CHAPTER THIRTEEN
STORAGE AND ADMINISTRATION IN THE AZTEC EMPIRE
José Luis de Rojas
Introduction
Administration is one of the tasks an empire does. Our interpretation of the Aztec Empire has changed a lot in the last few years. We now propose a model of the empire very different from the formerly held one, but both models have deep relations with the two main topics of this chapter. Robert Barlow (1949) made a map of the Aztec Empire from the point of view of a territorial domain, with continuous occupation of the land, excepting one province, Xoconochco, which was isolated from the rest. This model was based on the Matrícula de Tributos information in the second part of the Codex Mendoza, due to the very small photographs of the Matrícula he had (Barlow 1949: 7). Barlow interpreted every page of the Matrícula as a tributary province, the first town mentioned was the capital city. According to the Codex Mendoza Spanish texts:
The eighteenth towns drawn and named on the following two pages were ruled by a governor, called petlacalcatl, assigned by the lords of Mexico; and although in each town they placed a calpixqui, like a majordomo, who was in charge of collecting the rents and tributes that the said towns gave to the state of Mexico, all the said majordomos assisted the said petlacalcatl as their governor. (Berdan and Anawalt 1992: 4:44)
The text corresponding to the next province (Cuitlahuac) makes clear the structure and functions of these administrative staff:
The Lords of Mexico, after having conquered the twenty-six towns drawn and named on the following two pages, placed in each town calpixques and, in the principal one, a governor to rule over them, to maintain peace and justice, assure tribute payment, and prevent rebellion. (Berdan and Anawalt 1992: 4:47)
These texts and the next pages representing the towns and tributes listed guided Barlow in drawing the map of the empire with the thirty-eight provinces, named after the first town represented that was the capital or, better said, the home of the governor. Each of the other towns housed a calpixqui who was in charge of the collection of the tributes and the delivering of them to the provincial governor. The other obligations of the calpixqui had almost disappeared from the modern studies, and this is true also for the governor, rarely mentioned.
Barlow was disturbed by the existence of towns subjected by the empire that did not appear in the tributary rolls, mainly in the answers to the 1577 questionnaire of the Relaciones Histórico-Geográficas. He resolved this matter by adding some towns to the provinces and creating a “road to Xoconochco” that made him more comfortable with his map, because this province was no more isolated. It is fair to say that Robert Barlow himself (1949: 1) said that the map was a preliminary and a necessary precursor to the historical study he could never do.
For decades, Barlow’s model was accepted by the scientific community. So, the Aztec Empire had been interested more in the tribute exactions than in the political control of the provinces. The main objective of the administration system was to recover tributes of the provinces and manage them from lesser towns to the provincial capital and then to Tenochtitlan.
Scholars became less and less comfortable with this model and in 1986, things began to change. The process lasted until 1996. There was a nice coincidence since in that year the two books that changed our minds about the Aztec Empire appeared (Carrasco 1996; Berdan et al. 1996). Both are the result of two summer seminars assembled in 1986, one in Mexico City (Carrasco) and the other at Dumbarton Oaks.
The participants in the last conference proposed a different map, based on more sources than Barlow’s map was and with a slightly different interpretation of the tributary rolls:
The most significant innovation of the present study is our focus on the individual towns and places within the Aztec empire. From its inception, our research placed an emphasis on the analysis of political economy, and artistic characteristics of specific towns, and our most interesting findings flow from this emphasis. (Smith and Berdan 1996: 6)
This had a decisive influence in the changing of the Aztec Empire geographical interpretation and forced scholars to change Barlow’s map, based on these fundamental considerations:
First, Barlow constructed his map from the top down. He plotted the towns for each province as listed in the Codex Mendoza and other sources, but did not consider the local spatial context of towns as polities. Our place orientation allowed us to map the empire from the bottom up. We first located the relevant towns (drawing heavily upon Barlow’s work), plotted them plus their associated towns and dependencies on maps of a scale of 1:250.000. From these locations, we reconstructed local city-state territories. Finally, we considered the imperial status of each town to produce our final maps.[…]
Second, we were uncomfortable with Barlow’s practice of extending the boundaries of provinces far beyond the area covered by the Mendoza towns. He added other imperial towns (those known to have been conquered by or subject to the empire, but not included in the Codex Mendoza tribute tally) to the closest province, thereby doubling or even tripling the sizes of those provinces. By maintaining a spatial and analytical separation between the Mendoza towns and other imperial towns, we arrived at significantly different provincial borders and a new understanding of the organization of the empire. (Smith and Berdan 1996: 7)
Because of these considerations, we have now a map in which the most important thing is the relations between different towns, not a continuous territory occupying all the space. And most important: beside the classic tributary provinces, as they appear in the tributary documents, we have a new category, the strategic provinces. These provinces didn’t pay tribute in goods, but performed other commissions (see Berdan 1996; Smith 1996). The Aztec Empire is now different from ancient descriptions, more complex, and most of these changes do affect the administrative system and storage, as we will see later.
The title of Pedro Carrasco’s (1996: 17) book announces another way of thinking. He took the whole structure of the empire, and that means the three parts of the Triple Alliance, and determined the different roles played by different places.
One of Carrasco’s (1991) fundamental ideas, as it was pointed out in former papers, was the intermixture of territories, that is, each of the powers of the Triple Alliance had subjects in the same area, not in separate hinterlands. These are more evident in Maldonado’s (1990) study of Morelos, where there were towns with one calpixqui, towns with two and towns with three calpixque. The most numerous were from Tenochtitlan, but each possible combination is documented.
