A Small Rural Travel Stopover at the Late Postclassic Maya Site of Mensabak, Chiapas, Mexico: Overland Trade, Cross-Cultural Interaction and Social Cohesion in the Countryside

A small rural stopover along overland Maya and Aztec trade and travel routes was identified in surveys and excavations at adjacent settlements and shrines at Mensabak, Chiapas, Mexico. This collection of Late Postclassic to Spanish conquest-era (c. ad 1350–1650) Maya sites are similar in function to rural Old World and Andean caravan stopovers, such as caravanserai and way stations, where travellers and traders obtained supplies, trading partners, safety, solidarity through ritual and travel information along long-distance land routes. These sites are similar to trading ports and pilgrimage centres, but they are smaller, located in the countryside, not often managed by regional states, and have scaled-down economic exchange with fewer exotic trade items. Stopovers often include landscape and rock-art shrines for collective ritual among foreign travellers and local populations. While investigators have researched the anthropological importance of overland routes, caravans and trade centres, less attention has been given to stopover sites in the countryside. This article discusses the archaeological signatures and outlines the comparative social, economic and ritual implications of small rural stopover sites that united people on the road.

connections with the population centres of ports (Alexander 2005;Andrews 2008;Chapman 1957;Franklin & Boak 2021;Polanyi 1963). Stopovers are placed along interior trade routes in rural areas and do not have the large amounts of diverse exotic goods of the trading ports (Clark 2016). Stopovers also differ from pilgrimage centres (Palka 2014) since they have more permanent habitation, fewer religious buildings, more storage facilities and more defensive capabilities. Stopovers are also larger and have bigger buildings and more permanent residents when compared to travellers' camps, resting places and depots (Nielsen 2021). Small rural stopovers are more like way stations, outposts or caravanserais that were loosely connected through subsidiary routes to ports of trade and demographic centres in regional trade networks (Berenguer 2021;Edwards 2021;Mader et al. 2021;Nielsen 2021;Siveroni 2021). In colonial period Mesoamerica, some of these small indigenous rural sites with habitational structures and shrines became Spanish parajes, paradas and estancias maintained by local communities for travellers visiting markets, political centres and pilgrimage centres. These sites were located either on main roads or informal routes and paths (caminos, corredores, vías, senderos) across the realm (Bonilla Palmeros 2020; Lee & Navarrete 1978;Long Towell & Attolini Lecón 2010). One such rural stopover at Late Postclassic to early historic (c. AD 1350-1650) Maya sites at Mensabak in Chiapas, Mexico, is discussed below in comparative context (Fig. 1). I construct a model for specific human behaviours at rural stopover sites that have archaeological correlates and anthropological significance.
Importantly, the locations and material culture of rural stopovers make them archaeologically visible. Small rural stopover places have not been studied as often historically and archaeologically as larger trading ports, rural centres and pilgrimage shrines, but they can be examined with surveys and excavations by archaeologists conducting complete ground truthing in the countryside (Berenguer 2021;Erickson-Gini & Israel 2013;Jiménez Gómez 2010;Nielsen 2021;Palka 2014). Historical accounts provide information regarding the locations, nature, economic structure and social implications of these interesting sites (Jiménez Gómez 2010; Morante López 2010). However, we often lack detailed historical descriptions of the stopovers and life within them because chroniclers usually do not describe small rural settlements. Historical information on Maya travel routes and rural sites have been used to reconstruct regional interaction and economies (Caso Barrera 2002;Feldman 2000;Woodfill 2019), but we know less about stopover sites in countryside and lifeways there. Archaeology and comparative analyses help fill this informational gap. In this article, I place rural stopovers in an anthropological context and discuss salient human behaviours at these kinds of sites with the archaeological case study of Mensabak, Chiapas, Mexico.

Stopover sites in comparative analysis
Cross-culturally, small rural stopovers have existed in societies practising long-distance mercantilism, overland travel and pilgrimage between religious shrines. People require places to stop safely between the more populous towns and they need sustenance and crucial information about hostile regions. European traders, for instance, safely lodged in small rural Islamic caravanserais because of the economic importance and protection of local merchants there (Burns 1971;Burton [1893Burton [ ] 1964. In Mesoamerica, some small stopovers were incorporated into ritual landscapes, many with rock-art shrines, along distant trade routes that drew in local people, merchants, pilgrims and travellers for collective exchange, rituals and social ties (Pye & Gutiérrez 2007;Reyes Esquiguas 2010, 620). The group rituals helped people feel secure and they promoted success in travel, social interaction and trade from the assistance of resident deities and local inhabitants (Dibble & Anderson 1959;Palka 2014). Subsequently, at small rural stopover sites people joined in trade, ceremony, protection and social solidarity, making them cross-culturally significant for human settlement and interaction. Hence, the investigation of rural stopovers carries important anthropological and material insights into human behaviour regarding travel, economic interaction, religious ritual and social cohesion among people trekking in the countryside. Archaeology is significant for this endeavour since these often neglected sites can be excavated to learn about them and human behaviours associated with them.
