothing.
In sum, many lines of evidence drawn from architectural elements and ceramic data that support the interpretation of the building Vaillant excavated as an Aztec palace, likely the royal palace of the tlatoani, and shed light on some of the activities that took place within the compound.
• The form of the residence consisting of rooms on platforms around enclosed patios.
• The quality of the building materials such as stone for platform facing and wall foundations and plaster to coat wall and cover floors.
• The combination of ritual areas like the west platform and residential areas in one structure.
• The presence of decorative elements like stone and ceramic cones and crenelations that are known from ethnohistoric documents to have decorated elite residences and temples.
• The presence of storage features, stone fireplaces, and sweatbaths within the structure.
• The proportions of different ceramic types consumed by the residents suggest they used a relatively high frequency of Aztec Redwares, not just Black on Orange wares.
• The presence of hourglass-shaped cups and globular cups indicate that feasting and the consumption of pulque and cacao drinks took place in the compound.
• The presence of musical instruments, incense burners, and pipes also indicate that group-oriented ceremonial activities took place in the compound, possibly under the direction of the local ruler.
• The presence of figurines and spindle whorls provide evidence that ritual and production activities took place, likely directed by women residing in the compound.
The data Vaillant acquired show us that the Chiconautla palace was the location of relatively private domestic functions (like food preparation, spinning, and rituals involving figurines), and more group-oriented rituals involving incense burning, music, and drinking. Finally, we will consider in depth one provenience dating to the Late Aztec Period in order to show how a contextual analysis of all the material Vaillant found there can shed new light on Aztec culture.
The New Fire Ceremony at Chiconautla
One important ceremonial activity was the New Fire ceremony, which was conducted every 52 years when the 365-day secular and 260-day ritual calendars combined to form a cycle similar to our century. It is one of the few indigenous ceremonies in Mexico from which we have both archaeological and documentary evidence. Recently, Christina Elson and Michael Smith have discussed the archaeological implications of this ceremony.
Sahagún's Drawing of New Fire Ceremony and Temples
The Spanish friar Bernadino Sahagún provides the most extensive discussion of the ceremony in the Florentine Codex. The days lading up to the New Fire Ceremony were thought to be very dangerous and uncertain. This would have been a very disconcerting time for local people. On the eve of the new century, ceremonies were conducted at important state temples and shrines to insure the renewal of the world. Local people, who had extinguished all of their fires, had to wait in darkness until the "New Fire" generated though state ceremonies was distributed from temples to households. In preparation, families cleaned out their homes, swept out their hearths, and broke their household implements. Elson and Smith have used ethnohistorical evidence to suggest that these actions were done to eliminate potentially undesirable properties or essences that objects could become infused during the dangerous end of year period. New items replaced the old ones as a symbol of rebirth and renewal.
Sahagún's drawings show domestic wares like jars, pitchers, and cooking griddles as examples of objects that would have been broken and discarded during the ceremony. His drawing and descriptions inform us that human "idols" also were discarded. In another passage, Sahagún tell us that incense burners were used in New Fire celebrations and then thrown into the fire and broken.
Vaillant's early model for evaluating whether a ceramic assemblage represented a New Fire Ceremony deposit has been modified by Elson and Smith and tested with data from the sites of Chiconautla, Nonoalco, and Cuexcomate in Morelos. The basic criteria are as follows: the vast majority of vessels in a New Fire dump should be reconstructable; the artifacts in dumps should be the same as those found in a typical household (jars, tortilla griddles, cooking pots, cups, plates, and bowls); artifacts should constitute a single unstratified deposit; and the dumps should be located near residences.
At Chiconautla there is ample evidence to support the suggestion that a New Fire Ceremony took place in conjunction with the remodeling of the South House into a large platform. The dump was centered in the patio of the house which was filled with ash, large sherds and reconstructible vessels. The objects deposited in the dump mirror the domestic and ritual assemblage of the palace: serving wares and cooking vessels, spindle whorls, censers, figurines, rattles, whistles, and pipes. The evidence from Chiconautla illustrates the information documented in Sahagún by showing that people discarded food serving and preparation vessels and objects associated with feasting and ritual.
