The South American Monsoon System (SAMS) is the dominant mode of hydroclimatic variability in South America and the associated dynamics have an important influence on atmospheric organization during pluvial and drought periods. Paleoclimate records from the monsoon domain have helped to describe the isotope hydrology of this region over the last millennium, but are insufficient to explain why the system has changed. Utilizing transient simulations of last millennium climate from isotope-enabled global climate models as well as experiments forced by isolated external forcings made available from the NCAR Community Earth System Model Last Millennium Ensemble (CESM-LME) project, we explore the dynamical drivers of variability documented in the isotopic record. We find that while models continue to underestimate the temporal variability of regional climate, they are able to reproduce documented spatial modes of variability. A Monte Carlo empirical orthogonal function (MCEOF) analysis is used to decompose major modes of isotope variability from both isotope proxy records and isotope-enabled global climate models over the last millennium (850 – 1850 CE). The principal component time series from the isotopic records show clear mean state departures during the early and late part of the record, corresponding to the Medieval Climate Anomaly and Little Ice Age periods. We explore the role of internal variability and different external forcings in driving these departures, and use the model space to describe changes to regional circulation patterns during these periods of extreme isotopic change. This enhanced understanding of the drivers of variability over the last millennium could provide a foundation upon which to interpret the sensitivity of this system to future changes due to external forcings, such as anthropogenic activity.