To date, archaeological, historical and anthropological studies of travel have concentrated on roads, connections between centres, economic organization and phenomenology (Candy 2009;Franklin & Boak 2021;Hyslop 1984;Núñez & Briones 2021;Rueda 2010;Trombold 1991;Van Dyke 2007). Investigations of travel routes routinely focus on main routes and site locations-typically townsand connections between them, along with reconstructions of trade and economic behaviours (Feldman 1985;Lee 1978). Scholars often look at pragmatic considerations, such as travel times, strategic town placement and the geographical organization of travel sites (Gorenstein & Pollard 1991;Navarrete 1978). Other investigators theorize on the function and social importance of travel and trade sites, including ports of trade, market centres, trade diasporas, international trade centres, storage facilities and gateway communities (Andrews et al. 1988;Chapman 1957;Clark 2016;Edwards 2021;Hirth 1978;Nielsen 2021;Polanyi 1963;Stein 1999).
Yet small rural travel stopovers, traffic ports, outposts and caravanserais on less-travelled paths and their social and economic importance have not received adequate treatment in travel and settlement studies (Long Towell & Attolini Lecón 2010;Navarrete 1978;Nossov 2013;Yavuz 1997). However, investigations are increasing regarding human interaction and behaviour at caravan campsites, rest areas, small rural centres, oases and caravanserais on secondary routes through ethnoarchaeology, ethnohistory and archaeological excavations (Darnell 2021;Mader et al. 2021;Nielsen 2021;Núñez & Briones 2021;Palka 2014;Siveroni 2021). Economic and political power structures also are important at these rural places, which can be managed locally or sometimes impacted by state polities (Darnell 2021). For comparative purposes here, small travel stopovers in rural Mesoamerica were similar to Eurasian caravanserais and Andes outposts (Mader et al. 2021, 178;Siveroni 2021, 114). Their similar economic, religious and integrative characteristics were practical for travellers and important for social cohesion.

Characteristics of small rural stopovers
Across the Old World, caravans and groups of people covering hundreds of kilometres in rural areas needed safe places to stop so traders, travellers and pilgrims could rest, acquire food and water, exchange goods and make religious offerings. In the particular cases of Eurasian caravan stopovers, including caravanserais, they were dispersed in the countryside far from towns (Fig. 2;Johansen 2016;Shokoohy 1983). Many were commonly known destinations along principal (formal) or less-travelled (informal) long-distance routes (Bryce et al. 2013;Clarkson & Fowler 2021;Trombold 1991) since people seeking hospitable stopovers required rest and crucial information on where they could safely venture. Importantly, religious shrines united the travellers in collective ritual due to the uncertainties and dangers of rural travel (Bryce et al. 2013;Johansen 2016;Shokoohy 1983). Travellers need to communicate with spiritual forces through ritual for the wellbeing of people and their goods, and for the success of the trip (Nielsen 2021, 20, 26-8, 32-3). These rituals have taken place at landscape altars, temples in settlements and rock-art sites. After collective rituals, travellers shared meals, interacted socially and exchanged information (Nielsen 2021, 28-9).
Other desirable characteristics of small rural stopovers included their small markets and receptive local populations where traders exchanged goods and economic information (Yavuz 1997). The stopovers led to interaction between people who did not have kinship ties and who were not united in cultural practices, ethnicity, or political affiliation (Bryce et al. 2013;Nielsen 2021, 35;Núñez & Briones 2021, 222-3). For example, rural stopovers with their hospices, markets, water and shrines, such as 'travel villages', ritual rock cairns, ecclesiastical estates and caravanserais, were valuable for the solidarity and safety of the multinational travellers on the pilgrimage route to Santiago de Compostela (Candy 2009, 42-7, 60-61, 80-81). Clearly small rural stopover places helped different groups of people form social bonds and allowed them successfully to undertake their travels, trade and religious worship across the world.