Carrasco very carefully explained his aims and background. We have important tribute rolls for the Aztec part, not so good for determining the political and territorial structure, and, a different kind of documents for Texcoco and Tlacopan that allows us to identify different levels of territorial units in the Empire, based on their economical obligations and political condition (Carrasco 1996: 70).
If Carrasco’s vision is correct, we must conclude that both maps (Barlow’s and Dumbarton Oaks) refer only to the Tenochca part of the empire and that we must add the map of the Tlacopan dominion and the map of the Texcoco dominion, if we are able to do that someday. And we must intermix the three maps, because all the capitals had subjects in all parts of the territory. There are enough reasons to consider that the proposals of the Dumbarton Oak group and Carrasco may be studied following Hoekstra’s (1990) proposition that in prehispanic Mesoamerica, the dominion was personal, not territorial; that is, there are not subject towns but subject lords. This can explain the presence of calpixque of different parts of the empire at a single town: they did not share the dominion, but each one had subjects of their own. And so on for the lords. Lords who lived in a single altepetl could be vassals of different great lords. We return to this topic later on.
All of us believe that the Aztec Empire collected great quantities of goods, and following Carrasco, Tlacopan, and Texcoco, they must have done so. All these items must be preserved somewhere. Better, there must have existed different stores in different places, along the routes of transportation for different purposes. Also each household must have stored its own goods. The empire’s tribute system required a complex administrative organization for keeping the accounts, watching the caravans, and controlling the tributaries.
Storage
Water, food, and sumptuary goods are three kinds of things that should be preserved. It is not easy to find other goods that do not belong to one of these categories, but each one has different requirements due to its form, weigh, volume, durability, and so on. Further, needs grow in an exponential way when the amount of products to be preserved grows. Storage has not been a successful topic until recent times at the CEMCA symposium in Mexico City in June 2005 (Bortot, Michelet, and Darras 2012). Before this, we have some timely contributions (Rojas 1987a; see also Manzanilla 1988). We have to analyze different places and storage practice in different institutional contexts. We have domestic, community, and state storage. The primary distinction is between places like Tenochtitlan—and perhaps Texcoco—and the rest.
Tenochtitlan
As we have pointed out before (Rojas 2012), the first consideration we must have is that there must have been a lot of things to be stored in Tenochtitlan. We have mentioned tribute, but people have possessions—some of them many—and there was also food to be stored for an enormous population. This food must be consumed day in, day out. We know little of this. Archaeologists have not been able to identify storage units in the excavations, and we have only some mentions in the documentary sources. This topic apparently did not interest the chroniclers. So we have only some quotations about storage. Bernal Díaz del Castillo (1975: 187–88) talks about majordomos and treasurers in charge of the houses of the maize and two houses full of all kind of weapons, all registered in books and accounts. There were two kinds of storerooms for different goods, and probably controlled by different organizations. The following quotation may refer to the storehouses mentioned by Bernal Díaz in his first text:
In Mexico there were troxes, granaries, and houses in which bread was kept, and a great majordomo, with some lesser majordomos, which received it and expended it with the aid of painted books, in which there was such order that it was marvelous. (Cervantes de Salazar 1971: 1:323; English translation by author)
Probably the following quotation is the most cited, because it has more details about the existence and organization of different storehouses for different products:
Petlacalco: there all the food was stored. Dried maize grains thus were kept in wooden grain bins, more than two thousand [measures of] grains of dried maize of twenty years for the city. And in wooden storage bins were dried beans, chia, amaranth, seeds, wrinkled chia, salt jars, coarse salt, baskets of chilis, baskets of medium and large-sized squash seeds. (Sahagún 1950–82: 44)
The steward in charge of the Petlacalco was named in nahuatl petlacalcatl, and this is the name of the first town of the first tributary page of the Codex Mendoza (f. 20r). Because it is not a place name, it has been transformed in Petlacalco by different scholars (see Berdan and Anawalt 1997: 34).
We can identify these storehouses with the trojes reales, mentioned by Diego Durán (1967: 2:130).
It is interesting to check the assortment of products kept and the durability of some of them, because the adequate conservation and the restocking of the products guarded are proof of an efficient organization.
Among the contemporary authors that have written about prehispanic storage in Mesoamerica stands Teresa Rojas (1988), who, in her chapter about cosecha y almacenamiento (“harvest and storage”), follows the study of Hernández Xolocotzi (1949); she identifies two groups of storehouses for the prehispanic state or imperial and domestic. The first one disappeared after the Spanish conquest.
This colonial change may be responsible for our lack of evidence. The first Spaniards at Tenochtitlan, as we have mentioned, did know of the existence of storehouses, but these mentions disappeared very quickly. One must wonder who then kept control of these storehouses and their contents. Also, possibly there were provincial storehouses and their destiny was under Spanish rule.
Again, referring to Tenochtitlan, the presence of great lords, housed in palaces, seems to require several storehouses. In Alonso Axayacatl’s (a cacique of Iztapalapa) last will, dated 1581, storerooms are mentioned for his houses. These were in Iztapalapa, not Tenochtitlan, but this may be an interesting clue:
Item, I have as my property in the said town of Iztapalapa, other houses in the neighborhood of Ucuzacapan, where I have the maize bins in which hundred fanegas of maize and eighty of beans and ten of tomatoes and forty of chili and thirty of chia are stored. (Monjaras-Ruiz 1980: 309; English translation by the author)
In the Tenochtitlan colonial documentary sources analyzed by Calnek trojes in some houses in Tenochtitlan did appear (Calnek 1974: 19, 30) (Figures 13.1 and 13.2).
