The ninth-century Classic Maya Collapse has sometimes been viewed as a process that was limited to the southern Maya Lowlands. In contrast, the cities and polities of the northern lowlands were seen as having survived the Collapse and, in some regions, to have forged ahead in a new period of prosperity and growth during the Terminal Classic (a.d. 800–1000) and Early Postclassic (a.d. 1000–1200) periods.

By “Collapse” we mean the cessation of elite activities and rapid depopulation of a site or region (e.g., Adams 1973:22). The term does not necessarily imply abrupt abandonment. The chronology and intensity of the Collapse, and the degree to which different causal factors were involved, varied from region to region (Culbert 1973; Webster 2002). Many sites around the peninsula show continuity of settlement through these years. For example, Chunchucmil, Dzibilchaltun, Izamal, Yaxuna, Ek Balam, Chiquila, and Coba, while suffering drastic declines at the end of the Terminal Classic period, nevertheless continued to be occupied during the Postclassic period. Certain regions, such as the Caribbean coast, northern Belize, and the Peten Lakes region, show less evidence of a Terminal Classic collapse. These areas may have been spared because of their more favorable locations—coastal, lacustrine, or riverine—and the commercial, agricultural, or water resources these locations offered. Sites in these areas nevertheless saw marked shifts in ceramic and architectural styles, settlement patterns, and political organization, suggesting that they, too, were affected by and had to adjust to the changes that followed the end of the Classic period.

The reconstruction of the northern past as essentially different from that of the southern lowlands, based primarily on the research conducted by the Carnegie Institution of Washington during the first half of the twentieth century, began to change in the 1970s and early 1980s. The first challenges to the older views emerged as archaeologists began questioning the traditional chronology constructed by Carnegie archaeologists, which proposed a sequential and nonoverlapping series of cultural phases defined by both architectural and ceramic complexes: Puuc/Cehpech being replaced by Itza/Sotuta, which in turn gave way to Mayapan/Tases.

New data from the field and re-examinations of the older chronological data gave rise to the notion that there was partial or total overlap between the Puuc/Cehpech and Itza/Sotuta periods, and that this overlap might vary from region to region (Andrews V 1979; Andrews V and Sabloff 1986; Ball 1979; Lincoln 1983, 1986; Sabloff 1990). Additional data from excavations conducted in the 1980s and 1990s have provided new evidence for overlap of the Cehpech and Sotuta ceramic spheres (Anderson 1998; Andrews et al. 1988; Bey, Hanson, and Ringle 1997; Bey, Bond, Ringle, Hansen, Houck, and Peraza 1998; Carrasco Vargas 1996; Chung 2000; Kowalski 1994; Kowalski et al. 1996; Robles Castellanos 1987, 1988; Suhler et al. 1998). Restudies of ceramic collections and their contexts also support overlap models (Chung 2000; Ochoa 1999). Although the extent of overlap of these material complexes in time and space varies from region to region in complex ways that are not fully known, we know that many types of Sotuta-sphere ceramics appear in mixed deposits with Cehpech materials from at least a.d. 850–900 onward, and that diagnostic types from both spheres were produced well into the eleventh century.

With the realignment of the ceramic spheres, and partially as a result of it, came a re-examination of the history of Uxmal and Chichen Itza. Architectural, epigraphic, iconographic, and ceramic data suggest that elite buildings and carved monuments at Uxmal ceased in the early tenth century (Kowalski 1994; Kowalski et al. 1996). Some archaeologists believe that there is evidence of occupation and modest residential construction here and at Kabah (Carrasco Vargas 1996) for another half- or even full century. Although these and other Puuc sites were not all rapidly abandoned, they were no longer powerful political and economic centers by the close of the tenth century (G. Andrews 1994; Alfredo Barrera Rubio and José Huchím Herrera, personal communication 1998).

New research and reassessments of the data from Chichen Itza have led to a dramatic reconstruction of its history, as well. The re-dating of prominent Sotuta ceramic types to the ninth century brings the ceramic evidence in line with the epigraphic and radiometric data from the site, suggesting that the Itza city emerged as a major regional capital in the first half of the ninth century and held sway over a large part of northern Yucatan until the end of the tenth century and perhaps into the first half of the eleventh century.