Central elements of stopover sites include their placement along rural trade routes, small settlements, water availability, structures for storage and sleeping, fortified safe havens, public religious shrines, ritual landscapes and social places for small-scale trade of exotic goods (Dale 1994;Edwards 2021;Mader et al. 2021;Nielsen 2021;Thareani-Sussely 2007). Frequently, rural stopovers are politically autonomous. These sites, too, can managed by local governments that may be loosely connected to regional states (Darnell 2021;Nielsen 2021, 30). However, sometimes they can actually be controlled by regional polities if they are on main trade or pilgrimage routes (Edwards 2021;Franklin & Boak 2021). Burton ([1893Burton ([ ] 1964 describes some of these attributes at a rural small site, Al-Hamra, found on an alternate route to Meccah: It [Al-Hamra] is built on a narrow shelf at the top of a precipitous [defensible] hill . . .
[here] water of good quality is readily found . . . Al-Hamra is a collection of stunted houses made of unbaked brick. It appears thickly populated in the parts where the walls are standing . . . It is well-supplied with provisions . . . The bazar [market] is a long lane, here covered with matting . . . Near the encamping ground of caravans is a fort for the officer commanding a troop of Albanian cavalry, whose duty is to defend the village, to hold the country, and to escort travellers.
Archaeologically, such rural stopovers can be recognized as small clusters of structures near springs or rivers with reduced market spaces and few habitational structures, some of which were fortified (Berenguer 2021;Bryce et al. 2013;Burns 1971;Franklin & Boak 2021;Navarrete 1978;Nielsen 2021;Núñez & Briones 2021). Additionally, these sites can be found along travel routes on maps and aerial photographs, in travellers' accounts and through survey in rural areas between centres. Small caravanserais along the Silk Road and trade routes in Iran, for instance, have drawn the attention of art historians and archaeologists examining their constructions, functions and local religious life (Erickson-Gini & Israel 2013;Johansen 2016).
For the Americas, one historical and archaeological example of a stopover in the countryside is the Inca site of Cajamarca, Peru. Cajamarca was a small town founded by local indigenous people that became a stopover (tambo or tampu, 'resting place') on the Inca road far north from the capital of Cuzco (Hyslop 1984, 56-61;Julien 1993). Elites supported these rural tambos, or estancias, for officials' journeys, that were central for storage, rest, and the movement of people and goods along the extensive Inca road system (Garrido 2016;Hyslop 1984). Local indigenous populations resided at Cajamarca to maintain its buildings, storage facilities, plaza, temples and structures for Inca nobles and Spaniards when they travelled with their caravans. People stayed in the site's buildings or in tents. Water was available for travellers and adjacent hot springs served as baths. Goods were exchanged at the site, especially in its plaza that was fortified. Additionally, a hilltop fortress was found close by if violence broke out in this rural area. Travellers, too, undertook rituals in the plaza and adjacent temples (Julien 1993, 252). On Andean travel routes, rock-art shrines were important for collective rituals on the landscape at rural stopover sites and along rural routes (Berenguer 2021;Mader et al. 2021;Nielsen 2021, 32).
One archaeological example of a rural Mesoamerican stopover is the ancient site of Los Horcones, which rests on the Pacific coast of Chiapas between major ancient economic and political centres. People established this protected site on an impressive hill with shrines, some with rock art, located along a trade and travel route in this coastal cacao-growing area. Los Horcones has temples and ballcourts for collective community rituals, numerous buildings for sleeping and storage and enclosed plaza areas for trade and social interaction between local people and traders. The stone monuments, ceramics and architecture at the site point to strong ties with people from the Classic-period (c. AD 300-500) metropolis of Teotihuacan in Central Mexico (Garciá-Des Lauriers 2016). People were moving between Teotihuacan, sites in Veracruz, the lowland Soconusco cacao area in Chiapas and coastal Guatemalan towns. Monuments depicting war gods in addition to enclosed plazas on elevated areas point to the defence of the site's inhabitants. The large percentage of green Pachuca obsidian (about 40 per cent of the assemblage) brought from Central Mexico to Los Horcones underscores the strong ties with Teotihuacan merchants travelling to the Chiapas coast for trade in obsidian and cacao beans. Long-distance trade and travel across Mesoamerica and the setting up of stopovers and ritual landscape shrines with unified iconography (and rock art) along overland routes had existed for centuries (Pye & Gutiérrez 2007), but we have few details regarding everyday life at these rural places. Fortunately, scholars have documented comparative travel sites and trade routes from historic documents and archaeological survey in Mesoamerica (Chapman 1957;Feldman 1985;Long Towell & Attolini Lecón 2010;Navarrete 1978). Their insights are important for understanding ancient stopover sites at Mensabak and elsewhere.