Figure 13.1 Trojes. After Codex Mendoza, f.37r.
Figure 13.2 Troje. After Florentine Codex, book 7, chapter 8, f. 16r.
Storage outside Tenochtitlan
Outside Tenochtitlan we have far less information, and archaeologists, with a few exceptions like Michael Smith (2012b), have not detected storage units. Most of what we now know comes from ethnographical analogy (Seele and Tyrakowski 1985) and the information given by Batalla (2012) in his study of the representation of storage facilities in the codices. It is important that these representations do coincide closely with ethnographic data, with the added value that they were painted from the prehispanic to the Late Colonial periods. Batalla includes important information like the presence of the Nahuatl words cuezcomatl and quauhcuezcomatl in the sixteenth-century Nahuatl-Spanish dictionary of Friar Alonso de Molina. The first term in modern Spanish is cuescomate and the second, the troje (Figure 13.3).
As mentioned above, the societal context of storage is important to understand. In the domestic level, storage facilities were inside the house in baskets or pots, often shelled or unshelled in the case of maize. None of these have been preserved. Outside the house, we have specific buildings, like the cuescomate, for grain storage, particularly shelled maize, and in the trojes, the unshelled maize. We have no evidence of communal storage, but if it existed, probably the same types of constructions, but greater ones, were used. For local rulers and for state storehouses the same consideration could be made, as we will soon see. Cuescomates were made of mud and straw:
Its form may resemble a domed or convex recipient with its upper part covered with an onion-form roof, or a pediment made of tejamanil, zacate, or channels as protection against bad weather. The cuescomate may have a spherical or pear-like form, but also, as proved by the more ancient constructions observed, it may be elongated like a sleeve. The spherical recipient may rest directly on the floor—that was observed fewer times and is usual in closed ranches—or over a range of stones. This is done because it protects the food from rodents and water that splashes up from soil. Also it provides ventilation, (Seele and Tyrakowski 1985: 9; English translation by the author)
Comparing this description with figures 1–10 in Batalla (2012) the close connection of ancient and modern cuescomates becomes clear.
The trojes named quauhcezcomatl by Molina (1977: 115) seem to have been larger, as it is appreciated in codices like the Florentino and Pièce d’un procés: plan et titre d’une proprietésise a Hueocolco (Batalla 2012: fig. 14). The trojes are represented several times in the Matrículade Tributos and the Codex Mendoza, but we cannot think there are realistic representations of their size. Nevertheless, the form is close to its real form as is seen when reading the classical description by Clavijero:
They had trojes to keep grain. Their trojes were square and mainly made of wood. They used the oyametl, a very high tree that grows very straight, with a round trunk, subtle and smooth bark, few and slender branches. It produces a very hard wood difficult to break or to get moth-eaten. They formed these trojes over a base of pine beams, putting then the oyametl trunks in square form until they reached the high they desired. They made notches near the ends of the trunks in order to fit one into the other so close that light can’t go in. When they reached the height desired, they cover the troje with pine beams and over it a roof to keep the rain off the grain.
These trojes had the door in the upper side and a little window in the lower one. They were so huge that they contained five or six thousand hanegas of maize. (Clavijero 1982 [1780]: bk. 7, 29, 231; English translation by the author)
Figure 13.3 Cuescomate as a place glyph. After the Historia Tolteca Chichimeca (Ms 54–58, pp. 1–2).
Seele and Tyrakowski (1985: 22) show another construction, named zencal (Nahuatl cencalli), that they say existed in the prehispanic and colonial periods, but it resembled Spanish constructions and seems to be far from the indigenous traditions, as the authors themselves admit (1985: 27). Batalla (2012) refers to a cincolotli, related to the cencalli by Noguez (1995: 57), but it seems to be synonymous with troje more than a different type of construction.
We have not found in the documentary sources the subterranean granary that appears in Molina’s (1977: 124) dictionary, but Batalla (2012: fig. 19) presents two examples of subterranean constructions in two different codices, one of them the Florentino Codex, that may represent these subterranean storerooms. Its existence deserves more research.
The State System
The key question remains: Did the Aztec Empire have a large-scale, central storage system? To date we have no archaeological information (Smith 2012a). Teresa Rojas (1988: 104) argues for a “state reserves and distribution” system in Cuscomatepec (Veracruz), using a quotation from Mota y Escobar (1945 [1609]).
Although there is not enough information about storehouses and storage systems, there is overwhelming information about goods going in and out throughout Mesoamerica, which must have been stored many times in many places. There must have been different organizations at domestic, community, town, and city-state levels. The domestic level is the lowest step of the tribute system. There logically must have been a nobles’ storage system, with a domestic part for the daily needs, another part as a reserve, and a third part for keeping the private luxury items. Some of these goods lower-ranking lords must pay as tribute to higher-ranking lords.
Evidence of the quantities of tribute that sometimes appeared in the tributary codices is clear, but the difficulties in the identification of products or the evaluation of the measures used are also clear (Rojas 1997; Rojas and Batalla 2008). The size of the trojes, the delivery of loads or units, and the frequency of payment are the key questions. To accept one or another quantity is important for the storage needs. High-valued items need more storehouses, and this means storage in the point of origin, storage on the different intermediate steps, and storage at the final destination. Time is important; we must consider how much time one thing is kept in each place. The storage needs are close to the movement schedule; the more things are stored in each place, the more storehouses are needed, and if they are moving, the same places are occupied successively. But different things may need different kinds of storehouses, and we know only the trojes seem to be able to contain grains.
