All animals are model builders, but human beings are able to model possible worlds, often including the very worlds we have created by modeling. We do so, on many accounts, by blending select category-member relationships that we perceive to hold (iconically and indexically) between sets of phenomena that are otherwise distinct. This ability is identified, under different titles, by theorists across a wide spectrum of disciplines and persuasions, as the defining characteristic of human uniqueness, as a salient source of the language faculty, and as the only serious candidate for universal grammar (Sebeok 1991, Fauconnier & Turner 2002, Anttila 2003, Deely 2010, Bybee 2010, Hofstadter & Sander 2013, Harari 2014, Tattersall 2016, Pelkey 2017b). Since the ability is reflexive, it enables us to step back and investigate the nature of the ability itself—including its origins, mechanisms, limitations, liabilities, potentialities and responsibilities. In this paper I focus on the first two of these reflexive concerns—i.e., origins and mechanisms—identifying them as priorities for cognitive semiotics and proposing a grammar of kinesthetic patterning as a tentative solution and prospective research program. Little attention has been devoted to explaining the origins and mechanisms of this creative, reflexive modeling ability in ways that are both genuinely evolutionary and also grounded in embodied interaction. Both concerns are priorities for cognitive semiotics; so I propose in this paper that cognitive semiotics should prioritize inquiry into both concerns. I also propose that a pattern-grammar of embodied movement could have played a crucial role. Drawing on multimodal evidence from languages, cultures and theorists around the world, the paper identifies twelve potential sources of kinesthetic pattern grammar in three sets of four experiential clusters all grounded in the reorganization of the anatomical planes in the development of habitual upright posture. Together these experiential layers provide an evolutionary account that is sufficiently continuous to be plausible and yet systematically robust enough to account for multiple mechanisms of our creative, reflexive modeling abilities. Such hypotheses are in need of careful consideration, elaboration, development and testing within cognitive semiotics. Note: For inclusion in the theme session on Cognitive Semiotics References Anttila, R. (2003). Analogy: the warp and woof of cognition. In B. D. Joseph & R. D. Janda (Eds.), The Handbook of Historical Linguistics (pp. 425–440). London: Blackwell. Bybee, J. (2010). Language, Usage and Cognition. Cambridge: Cambridge University Press. Deely, J. (2010). Semiotic animal: A postmodern definition of human being transcending patriarchy and feminism. South Bend, IN: St. Augustine’s Press. Fauconnier, G., & Turner, M. (2002). The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities. New York: Basic Books. Harari, Y. N. 2014. Sapiens. London: Harvill Secker. Hofstadter, D., & Sander, E. (2013). Surfaces and Essences: Analogy as the Fuel and Fire of Thinking. New York: Basic Books. Pelkey, J. (2017b). The Semiotics of X: Chiasmus, Cognition, and Extreme Body Memory. London: Bloomsbury Academic. Sebeok, T. A. (1991). In what sense is language a primary modeling system? In T. A. Sebeok (Ed.), A sign is just a sign (pp. 49–58). Bloomington: Indiana University Press. Tattersall, I. (2016). The thinking primate: Establishing a context for the emergence of modern human Cognition. Proceedings of the American Philosophical Society 160.3, 254-265.