Most of the Long Count dates from Chichen Itza and the nearby sites of Yula and Halakal fall between a.d. 832 and 897. One controversial inscribed date, from the High Priest's Grave, has been variously read as a.d. 842, 894, and 998. A recent analysis by Daniel Graña-Behrens, Christian Prager, and Elisabeth Wagner (1999) favors the latest reading. But even if the epigraphic record at Chichen Itza ends before a.d. 900, this does not necessarily date the end of elite activity. As Ruth Krochock and others have suggested, the change may reflect a shift to a public narrative iconographic program directed to an increasingly cosmopolitan and multiethnic population (Krochock 1998; Krochock and Freidel 1994).

The midpoints of relevant Sotuta-related calibrated radiocarbon dates from Chichen Itza range from a.d. 883 to 891, and four calibrated dates from Balankanché range from a.d. 968 to 1009 (uncorrected dates listed in Andrews IV and Andrews V 1980, Table 4; see Ringle, Bey, and Peraza 1991; Ringle, Gallareta Negrón, and Bey 1998:242 for calibrated dates). Calibrated midpoints of Sotuta-related dates at Isla Cerritos, the main port for Chichen Itza, range from a.d. 740 to 1075 (Andrews et al. 1988).

In sum, these data favor a substantial overlap of Uxmal and Chichen Itza. The former site emerged as a prominent regional center in the late eighth century and survived until the middle or late tenth century, and Chichen Itza emerged in the first half of the ninth century and survived into or beyond the first half of the eleventh century.

For all practical purposes, then, the period of monumental construction and other elite activity at Chichen Itza has now been redated to the Terminal Classic period (Andrews V and Sabloff 1986:449–456; Cobos Palma 1997, 1998, 1999; Ringle 1990; Ringle, Bey, and Peraza 1991; Ringle, Gallareta Negrón, and Bey 1998; Schele and Mathews 1998:198–200). Although more precise dates will be needed to determine how late the Itza remained a strong political and economic presence in the northern lowlands, there should be little doubt that their period of supremacy was over before a.d. 1100, and possibly before a.d. 1050. The cessation of monumental construction at Chichen Itza, which we date to about a.d. 1000, may not necessarily indicate the demise of the city's political power, for it is likely that the Itza continued to play a major role in the economy and politics of the northern lowlands long after monumental construction slowed down or came to a halt, much like downtown Washington, DC, today, where most of the monumental construction dates to the late nineteenth and early twentieth centuries. The situation at Chichen Itza may have been similar to the collapse at Copan, where major construction at the center of the site ended in the late eighth century, the collapse of the dynasty occurred about a.d. 820, and near-total demographic collapse followed within a generation or, at most, two (Fash et al. 2003).

Although the major loss of population at Chichen Itza probably occurred within a few years of the political collapse, an event that may have resulted from conflict, possibly with the nascent city of Mayapan (Suhler and Freidel 1995), the site was never abandoned. Scattered archaeological evidence (see Chung 2000) and historical data attest to a continued, though much reduced, occupation of the site until the Spanish Conquest, when Francisco de Montejo the Younger established his headquarters at “Ciudad Real de Chichen Itza” in 1532.

This scenario leaves us with the question of what happened during the Early Postclassic period, which traditionally began with the rise of Chichen Itza in a.d. 1000 and ended with the rise of Mayapan in a.d. 1200. If our current reconstruction is correct, we are left with a 100- to 150-year gap in the chronology of the northern lowlands in which very little appears to have happened. There are several scenarios that may explain this: (1) evidence of late elite activity at Chichen Itza (i.e., post-a.d. 1000 or even 1100) is present but has not yet been documented; this is a possibility because large parts of Chichen Itza have not been excavated; (2) there was a “dark age” of 100–200 years in which elite activity ceased between the fall of Chichen Itza and the rise of Mayapan; (3) Mayapan and other Late Postclassic centers emerged before a.d. 1200, perhaps by a.d. 1100 (A.P. Andrews 1993). Recent excavations at Chichen Itza and a reinterpretation of earlier ceramic data suggest that the city was occupied until a.d. 1300 (Chung 2000). Evidence is also growing that Mayapan arose earlier than previously thought, perhaps as early as a.d. 1050 (Milbrath and Peraza 2001). Similarly, some of the monumental buildings in the “East Coast style” of architecture of coastal Quintana Roo, which has traditionally been dated to the Late Postclassic period, may date to the Early Postclassic period, before a.d. 1200 (Toscano 1994).