Further, the storehouses need specific designs, proper construction, using adequate materials, and constant maintenance. This may be included in the necessary management and planning. The Mexicas knew a lot about building, maintenance, and management, as the Chapultepec aqueduct shows; there are two streams, and one flowed while the other was cleaned and repaired.
Important to this effort to understand storage is first to understand what items were moved and stored (Rojas and Batalla 2008). The amounts of items just in the great tributary codices are amazingly high. However, there are more than tribute items. There are vegetables going to Tenochtitlan from the chinampas zone via the pochteca or merchant traffic, and Tlatelolco’s great market, as well as the food and sumptuary good supply of other great cities like Texcoco.
Regarding grain, the troje’s capacity is a key point; the low estimate for the capacity of a troje is two thousand fanegas, and the high estimate is five thousand (see Rojas and Batalla 2008: 200–201). Here we use the two extremes to show the range and not prejudice the results, but we must also accept the size of the fanega, which seems to be more difficult than expected to calculate. For instance, Katz (1966) used a measure of 144 liters, and that seems to us very improbable. It seems reasonable to use the fanega de Castilla in New Spain in the sixteenth century, a time when most of our sources were written. It has fifty-five liters of capacity, and fifty-five liters of maize weigh around forty-six kilograms. It is important to say that a fanega has four arrobas, or two loads, two arrobas each. So, the standard load carried by porters was twenty-three kilograms. It is important to count the number of carriers needed to move all the tribute. Each troje had between 110,000 liters (Sahagún’s figure) and 275,000 liters (the higher figure from the Codex Mendoza). The following step determines how many trojes are paid each year, which, too, is difficult to estimate. The Spanish text of the Codex Mendoza does not fit with the glosses and they are far from the pictographs. The text multiplies the trojes that contained different kinds of grains. We consider that the correct procedure is to read the pictographs, and so we have forty-two trojes of different grains. The Información de 1554 did not count trojes. Instead it says “4,100 fanegas” We must also decide the frequency of the tribute. We think that the grain tribute was paid one time during the year. Of course, if one accepts more payments or more trojes, the storage problems will grow in the same proportion. With the data we accept, the annual grain tribute was between 4,620,000 liters and 11,550,000 liters. Putting it all together in a storehouse with a base of one hundred square meters, we will need to calculate 46.2 cubic meters in the lower estimate and 115.5 cubic meters in the higher one. If we change the base of the unit, we must change the height. Probably there never were stores like this one, and different products may have been stored in different places, with different forms and sizes. For instance, the Conasupo cellars have a capacity between one hundred and five thousand tons (Seele and Tyrakowski 1985: 42).
If we accept the data from Tenochtitlan, there were different storehouses of diverse sizes in which they kept different items, including grains. The food supply of the city (Rojas 2001) holds a clue to consider the storage necessities. Again, estimates are difficult to calculate. The annual requirements per person are between 160 and 256 kilograms of maize—used as equivalency (Ortiz de Montellano 1993: 101)—which seems low, and the size of the population should give us a rough figure. For instance, with a population of two hundred thousand and the annual requirement of two hundred kilograms of maize per person, we need forty million kilograms of food per year. With the highest figures, the tribute was 11,550,000 kilograms. That is a difference of more than twenty-eight million kilograms. It is interesting to mention that Texcoco did have storehouses in Tenochtitlan if we believe Ixtlilxochitl (1975–77: 2:151). We also have figures from Texcoco: 31,600 fanegas of maize, 243 loads of cacao, 8,000 cocks, 5,000 fanegas of wide and slender chili and seeds, and 2,000 measures of salt were consumed in a year. There were also 574,010 measures of cloth, most of the last of very fine quality. All of these were kept in the storehouses the Texcocan tlahtoani had in Texcoco and in Mexico (Ixtlilxochitl 1975–77: 2:150–51).
Was it possible that Tlacopan and Tenochtitlan had storehouses in the other capitals, such as Texcoco? The answer is probably yes. Moreover, the three must have had storehouses in other places, in order to manage their tribute, but as usual, most of the data come from Tenochtitlan.
Outside Tenochtitlan, because of the enormous differences in city sizes, the storehouses for local consumption may have been smaller. For the towns placed along trade routes, one must assume that there was a continuous replacement of goods.
The analysis of the storage so far has been focused on grains, avoiding the inclusion of other items. Even the interpretation of data has the same bias. When Smith (2012a) concludes that there was not a grain storage system in the Aztec Empire, he means that the state did not attend to the food necessities of people. We agree with his final chapter heading: “The lack of an Aztec food policy,” but we think that this does not imply the nonexistence of storehouses, particularly for grains. The accumulation, the sumptuary expenses, and the maintenance of the royal houses may be reasons enough to store things. Batalla (2012: figs. 16–18) presents the representation of buildings named petlacalco in the codices, with the indication of the things that were stored in them. Different kinds of storehouses must have been were required to keep different goods. There are delicate items that could not be piled up, and so they would have needed more surface or rigid containers or a sort of shelf. We have no evidence of these storerooms, but archaeologists have not spent much time looking for them.
Even though there is scarce evidence of storehouses, there is a lot of evidence of goods in search of storerooms in every step of their life, sometimes just for a while, sometimes for years. These storehouses must have been in different places, and the ones that were along the trade routes must have renewed their contents quickly. Maybe it is better to talk about outposts where the travelers could deposit their merchandise during the night and continue on their way the next morning rather than managed storehouses. The logistics of the pochteca caravans need further research.