Mesoamerican trade caravans and stopover sites
Long-distance travel for trade and pilgrimage with human caravans in Mesoamerica was widely documented following the Spanish conquest (Berdan A Small Rural Travel Stopover at the Late Postclassic Maya Site of Mensabak, Chiapas, Mexico et al. 2003;Chapman 1957;Clarkson & Fowler 2021). Large numbers of people travelled along routes through mountains and forests across political boundaries. The travellers included merchants, pilgrims, political officials, migrants, tribute collectors, warriors, spies and porters. In these rural areas, travel was often hazardous (Dibble & Anderson 1959;Viqueria 2002, 119, 147). Some settlements were not always friendly to outsiders. Hence, local governments, merchants and soldiers helped maintain small rural settlements for safety along the long routes. Ports of trade, trading diasporas, towns and small stopover sites in between were developed to facilitate trade and travel to central economic destinations. Sites with markets and pilgrimage shrines were more accepting of travellers and strangers due to social interaction and the importance of peaceful trade (Chapman 1957;Gaxiola González 2010). In Chiapas, Aztec and Maya merchants crossed the region exchanging their wares and travel information. These merchants organized human caravans and negotiated trade in foreign areas as part of their specializations (Chapman 1957, 115;Clarkson & Fowler 2021).
Mesoamerican historical chronicles provide evidence for long-distance travel routes and their small rural stopovers. Some of the better-described historic routes are found in southern Mexico (Attolini Lecón 2010;Chapman 1957;Dibble & Anderson 1959;Viqueira 2002). However, smaller radial routes in rural areas off the main roads connected towns and stopovers (Rueda 2010;Sheseña et al. 2021), but they are not as well described as the main routes. Colonial period maps from Chiapas (Fig. 3) show several possible paths and rural stopovers and not just large centres (Lee & Navarrete 1978, 78, 96, 103). One extensive rural route went through several towns, farmsteads (ranches, estancias) and resting places (parajes) ranging between one and four leagues apart (around 5-20 km; Navarrete 1978, 77-84). It was important for merchants to reach rest points at the ranches and small stopovers along these overland routes so they could safely exchange their products with local populations (Lee 1978). For instance, roads, paths, ritual sites and stopovers connected centres in Guerrero, Oaxaca and central Chiapas with the Pacific coast Aztec province of Soconusco (Xoconochco; Berdan et al. 2003;Feldman 1985;Pye & Gutiérrez 2007), which was critically important for the trade in cacao and other regional products.
One account mentions rural stopovers along a rural Chiapas route: the road passes through valleys with few inhabitants and since the traffic does not earn enough to permit the carrying of forage and provisions, settlements sprung up at 8 km. or so intervals to feed the men and animals; their aspect and function are determined completely by the traffic. These serve for resting and staying overnight. (Navarrete 1978, 83) Catholic priests also helped maintain and integrate the routes and their travellers by creating churches, shrines and missions at stops along them (Viqueira 2002, 125-34). The colonial governments ensured their protection and viability. Therefore, these rural stopovers had important economic, religious, political and social functions for the caravans.