Current data therefore favor a combination of the latter two scenarios: a “dark age” followed by an early emergence of Mayapan. As we have discussed elsewhere (Andrews 1990; Robles and Andrews 1986), the Terminal Classic period in northern Yucatan was a time of growing stress. The Itza emergence came when overpopulation, land shortages, ecological stress, and climatic change were testing the capacity of the existing political and economic frameworks. Cities and towns erected fortifications, suggesting competition over land and other resources. The Itza were initially able to flourish despite these conditions, with an economy fueled by the control of coastal resources and access to external trade networks, coupled with the probable collection of tribute and the exploitation of labor and agricultural resources in the interior. As the Itza state grew, however, many cities in northern Yucatan ceased monumental construction and were severely depopulated, and some were eventually abandoned. Whether this was the result of a demographic and ecological crisis or of commercial and military competition with the Itza is not known; it may have been the outcome of a combination of these factors. Eventually, Chichen Itza, whose economy could no longer be sustained in the face of overpopulation and exhaustion of resources, also collapsed, sometime in the eleventh century.

Several authors have noted the possible role of climatic change in the Classic Maya Collapse, as there is increasing evidence to suggest that the Terminal Classic period was a dry and warm period in which the lowlands may have suffered severe droughts (Andrews 1990; Dahlin 1988; Folan et al. 1983; Gill 2000; Hodell et al. 1995). Most recently, David Hodell and colleagues (Hodell et al. 2001) have identified a major drought event brought on by solar forcing at ca. a.d. 1020, which coincides roughly with the period of decline at Chichen Itza. Other researchers, however (e.g., Dunning and Beach 2000:198), urge caution, believing that the evidence for a major pan-lowland drought at the end of the Classic period is not demonstrated.

In recent years, a number of Mayanists have suggested that the Itza owed their success not only to their mercantilist and other economic strategies, but also to their attempts to develop new forms of government. Epigraphic research by David Stuart (1993) and Krochock (1988, 1998; Krochock and Freidel 1994) suggests that a new type of political structure was forged at Chichen Itza, one that shifted the focus of power away from the Classic-period emphasis on individual rulers to the development of shared rulership, perhaps under the control of rulers from different social groups (Lincoln 1988; Schele and Freidel 1990; Wren and Schmidt 1991). Some scholars envision the emergence of a confederate type of government, similar to the multepal confederacy of later Postclassic states (Fox 1989; Freidel and Schele 1989; Schele and Freidel 1990). Rafael Cobos Palma (1999) has argued that the transition to shared rulership may have been temporary, corresponding to the early (ninth-century) period of Itza rule, when references to such rulership are alluded to in the inscriptions. In the tenth century, however, relying on historical and archaeological evidence, he suggests that the Itza reverted to having single rulers (for yet another perspective on this issue, see García Campillo 2001).

This departure from the Classic-period focus on individual rulers and lineages marked an evolutionary step in Maya political organization, in which the small dynastic state was replaced by a larger polity with a more complex state level of organization. This shift is evident in the development of historic and symbolic imagery at Chichen Itza. Classic Maya glyphic texts, which often provide a narrative record of major events in the lives of rulers, do not serve such a purpose at Chichen Itza; they record names of individuals and relationships among them, without much historical narrative. Moreover, glyphic texts are eventually replaced by iconographic imagery with a broad variety of themes that incorporate foreign elements. In short, Chichen Itza is thought by many to have been developing into a cosmopolitan Mesoamerican capital with a multiethnic social and political structure different from that of previous Maya polities (Freidel 1986; Schele and Freidel 1990; Wren and Schmidt 1991). The multiethnic character of Itza society is further suggested by García Campillo's (2000) lexical study of the inscriptions of Chichen Itza, which documents a strong Cholan component.

Although these developments appear to be radically different from those recorded in traditional Classic dynastic texts, there are earlier indications in the inscriptions of Xkalumkin of a shift in political authority from individual rulers to larger family or lineage groups (Grube 1994) and Copan (Fash et al. 1992; Fash 2001), suggesting that the evolution toward a confederate type of government had its beginnings in the Late Classic period (Krochock 1998). Itza attempts to forge a new political framework were thus rooted in the past and ultimately were not sufficient to overcome the crises they faced at the end of the first millennium.