The Administration
Another indirect evidence of the existence of Aztec storehouses is the presence in the documentary sources of people in charge of the collection and administration of tribute. For the Mexica case, there are two different types of evidence: administrators residing in the royal palace at Tenochtitlan and those who were in towns and provinces subject to the Aztec Empire.
Sahagún (1975: bk. 8, chap. 14, 467) refers, as discussed above, to the existence of a dependency specialized in economic affairs, named petlacalco, where tribute, among other obligations, was stored. This dependency was presided over by the petlacalcatl, also named huey calpixqui. The entity had majordomos, treasurers, counters, tax collectors, and the rest of officials in charge of the treasure (Cervantes de Salazar 1971: 1: 319; Hernández 1946: 97; Vetancurt 1971: 50). Durán (1967: 2: 323) mentions majordomos, factors, treasurers, and mandoncillos of the wards. There was another dependency named calpixcacalli where Durán (1967: 2: 116) places the calpixque who received their instructions. Sahagún (1975: bk. 8, chap. 14, 467) describes it as follows:
Another room was named calpixcacalli, and its other name was texancalli. In this place all the majordomos of the lord got together, bringing each one the count of the tribute in his charge, to justify the accounts when the lord asked for them, and so, each day every majordomo prepared the tribute in his charge.
There are different texts showing the differences in categories of these administrators. An example of this is appears in the History of Friar Diego Durán (1967: chap. 41, 313). There were officers for everything, even sweeping the city. There were officers in charge of one hundred, fifty, forty, or twenty houses. The officers of lesser houses were chosen by those who were in charge of one hundred houses and were their subjects.
Another category of official is the calpixqui. It has been understood as “tax collector” Armillas (1987: 21–23) argued over these political duties using the chronicles of the Spanish Conquest in which different calpixque appeared, but none is related to tribute. The existence of calpixque in the provinces of the empire is strongly linked to the Codex Mendoza in which they and their duties are specified, as mentioned in the beginning of this chapter. Their functions were collecting tribute and also governing and administering justice. In the Relaciones Histórico-Geográficas (1577 questionnaire) local lords, ruling their dominions, and paying tribute to the governors and majordomos that Moctezuma had placed in the different provinces, appear (Rojas 1991). In this regard, the Codex Mendoza evidence has been misinterpreted or ignored. The calpixque have been successful while the governors have almost disappeared. Even more, the Codex Mendoza states that the towns depicted in the tributary pages are those that have a calpixqui, not the tributary towns. This helps explain why conquered towns present in other sources do not appear in the tributary roll, and important towns listed are not the local Aztec capitals. This information may be useful to determine the extent of the administration of the empire and its relation to local governing bodies (see Hicks 1984, 1992).
To understand all this, it is necessary to ask what the Aztec Empire really was. The new views that appeared in 1996 owe a debt to a lot of previous research (see Rojas [Mss.] for an overall panorama). Smith (1986) pointed out a hypothesis that had been more and more accepted:
It is my contention that the primary force binding the empire together was the common interest of the Postclassic Mesoamerican nobility. Provincial ruling dynasties cooperated with the rulers of the Triple Alliance states in the economic exploitation of their local commoners. Social stratification may thus be singled out as one of the most important factors in the integration of much of the empire. (70)
The extension of familial ties among the Mesoamerican ruling class is well known. The existence of polygamy allowed the rulers to have personal ties with many other families. It was customary to exchange women, so the ties functioned in two directions. This does not mean a lack of stratification; the ruling class in Mesoamerica was strongly stratified (see Carrasco 1974, 1984). The marriage system tended to form a web of power, very similar to the organization of the administrative officers. Each lord had subjects (who were close relatives) and were subjects (of a relative lord), except the Tenochtitlan lord in the later empire. But the status quo does not last forever. When the Triple Alliance defeated Azcapotzalco, the dynasty shifted from Azcapotzalco to Tenochtitlan (Santamarina 1998, 2006), and the local dynasties were realigned. It was possible to gain more status and higher rank via warfare, alliance, or marriage. The structure was similar in every level. The intermediate lords had spouses of superior status and of inferior status, and they gave spouses to their superior lords and to their subject lords in an intricate network of familiar relations with quite strict rules (see Carrasco 1974).
This model resembles the organization described by Durán. Each lord was promoted by his superior lord and, at his time, he named his subordinate lords, and so on, down to the lowest-rank lord. Every lord ruled over his subjects and obeyed his superiors, forming a pyramid. Colonial information found in the Matrícula de Huexotzinco (1974) and the Padrones de Morelos (see Hinz, Hartau, and Heimann-Koene 1983; Cline 1993; Díaz-Cadena 1978) confirms this model, with the presence of low-ranking administrators who depended on lords and gave them account of the tribute and duties of the people they had in charge. This forms a hierarchical ladder that works very efficiently and gives support to Smith’s ideas. This chain of lords resembles much of the new map of the Aztec Empire, listing lords instead of towns.
The acceptation of this model has some implications for our understanding of storage practices. In different places it is said that the governors and calpixque who ruled the towns were from Mexico (Codex Mendoza, f. 22v, 41v); this is also said for the officers who appear in the controversial pages 17v–18r of the same Codex (f. 18r). It seems it was customary to put new administrators and lords in charge when the empire made a conquest: Mexica in the tenochca case, and we must suppose Acolhua in the Texcoco case, and Tepaneca when Tlacopan was the conqueror. Friar Diego Duran and Friar Bernardino de Sahagún agreed about this topic:
They told them the tribute that they must deliver to Mexico and they accepted it and promised to accomplish it. And leaving them collectors and a Mexican lord to rule them, they left Tlachquiauhco. (Durán 1967: 2:481–482; English translation by the author).