In Spanish conquest-era Chiapas, caravan trade and travel routes crossed extremely rural areas since highly prized Aztec and Maya products, including cacao, tobacco, feathers, incense and amber, were available here (Feldman 1985;Lee 1978;Viqueira 2002). Aztec pochteca merchants traded green Pachuca obsidian, copper bells and cotton cloth for these highly sought Maya items Dibble & Anderson 1959). Importantly, Aztec merchants established garrisons and trade enclaves at Zinacantan (Zinacantlan) in the highlands and in their cacao-growing province of Soconusco on the Pacific coast to protect their caravans (Kohler 1978). Similar caravans of Maya traders crossed the area, which was not always safe due to the politically fragmented landscape with its violence, random trade taxation and piracy. Aztec and Maya traders also maintained markets and garrisons on the Gulf Coast, which were connected economically and socially with trade towns and small rural stopovers in Chiapas (Attolini Lecón 2010; Berdan et al. 2003;Chapman 1957). Travel routes ran through the Chiapas highlands connecting this region with the Gulf Coast to the north and with Aztec Central Mexico and the Guatemalan highlands (Feldman 1985;Lee 1978). Another important route was through the countryside along the Pacific Coast of Chiapas that continued to Oaxaca and Central Mexico (Attolini Lecón 2010;Feldman 1985). An additional rural route passed through the towns of Zinacantan and Comitan (near San Cristobal or Ciudad Real) heading north to the lowlands and Gulf Coast via Ocosingo (Ocosinco) in Chiapas (De Vos 1980;Navarrete 1978;Taladoire 2016). Modern roadways follow these ancient rural routes; one secondary road from Ocosingo to Palenque passes near Mensabak, as discussed below (see Figure 3). Another less-travelled path incorporated the rural Ch'ol Maya towns to Tumbalá and Tila, which had pilgrimage shrines, ritual caves with paintings (rock art) and market areas that integrated travel, commerce, religion and different social groups (Bassie-Sweet 2015;Navarrete 1978, 93).
Significantly, Acalan ('Place of [Trading] Canoes') Chontal Maya merchants of the Gulf Coast of Tabasco supported the criss-crossing trade routes through the countryside that connected their lands with Central Mexico, Guatemala, Yucatán and Honduras (Attolini Lecón 2010; Chapman 1957). These Chontal Maya actively maintained exchange ties with Maya and Zoque peoples in the rural sierras of Chiapas to the south to provide their markets with exotic bird feathers, animal pelts, incense and tobacco (Navarrete 1978). Tenosique (Tanoche or Tanosic), Tabasco, played a major role in these economic interactions since the small town, which was jointly occupied by Chontal and Yucatec Mayas (also Aztecs intermittently), is located where the Usumacinta River is first navigable downriver to the Gulf Coast (De Vos 1980). One overland secondary route connected Tenosique with Ocosingo, which Maya from Mensabak (perhaps called Nohha; De Vos 1980) used to visit these towns (Fig. 4). Another path followed valleys from rural Mensabak towards the large Ch'olti'-Lacandon settlement of Lakamtun at Lake Miramar just to the south. This informal route was linked to a path from Palenque and Tumbalá to the Lake Miramar settlements through Pochutla on Lake Ocotal Grande ( Fig. 4; Bassie-Sweet 2015; De Vos 1980). Significantly, Mensabak rests right on the crossroads of the rural Ocosingo to Tenosique and Tumbalá/Palenque to Ocotal/Miramar routes (De Vos 1980, 507, 510). Historic Lacandon Maya from Mensabak actually travelled along these informal paths to Palenque, Ocosingo and Tenosique to trade their tobacco for salt, cloth and metal tools until recently (Palka 2005;2017). The documentary evidence regarding trade and travel routes, in addition to archaeological data, indicates that Aztec and Maya merchants and travellers walking the countryside between the Gulf Coast, Zinacantan in the highlands, and then to Soconusco on the Pacific Coast, used Mensabak as a rural stopover site because of its special geographical, social, ritual and economic characteristics. Hence, this site provides an ideal archaeological case study to examine social life at these sites.

Case study: a small rural stopover at Mensabak, Chiapas
Long-distance overland travel stopovers can be examined through archaeology for insights on specific human behaviours associated with them. Their characteristics, especially their locations on rural travel routes, abundant water, buildings for sleeping and storage, fortifications for safety, trade in exotic items, small populations and public religious shrines can be identified through survey, excavation and artifacts. Postclassic sites dot the Tabasco plain (Scholes & Roys 1968), but few coeval sites have been encountered in adjacent northern Chiapas, including along the Usumacinta and Tulijá river routes (Palka & Lozada Toledo 2018). One site was discovered along a secondary long-distance travel, trade and pilgrimage route at rural Mensabak, Chiapas, Mexico (Palka 2014). Mensabak (also Mensäbäk) is a small cluster of archaeological sites at a small lake located in the Sierras of Chiapas, about 250 km to the south of the lowland coastal Maya and Aztec trade enclaves and port towns of Potonchan and Xicalango (Xicalanco) on the Gulf Coast of Tabasco Chapman 1957; see Figure 1). The sites and shrines at Mensabak were discovered during the author's collaborative rural archaeological surveys with local  (Palka 2014;Palka et al. 2020). We research an area far from regional Maya centres and ports, which have drawn more archaeological attention due to the presence of Maya elite material culture like palaces and hieroglyphic monuments. Mensabak is a rural stopover site with multiple communal rock-art shrines, canoe ports, plazas and several protected settlements spread out along the shore of a small lake. The Mensabak site as a whole acted as a safe haven, religious sector and small-scale market area. Mensabak also has plentiful water from its lakes, springs and streams along with abundant foods from the waters, fields and forests for residents and travellers. Mensabak did not function as a trading port since it is not located on main travel routes managed by regional polities. Hence, the adjacent lakeside settlements, plazas and shrines here functioned together as a rural stopover; there was not just one central site or port.