The Terminal Classic decline of the northern cities was a delayed version of the massive societal collapse that occurred in the southern lowlands in the ninth century (Andrews 1990; Andrews V 1994; Robles and Andrews 1986). The Itza, while evidently experimenting with new forms of political organization, were unable to stop a series of processes that had begun in the south during the eighth century. Although Mayanists have not reached a consensus on which of these processes were the predominant causes of the collapse, they agree that rapidly rising populations, ecological degradation, a burgeoning noble class with declining prospects of wealth and power, warfare, and a failure of rulership contributed to the demise of the Classic Maya. These factors varied in intensity from region to region. The same factors appear to have been at work in northern Yucatan, where native chronicles attest to corrupt rulership, social strife, diseases, and endemic famines. The abandoned ruins that greeted the Spaniards in Yucatan bespoke the collapse of great cities.

With the collapse of the northern cities of the Terminal Classic period between a.d. 900 and 1000, the Yucatec Maya appear to have entered a dark age lasting more than a century. Sotuta and Cehpech ceramics fade by a.d. 1050–1100, replaced by ceramics of the Hocaba and Tases spheres (Robles 1987, 1988). We have little evidence of large-scale construction until the rise of Mayapan and the coastal cities of the Caribbean and southern Gulf Coast. These Postclassic cities probably began to emerge in the twelfth century. Smaller than many cities and states of the Classic period, the coastal communities were adapting to changed political and economic conditions, more responsive to developments elsewhere in Mesoamerica than their predecessors. The new trajectory was cut short by the arrival of the Spaniards.

Mayapan is traditionally thought to have seen its major settlement from a.d. 1200 or 1250 to 1450, but the beginning date is uncertain (Brown 2001). Some archaeologists are now willing to entertain a date 100 or 150 years earlier. This view results in part from the inability of excavators and ceramicists to demonstrate that the Hocaba and Tases ceramic complexes at Mayapan (Smith 1971) characterize separate and sequential phases. The appearance of Hocaba pottery (Peto Cream Ware) at the end of the Chichen Itza sequence therefore suggests that the rise of Mayapan follows the decline of Chichen Itza by only a short interval. Milbrath and Peraza (2001) believe that a significant occupation at Mayapan may date back to the eleventh century. Stelae 5 (4 Ahau) and 6 (13 Ahau) at Mayapan probably date to the thirteenth century (a.d. 1244 and 1283). Stela 1 (10 Ahau), and an altar fragment with the date 1 Ahau may date to a.d. 1185 and 1141, or to a.d. 1441 and 1401 (Proskouriakoff 1962:135; Schele and Mathews 1998:135). The earlier dates for Stela 1 and the altar fragment would precede the 13 Ahau (a.d. 1283) date mentioned in native chronicles as the founding date of Mayapan.

These changing views of late northern Maya archaeology lead to several judgments about Maya cultural history. None of these conclusions is entirely new, but none has until now been widely accepted. First, Chichen Itza is now correctly viewed as a Late Classic Maya capital—it can no longer be seen as representing a major break with the Classic past and the beginning of a new Postclassic era. Its demise was the product of the same processes that characterized the end of the Classic horizon. Second, it is no longer appropriate to consider the “southern Classic collapse” as a regionally restricted phenomenon. We believe there was a pan-lowland collapse. Third, if we accept Chichen Itza as the last of the Classic-period states, the term “Early Postclassic” must then refer to a period that is transitional. This time witnessed the decline and abandonment of many northern cities, a likely dark age in which the worst prophecies of Maya priests came to be, and, at the same time, the stirrings of a new era (especially on the Caribbean coast). We think that the term “Early Postclassic” is too laden with obsolete associations and that it has lost much of its meaning. It no longer signifies the beginning of a new era, and it does not convey a sense of the collapse and chaos of the tenth and eleventh centuries in the northern lowlands. We therefore believe the term should no longer be used. The eleventh century was a transition—an interregnum—from a Late or Terminal Classic period, characterized by Cehpech, Sotuta, and Hocaba ceramics and related styles of architecture, to a Postclassic period defined by Hocaba and Tases ceramics and a Mayapan-East coast style of architecture.