Because Vaillant carefully conducted fieldwork, preserved the artifacts he uncovered, and prepared fieldnotes, photographs, and maps, we can use the data from Chiconautla to formulate new ideas about Aztec society and culture. Some analyses, like an archaeological evaluation of the New Fire Ceremony, are possible because we now have comparative data from other sites in Mexico. Others, like an examination of Chiconautla's role in exchange networks, are possible due to th development of analytical techniques like Instrumental Neutron Activation Analysis.
|Anderson, Arthur J. O. and Charles E. Dibble, trans. and eds.|
|1950 - 1982||Florentine Codex: General History of the Things of New Spain, 13 vols. Santa Fe and Salt Lake City: School of American Research and University of Utah.|
|Burkhart, Louise M.|
|1997||Mexica Women on the Homefront: Housework and Religion in Aztec Mexico. In Indian Women of Early Mexico edited by Susan Schroeder, Stephanie Wood, and Robert Haskett, pp. 25-54. University of Oklahoma Press, Norman and London.|
|Elson, Christina M.|
|1999||An Aztec Palace at Chiconautla, Mexico. Latin American Antiquity 10: 151-167|
|Elson, Christina M. and Michael E. Smith.|
|2001||Archaeological Deposits from the Aztec New Fire Ceremony. Ancient Mesoamerica 12: 157-174.|
|1991||Architecture and Authority in an Aztec Village: Form and Function of the Tecpan. In Land and Politics in the Valley of Mexico, edited by H.R Harvey, pp. 63-92. University of New Mexico Press, Albuquerque.|
|Cloth in the Political Economy of the Aztec State. In Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith, pp. 90-111. Institute for Mesoamerican Studies, State University of New York, Albany.|
|Minc, Leah D., Mary G. Hodge, and James Blackman|
|1994||Stylistic and Spatial Variability in Early Aztec Ceramics: Insights into Pre-Imperial Exchange Systems. In Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith, pp. 133-174. Institute for Mesoamerican Studies, State University of New York, Albany.|
|Nichols, Deborah L., Elizabeth M. Brumfiel, Hector Neff, Mary Hodge, Thomas H. Charlton and Michael D. Glassock.|
|2002||Neutrons, Markets, Cities, and Empires: A 1000-Year Perspective on Ceramic Production and Distribution in the Postclassic Basin of Mexico. Journal of Anthropological Archaeology 21: 25-82.|
|Parsons, Jeffrey R.|
|1994||Late Postclassic Salt Production and Consumption in the Basin of Mexico: Some Insights from Nexquipayac. In Economies and Polities in the Aztec Realm, edited by M. G. Hodge and M. E. Smith, pp. 43-72. Institute for Mesoamerican Studies, State University of New York, Albany.|
|2001||The Last Saltmakers of Nexquipayac, Mexico: An Archaeological Ethnography. Anthropological Papers 92. Museum of Anthropology, University of Michigan.|
|Parsons, Jeffrey R. and Mary H. Parsons|
|1990||Maguey Utilization in Highland Central Mexico. Anthropological Papers 82. Museum of Anthropology, University of Michigan.|
|Parsons, Jeffrey R., Elizabeth M. Brumfiel, and Mary G. Hodge|
|1996||Developmental Implications of Earlier Dates for Early Aztec in the Basin of Mexico. Ancient Mesoamerica 7:217-230.|
|1996||The Aztecs. Basil Blackwell: Oxford.|
|2003||Decorated Serving Vessels from Postclassic burial offerings in Morelos. Mexicon: XXV (1): 3.|
|Smith, Michael and Jennifer Wharton|
|2003||Aztec-style pitcher from a Late Postclassic burial offering in Morelos. Mexicon: XXV (1):2-3.|
|1968||Music in the Aztec and Inca Territory. University of California Press, Berkeley.|
|Townsend, Richard F.|
|1992||The Aztecs. Thames and Hudson, London.|
|Vaillant, George C.|
|1962||The Aztecs of Mexico. Doubleday, New York.|