Two small settlements dating to around the time of Spanish colonization (Late Postclassic to early historic times: c. AD 1350-1650) were founded by rural Maya leaders and their allies at Mensabak (Palka 2014). The two settlements, Tzibana and La Punta, were placed on fortified peninsulas on the lake shore by Maya elites and merchants to attract travellers, traders and pilgrims. Both sites have large canoe ports where travellers docked their canoes, unloaded their goods and sought food, shelter and social interaction. Diagnostic Late Postclassic ceramic types are found at these sites, including Matillas Fine Orange wares from the Chontal Maya Gulf Coast, in addition to small footed bowls and colanders (Fig. 5). Maya lineages and their allies migrated to Mensabak specifically because of the culturally significant pilgrimage shrines located there (Palka in press). Mirador Mountain (also El Mirador), an impressive, unusual mountain with a sheer, red-stained cliff rising out of the lake, dominates the Mensabak landscape (Fig. 6). Maya travellers arrived at a canoe port at the Ixtabay temple complex (see Figure 1) and ascended this mountain for rituals at a stone-block temple near a cave on its summit. This 'water mountain' with temples on an island in an idyllic lake recalls the Aztec mythological/historical homeland of Aztlan and altepetl ('community'). Mirador Mountain even has a spring that initiates a major river (the Río Tulijá) recalled in Aztec Aztlan myths. Maya and Aztec pilgrims, Aztec pochteca and Maya merchants and travellers arrived at Mensabak where they ascended Mirador Mountain to conduct ceremonies at landscape shrines here. They also performed collective rituals with the local community at nearby landscape and public rock-art shrines.
One site, La Punta, is protected with defensive walls and residences along the lakeshore every 4-6 metres, like bastions (Fig. 7). An artificial stone-block island (Str. 11) provided a check-point at the canoe port entrance. The large, well-formed canoe port was the focal point of this settlement, enhancing its function as a stopover site. Storage buildings (Strs. 4, 8 and 57), public structures (Strs. 3, 58 and 59) near a small plaza for unloading and gathering (Strs. 60 and 61) and residences (Strs. 10, 12, 66 and 67) were located around the canoe port. A stairway from the canoe port rises to a large structure compound (Strs. 2 and 9) on a small plaza with a central altar and stela for ceremonies. Some traders and pilgrims went up the stairs to join local leaders in collective ritual feasts and to exchange goods in the plaza. Buildings on the plaza and on the site's hilltop shrine near cliffs (Strs. 51 and 55) were aligned to the temple and cave on Mirador Mountain, enhancing the site's religious importance. The hilltop shrine's plaza would have held numerous ritual participants, which indicates its communal function.
Archaeologists recovered large amounts of animal bone from feasting and some trade goods on the plaza (Palka 2014). The trade goods included a small amount of Matillas Fine Orange ceramics produced by Chontal Maya in Tabasco, a marine shell bangle from Soconusco (Voorhies & Gasco 2004), a few copper bells from Central and West Mexico (Hosler 1994) and many green obsidian blades (about 40 per cent of the obsidian artifacts) from the Pachuca source in Central Mexico, brought by Aztec pochteca merchants ( Fig. 8; Berdan et al. 2003;Clark et al. 1989;Smith 1990). Sites located along Aztec travel and trade routes and colonies along the Pacific Coast in southern Mexico, such as Tututepec, Oaxaca and Ocelocalco in Soconusco on the Chiapas Pacific coast, typically contain a large quantity (over 30 per cent) of Pachuca obsidian (Braswell 2003;Clark & Lee 2007;Clark et al. 1989;Levine et al. 2011;Ohnersorgen 2006;Smith 1990). Most Late Postclassic sites in Chiapas have low quantities (less than 10 per cent) of green Pachuca obsidian. Aztec Nahuatl terms also appear in the local Lacandon Mayan language and in historic Maya groups in the region (De Vos 1980), pointing to significant long-term interaction with Aztec people. Small quantities of Central Mexican copper bells also occur in households at other Mensabak settlements, underscoring their importance in interregional trade and ritual. It is possible that Aztec pochteca merchants visited Mensabak and its Aztlan-like Mirador Mountain, since the site rests along a rural route between their garrisons on the Gulf Coast, Zinacantan in the Chiapas highlands and Soconusco. Traders probably acquired local tobacco, animal skins and feathers, which were prized in Postclassic times, including in Central Mexico (Palka 2017). Moreover, Maya merchants from Tabasco also arrived at Mensabak, since they actively participated in the trade and ritual life involving Aztec pochteca in this region. La Punta was a safe residential site, a place for storing goods, a setting for communal feasting and a stopover for travellers to interact, trade and perform rituals with local political leaders and inhabitants.