After the province was pacified then the rural lords imposed tributes to those who had been conquered in order to give them annually to the lord who had conquered them; and the tribute consisted of what were produced locally and then they elected governors and officers to rule that province, not from people born there but of the conquerors. (Sahagún 1975: 470; English translation by the author)
One interesting question arises here: Were these Mexican governors imposed by the conquerors really foreigners or did they have some relationship to the local community? We have seen that it was not difficult to be Mexican and, at the same time, be a member of local dynasties as a result of marriage alliances. The conqueror may have appointed Mexican governors who were members of the local ruling family. This could help reinforce the ties with the empire without losing the local links. This coincides with Smith’s idea, because the local lords also formed part of the administrative structure of the empire. This analysis cannot ignore the role of the factions; the local and Mexica appointed ruler could also have been a conqueror, allied to the empire with the aim of becoming a local lord instead of being a relative. There were wars between towns, but also between different branches of a ruling family, and this helps to explain the realignments that happened after a conquest (e.g., Rojas 1994). In fact, one way to advance inside the empire was to collaborate in new conquests and receive reward for it. The power shifted from local roots to central decision makers. The most important was the ruler; the less important ruled the towns. It is possible that the real significant items could be the lords, not the towns in the way Hoekstra (1990) had shown us.
An alternative conception of polity definition, called “capital-centric” by Berman (2005), emphasizes people, not territory. As I suggest above, an altepetl consisted of all of the people subject to a king, wherever they happened to live. In many cases the members of an altepetl lived in a single continuous territory. In some cases, however, the subjects of neighboring kings lived interspersed with one another to such a degree that it is impossible to draw discrete territorial boundaries between the polities. The best known example concerns the altepetl of Tepechpan, Acolman and Teotihuacan in the Teotihuacan Valley (this is the Aztec altepetl of Teotihuacan, not the Classic period metropolis). This situation was first pointed out by Charles Gibson (1964, 44–47); see Figure 3.6. It is simply not possible to draw any kind of rational territorial boundaries around the subject villages of these three altepetl. This distribution of settlements, however, makes perfect sense from the alternative people-focused perspective of Pre-hispanic central Mexico was transformed under Spanish rule into the European territory-based viewpoint of the organization of native communities. (Smith 2008: 91)
We think that this idea helps explain also the intermixture of territories (Carrasco 1996); there were lords with subjects in a same area. It also explains the situation of altepetl with more than one lord. In some cases, there was a powerful lord, with some subject lords, and in other cases, there were different lords, with their respective subjects, and all of them were relatives. A good example of this situation for the colonial period is in Martinez’s (1984) study of Tepeaca. The shift from the personal to the territorial model of domain occurred in Europe after the Conquest of Mexico, which means that the Spanish conquerors understood very well the Mexican way of organization. The study of colonial lords in Mexico (Rojas 2010) supports the personal model and gives us a clue over the double office of rulers as local lords and imperial officers: most of colonial tlahtoque in the sixteenth century were governors in the Spanish organization. Their relatives, who were tlahtoque or tecuthli, occupied the main posts in the cabildo—alcaldes and regidores—and some of them acted as judges in other towns, in service of the royal audience (Rojas 2010: chap. 5).
Storage and administration formed part of the Aztec Empire structures and we must study them in close connection with the empire interpretation. Changing our point of view has added value to new documents that allow us a new interpretation. There is new evidence and new questions in search of answers.
Some ideas come from South America and the Inka Empire (see Covey, Quave, and Covey, Chapter 7, this volume). We hope that Teotihuacan will supply archaeological evidence lacking for Tenochtitlan (see Manzanilla, Chapter 9, this volume), and concepts like the palace as an administrative center, as in the Near East, need to be investigated.
References
Armillas, Pedro. 1987. La realidad del Imperio Azteca. In La aventura intelectual de Pedro Armillas, ed. José Luis de Rojas, 13–33. Zamora: El Colegio de Michoacan.
Barlow, Robert H. 1949. The Extent of the Empire of the Culhua Mexica. Berkeley: University of California Press.
Batalla, Juan José. 2012. Análisis de la representación de depósitos de almacenamiento en los códices. In Almacenamiento prehispánico del Norte de México al Altiplano Central, ed. Séverine Bortot, Dominique Michelet, and Véronique Darras, 187–202. Paris, Mexico City, and San Luis Potosí: Université de Paris I Panthéon-Sorbonne, Universidad Autónoma de San Luis Potosí, Centro de Estudios Mexicanos y Centroamericanos.
Berdan, Frances F. 1996. The Tributary Provinces. In Aztec Imperial Strategies, ed. Frances F. Berdan, Richard E. Blanton, Elizabeth Hill Boone, Mary G. Hodge, Michael E. Smith, and Emily Umberger, 115–135. Washington, DC: Dumbarton Oaks.
Berdan, Frances F., and Patricia Anawalt. 1992. The Codex Mendoza. 4 vols. Berkeley: University of California Press.
Berdan, Frances F., and Patricia Anawalt. 1997. The Essential Codex Mendoza. Berkeley: University of California Press.
Berdan, Frances F., Richard E. Blanton, Elizabeth Hill Boone, Mary G. Hodge, Michael E. Smith, and Emily Umberger. 1996. Aztec Imperial Strategies. Washington, DC: Dumbarton Oaks.
