Cosmological models have been highly influential in Mayanist studies and frankly, I do not like them at all. Where we find the strongest bastion for cosmological modelling in contemporary Mayanist research is probably in the subfield of cave studies. Karen Bassie-Sweet followed in the footsteps of Schele and Freidel and although her work has been criticized by Brady and Prufer, there are no major differences in how these researchers all view caves as central to various cosmological models. The difference is one of degree, not one of kind. They all agree that cosmology is the foundation for politics, economics, etc. These heavily idealist approaches could use some materialist injections to explain the far more complicated picture that exist.
As I see it we have two extreme options in the study of the relationship between caves and cosmology. We can, as Freidel, Schele, Stanton and others do, use a cosmological model that is as generalized as possible so that it can incorporate all changes through three millennia. This research into cosmology is employed by most Mayanists. It relies on an arborescent/tree like or hierarchical view of reasoning where everything within the Maya cosmology can be signified by a few core concepts. This means that each specific cave is interpreted from a more general idea of cosmology. However, generalities are linguistic conventions and have no reality apart from the linguistic system. In reality, only haecceities exist.
The other option is to see “cosmology” as an emergent phenomenon in all its differences. This is a “topological” rather than a “geometrical” view, a study of the properties that create forms rather than the studies of final generalized forms. I follow the “topological” approach since the static generalized models employed in Maya studies do not apply other than on a general level where differences are added to an already established pattern.
As an archaeologist and a self-proclaimed “neo-materialist” and “neo-realist” I therefore choose a perspective that sees linguistics as secondary. Manuel DeLanda argues that the linguistic turn in the social sciences has been predominant and this dominance has neglected other forms of information. If we are to be able to extract some kind of “cosmological information” from the archaeological record we should perhaps use Peircean semiotics instead of the Saussurean linguistic use of signs. In Peirce’s semiotics, which goes beyond linguistics, symbols are the least important of signs from an archaeological perspective. We should rather focus on Peirce’s index or icon.
Deleuze and Guattari argue that language is one of several regimes of signs. Regimes of signs emerge from assemblages and these are the emergent wholes of heterogeneous parts. Basically, this means that linguistic expressions are always part of a greater whole that includes haecceities like mountains, caves, lakes, rain clouds, institutions, economies, etc. Signs themselves are the emergent result of a flow. There is therefore a spatio-temporal flow that pre-exist the signs and other haecceities themselves. Abstract machines are what diverge and actualize this flow and hold the assemblages together. Cosmology as it usually is applied in Mayanist research belongs to a discursive/linguistic order and as such it does not penetrate these non-linguistic and non-representational understandings of the world.
The depiction of a cave or the word cave is not primarily a representation or sign of a cave “out there.” At best, the image or word points toward a cave and can be seen as an index of the cave. The depiction or the word cannot be something else than what they are and they are not the same as the “real” cave. Iconography and words make up their own realities. They form their own singular existences but as singular existences they also form an assemblage with the real cave. From this assemblage there might then emerge signs with “human meaning.” Thus, the human mind is not primarily concerned about representing a world “out there.” The mind is intertwined with assemblages. For example, Malafouris shows that in all likelihood representational thinking did not emerge before pictures emerged. We do not need representational thinking in order to depict. The sapien mind was already engaged in the material world before it could represent it. The mind is therefore distributed to materialities and images that archaeologists study and hence we can talk about meaning that is non-linguistic and partially located in materialities.
This is where a “deeper” understanding of the world exists, as something non-expressible in words, but this is still something real. Reality is chaosmos: Chaosmos is neither chaos (primordial disorder) nor cosmos (order imposed on primordial chaos from an outside/transcendent source). Chaosmos is an immanent process where order emerges from self-organization and disorder is the breakup of structure. It is a system at the edge of chaos, like the point where water either freezes to ice or turns into liquid. This is topological rather than geometrical. A chaosmological perspective differs from a cosmological in the sense that the latter focus on predefined and formed cosmograms and the former emphasizes the connections between properties that create an emergent form.