Another Mensabak settlement, Tzibana (also Tz'ib'ana), is a small site across the lake from La Punta with a large plaza surrounded by a cluster of buildings and defensive walls uphill from its large canoe port (Fig. 9). A defensive wall blocks access to the site from the canoe port. Important Maya families resided at Tzibana, which is evidenced by the presence of large Late Postclassic residences made of finely dressed blocks. The site has only a few residential structures, but their sizable construction and architectural complexity point to the wealth of the inhabitants, which was probably gained through trade. Travellers were drawn to Tzibana because of its important Maya families, ample plaza, location near water and the Late Postclassic ceremonial shrines at ruined ancient temples, a nearby cliff with rock art rising from the lake (Fig. 10) and a large lakeside cave, which was full of ritual materials. The multitude of rock-art designs, including hunting scenes, animal totems or spirit familiars and hand prints of men, women and children (Palka & Lozada Toledo 2022), speak to the communal nature of the rites at the cliff. Additionally, the numerous human burials and utilitarian ceramics in the cavern behind the cliff also evidence the ceremonial involvement of the community and not just a few religious specialists or elites. Postclassic people did not construct residences on the tall Preclassic (c. 200 BC-AD 200) temples, but they may have used them for public rituals and feasts according to sporadic finds of broken ceramics and animal bone on their surfaces. Whereas La Punta was a political and economic hub at Mensabak, the inhabitants of Tzibana demonstrated their families' wealth and their abilities to conduct collective ceremonies at important shrines to visitors. The travellers were likely to have used the many residential and storage structures around its plaza where trade was conducted. Traders brought small quantities of Chontal Maya Matillas Fine Orange ceramics from Tabasco, Mexican copper bells and exotic marine shell. Furthermore, they exchanged a large percentage (about 35 per cent) of Central Mexican Pachuca green obsidian, pointing again to the presence of Aztec traders, or their direct influence on Chontal Maya merchants, in the local economy as historically and archaeologically seen in Chiapas and nearby Oaxaca (Clark et al. 1989;Levine et al. 2011). Hence, Tzibana was the secure locale for the exchange of goods and collective rituals at religious shrines. It is likely that a small number of travellers rested at Tzibana before continuing on their journeys.
Other comparable Postclassic Maya sites in Chiapas and nearby Petén, Guatemala, differ from what we have seen at the Mensabak rural stopover. For instance, the large island site of Topoxte in Petén was a town with large stone block temples and palaces (Wurster 2000). Topoxte was a political centre of the Kowoj Maya polity that interacted with the nearby regionally powerful Itza Maya and their many sites of the Petén Lakes area (Rice & Rice 2009). Another example, the large site of Canajaste in highland Chiapas was strongly fortified and not very accessible, and it had a large population comparable to other contemporaneous centres in the region (Blake 2010). Canajaste Maya interacted with adjacent populations along the trade corridors of the Grijalva River system and this was not built in a rural area. Moreover, this centre, in addition to the Petén sites, did not exist in the far countryside, nor did they have communal landscape and rock-art shrines for use in collective rituals. One Late Postclassic Maya site, Isla Cilvituk, located in rural Campeche (Alexander 2005), may have functioned as a defensible stopover along regional trade and travel routes. However, the site is rather large compared to Mensabak settlements and it may have been a secondary centre. Additionally, no evidence was found at Cilvituk for canoe ports and storage, nor collective ritual at temples or landscape shrines.