Berman, Merrick Lex. 2005. Boundaries or Networks in Historical GIS: Concepts of Measuring Space and Administrative Geography in Chinese History. Historical Geography 33: 118–133.
Bortot, Séverine, Dominique Michelet, and Véronique Darras, eds. 2012. El almacenamiento Prehispánico del Norte de México al Altiplano Central. Paris, Mexico City, and San Luis Potosi: Université de Paris I, Universidad Autónoma de San Luis Potosí and Centro de Estudios Mesoamericano y Centroamericanos.
Calnek, Edward. 1974. Conjunto urbano y modelo residencial en Tenochtitlan. In Ensayos sobre el desarrollo urbano de México, ed. Edward E. Calnek et al., 11–59. Mexico City: SEP Setentas.
Carrasco, Pedro. 1974. Sucesión y alianzas matrimoniales en la dinastía teotihuacana. Estudios de Cultura Náhuatl 11: 235–241.
Carrasco, Pedro. 1984. Royal Marriages in Ancient Mexico. In Explorations in Ethnohistory: Indians of Central Mexico in the Sixteenth Century, ed. Herbert H. Harvey and Hans J. Prem, 41–81. Albuquerque: University of New Mexico Press.
Carrasco, Pedro. 1991. The Territorial Structure of the Aztec Empire. In Land and Politics in the Valley of Mexico: A Two Thousand Year Perspective, ed. Herbert H. Harvey, 93–113. Albuquerque: University of New Mexico Press.
Carrasco, Pedro. 1996. Estructura político-territorial del Imperio tenochca. La Triple Alianza de Tenochtitlan, Tetzcoco y Tlacopan. Mexico City: Fondo de Cultura Económica and El Colegio de México.
Cervantes de Salazar, Francisco. 1971. Crónica de la Nueva España. 2 vols. Madrid: Editorial Atlas.
Clavijero, Francisco Javier. 1982 [1780]. Historia antigua de México. Mexico City: Editorial Porrua.
Cline, Sarah L. 1993. The Book of Tributes: Early Sixteenth-century Nahuatl Censuses from Morelos. Los Angeles: University of California Press.
Díaz-Cadena, Ismael. 1978. Libro de los tributos del Marquesado del Valle. Mexico City: Instituto Nacional de Antropología e Historia.
Diaz del Castillo, Bernal. 1975. Historia verdadera de la conquista de la Nueva España. Madrid: Espasa Calpe.
Durán, Diego. 1967. Historia de las indias de Nueva España e islas de la Tierra Firme. 2 vols. Mexico City: Editorial Porrua.
Gibson, Charles. 1964. The Aztecs under the Spanish Rule: A History of the Indians of the Valley of Mexico 1519–1810. Stanford, CA: Stanford University Press.
Hernández, Francisco. 1946. Antigüedades de la Nueva España. Mexico City: Editorial Pedro Robredo.
Hernández Xolocotzi, Efraín. 1949. Maize Granaries in Mexico. Botanical Museum Leaflets 1 (7): 153–192.
Hicks, Fréderic. 1984. La posición de Temazcalapa en la Triple Alianza. Estudios de Cultura Náhuatl 17: 235–260.
Hicks, Fréderic. 1992. Subject States and Tribute Provinces: The Aztec Empire in the Northern Valley of Mexico. Ancient Mesoamerica 3: 1–10.
Hinz, Eike, Claudine Hartau, and Marie Luise Heiman-Koene. 1983. Aztekischer Zensus zur Indianischer Wirtschaft un Gessellchaft in Marquesadio um 1540. 2 vols. Hannover: Verlag für Ethnologie.
Hoekstra, Rik. 1990. A Different Way of Thinking: Contrasting Spanish and Indian Social and Economic Views in Central Mexico (1550–1600). In The Indian Community of Central Mexico, ed. A. Ouweneel and S. Miller, 60–86. Amsterdam: CEDLA.
Ixtlilxochitl, Fernando de Alva. 1975–77. Obras Históricas. Mexico City: Universidad Nacional Autónoma de México.
Katz, Friedrich. 1966. Situación social y económica de los aztecas en los siglos XVy XVI. Mexico City: Universidad Nacional Autónoma de México.
Maldonado, Druzo. 1990. Cuauhnahuac y Huaxtepec (Tlalhuicas y Xochimilcas en el Morelos Prehispánico). Cuernavaca: Universidad Nacional Autónoma de México.
Manzanilla, Linda. 1988. Los contextos de almacenamiento en los sitios arqueologicos y su estudio. Anales de Antropología 25: 71–87.
Martínez, Hildeberto. 1984. Tepeaca en el siglo XVI. Tenencia de la tierra y organización de un señorío. Mexico City: CIESAS.
Matricuía de Huexotzinco. 1974. Mss. Mexicanus 387 derBibliothèqueNationale de Paris. Facsimilar edition. Ed. Hanns J. Prem and Ursula Dyckerhoff. Graz: Academische Druk und Verlaganstalt.
Molina, Alonso de. 1977. Vocabulario en lengua castellana y mexicana y mexicana y castellana. Mexico City: Editorial Porrúa.
Monjarás-Ruiz, Jesús. 1980. Testamento y fundación de una capellanía por don Alonso de Axayacatl de Iztapalapa. Tlalocan 8: 289–322.
Mota y Escobar, Alonso. 1945. [1609]. Memoriales del obispo. Mexico City: Instituto Nacional de Antropología e Historia.