Conclusions: rural stopovers in cultural context
The identification of a small rural stopover for travellers and traders along an informal, secondary route at Mensabak, Chiapas, underscores the economic, social and religious components of long-distance overland caravans and road networks seen in comparative contexts. Insights from archaeological studies of countryside stopovers add to knowledge of the complexity of social interaction and settlement types in these material and cultural contexts besides the trading ports, colonies and major routes that have been studied across the world. The defining characteristics of small rural stopovers seen in Mesoamerica and Eurasia and in the Andes are their locations in the distant countryside along travel routes, general availability of food and water, buildings for resting and storage, protective constructions for safe havens, places for small-scale trade and social interaction, the presence of people of different ethnicities, and religious shrines in the landscape, often with rock art, for collective worship. In the case of Mensabak, traders and pilgrims arrived here and interacted with the populations of these small rural settlements, such as La Punta and Tzibana. No direct architectural evidence or human burials currently point to an Aztec or Chontal Maya trade diaspora colony at Mensabak, although contact and trade with these people are seen in the substantial amounts of green Pachuca obsidian blades from Central Mexico and the Matillas Fine Orange ceramics from Tabasco. Maya leaders and merchants, and probably Aztec pochteca traders, maintained safe access to this site and they may have influenced local political and economic organizations.
Mensabak was not a port and not just a location of pilgrimage shrines. The adjacent site of Toniná near Ocosingo, on the other hand, was a Postclassic Maya pilgrimage centre where people buried their dead and performed rituals in ancient Maya ruins. At this site and others in the Ocosingo valley, only small amounts of Pachuca green obsidian, Matillas Fine Orange ceramics and copper bells have been found (Becquelin & Baudez 1979). These items recovered in domestic contexts and in plazas at Mensabak point to the presence of traders and Aztec merchants travelling between the Gulf Coast and the Chiapas highlands, and not just pilgrims visiting temples. Furthermore, Mensabak is more likely a small rural trade route stopover (caravanserai) rather than a Mesoamerican gateway centre like Chalcatzingo in Morelos (Hirth 1978), the international trade town of Xicalanco , a major pilgrimage centre like Cozumel (Freidel & Sabloff 1984), or a coastal port of trade as seen at Potonchan (Chapman 1957), because of its small settlement size, rural location, location along rural overland routes and comparatively small amounts of exotic trade goods compared to ports. Trade at Mensabak was limited to select local items, like the historically known tobacco, animal skins, cacao, ceramics, copper bells, and green obsidian, rather than the many different types and larger quantities of exotic goods known for larger, more populous trading ports (Clark 2016;Hirth 1978). The larger towns closer to markets, main routes, and denser populations in the Gulf Coast and Chiapas highlands served as gateway centres and international trading centres on the formal overland trade routes.
Small rural stopover sites can be discovered and examined further with archaeology and ethnohistory, but investigators have to shift from centres and ports to often neglected rural areas, particularly in Mesoamerica. Historical documents (Feldman 1985;Lee & Navarrete 1978;Long Towell & Attolini Lecón 2010) and aerial LiDAR mapping (Ensley et al. 2021) demonstrate that small rural sites, and possible stopovers, were common across time throughout this region. In surveys and excavations, we can recover evidence for rural stopovers on travel routes, such as small sites with buildings and walls for protection, items related to longdistance trade, evidence of trails and countryside settlements, public interaction and trading spaces and communal religious shrines. The public areas for exchange, central buildings for storage and sleeping and religious shrines integrated local populations socially with travellers and merchants. Landscape shrines for collective ritual of travellers seeking safety and economic success while on the road, particularly at rock-art sites, are important elements for rural stopovers as well. The ritual, trade and the exchange of information lead to social interaction and solidarity at these intriguing sites. While some rural sites may have served as pilgrimage shrines or gateway communities, the smaller ones can function as practical stopovers and places for exchange among travellers between centres, especially if they are located on trade or pilgrimage routes. These places could also have served in small trade fairs where travellers and merchants met at rural pilgrimage shrines on secondary travel routes to exchange their goods and participate in ritual at prescribed times (Feldman 1985, 19-20;Freidel & Sabloff 1984;Palka 2014). Furthermore, the stopovers can eventually become settlements attracting people due to their social, economic, protective and ritual significance. The presence of these small countryside sites and the growth of settlements around them accentuate their importance in human history as being crucial for integrating different ethnic groups, regional economies and powerful polities as long as they, the trade routes, collective religious rituals and interconnected centres are maintained.