Noguez, Xavier. 1995. El códice de Huexotzinco. In The Huexotzinco Codex/ El códice de Huexotzinco: A Facsimile of the 1531 Huexotzinco Codex in the Harkness Collection, ed. John R. Hébert, Luis Lobao, Xavier Noguez, Sylvia Rodgers Albro, and Thomas C. Albro, 17–62. Mexico City: Library of the Congress, Coca-Cola Export Corporation and Ediciones Multiarte.
Ortiz de Montellano, Bernardo. 1993. Medicina, salud y nutrición Aztecas. Mexico City: Siglo XXI.
Rojas, José de Luis. 1987a. El control del granero del Imperio y la consolidacion del estado mexica. In Almacenamiento de productos agropecuarios en México, ed. G. Mummert, 29–38. Zamora: El Colegio de Michoacan y ANDSA.
Rojas, José de. Luis 1987b. La aventura intelectual de Pedro Armillas. Zamora: El Colegio de Michoacan.
Rojas, José de Luis. 1991. La organización del imperio mexica. Revista Española de Antropología Americana 21: 145–169.
Rojas, José de Luis. 1994. After the Conquest. Cuauhtinchan and the Mexica province of Tepeacac. In Economies and Polities in the Aztec Realm, ed. Mary G. Hodge and Michael E. Smith, 405–431. Albany: State University of New York Press.
Rojas, José de Luis. 1997. Informacion de 1554 sobre los tributos que los indiospagaban a Moctezuma. Mexico City: Centro de Investigaciones y Estudios Superiores en Antropologia Social.
Rojas, José de Luis. 2001. El abastecimiento de Tenochtitlan: Un modelo probablemente poco modélico. In Reconstruyendo la ciudad maya: El urbanismo en las sociedades antiguas, ed. A. Ciudad, M. J. Iglesias, and C. Martínez, 391–501. Madrid: Sociedad Espanola de Estudios Mayas.
Rojas, José Luis. 2010. Cambiar para queyo no cambie: La nobleza indígena en la Nueva Espaha. Bueños Aires: Editorial SB.
Rojas, José Luis. 2012. El almacenamiento en el Imperio Mexica: una necesidad evidente en busca de evidencias. In El almacenamiento prehispánico del Norte de México al Altiplano Central, ed. S. Bortot, D. Michelet, and V. Darras, 173–178. Paris, Mexico City, and San Luis Potosí: Université de Paris I, Universidad Autónoma de San Luis Potosí y Centro de Estudios Mesoamericanos y Centroamericanos.
Rojas, José Luis. Mss. El Imperio Azteca: historia de una idea.
Rojas, José de Luis, and Juan José Batalla. 2008. Los numeros ocultos del Códice Mendoza y la Matrícula de Tributos. Revista Española de Antropología Americana 38 (2): 199–206.
Rojas, Teresa. 1988. Lassiembrasdeayer. La agricultura indígena del siglo XVI. Mexico City: Secretaría de Educación Pública y Centro de Investigaciones y Estudios Superiores en Antropologia Social.
Sahagún, Fray Bernardino de. 1950–82. The Florentine Codex. Trans. Arthur Anderson and Charles Dibble, 12 vols. Salt Lake City: University of Utah Press.
Sahagún, Fray Bernardino de. 1975. Historia General de las cosasdela Nueña Espaha. Mexico City: Editorial Porrúa.
Santamarina, Carlos. 1998. La muerte de Chimalpopoca: Evidencias a favor de la tesis golpista. Estudios de Cultura Náhuatl 28: 277–316.
Santamarina, Carlos. 2006. El sistema de dominación azteca: El Imperio Tepaneca. Madrid: Fundación Universitaria Espanola.
Seele, Enno, and Konrad Tyrakowski. 1985. Cuescomate y zencal en la región Puebla-Tlaxcala/México. Puebla: Fundacion Alemana para la Investigación Científica.
Smith, Michael E. 1986. The Role of Social Stratification in the Aztec Empire: A View from the Provinces. American Anthropology 88 (1): 70–91.
Smith, Michael E., 1996. The Strategic Provinces. In Aztec Imperial Strategies, ed. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M. E. Smith, and E. Umberger, 137–150. Washington, DC: Dumbarton Oaks.
Smith, Michael E. 2008. Aztec City-State Capitals. Gainesville: University Press of Florida.
Smith, Michael E. 2012a. El almacenamiento en la economía azteca: Una perspectiva comparativa. In Almacenamiento prehispánico del Norte de México al Altiplano Central, ed. Séverine Bortot, Dominique Michelet, and Véronique Darras, 203–220. Paris, Mexico City, and San Luis Potosí: Université de Paris I Panthéon-Sorbonne, Universidad Autonoma de San Luis Potosí, Centro de Estudios Mexicanos y Centroamericanos.
Smith, Michael E. 2012b. Graneros y almacenamiento de maíz en Morelos Posclasico. In Almacenamiento prehispánico del Norte de México al Altiplano Central, ed. Séverine Bortot, Dominique Michelet, and Véronique Darras, 159–172. Paris, Mexico City, and San Luis Potosi: Université de Paris I Panthéon-Sorbonne, Universidad Autónoma de San Luis Potosí, Centro de Estudios Mexicanos y Centroamericanos.
Smith, Michael E., and Frances F. Berdan. 1996. Introduction. In Aztec Imperial Strategies, ed. F. Berdan, R. E. Blanton, E. H. Boone, M. G. Hodge, M. E. Smith, and E. Umberger, 1–10. Washington, DC: Dumbarton Oaks.
Vetancurt, Agustin de. 1971. Teatro mexicano. Mexico City: Editorial Porrúa.