The 1897 Shillong (Assam), northeast India, earthquake is considered to be one of the largest in the modern history. Although Oldham's <1899> classic memoir on this event opened new vistas in observational seismology, many questions on its style of faulting remain unresolved. Most previous studies considered this as a detachment earthquake that occurred on a gently north dipping fault, extending from the Himalayan front. A recent model proposed an alternate geometry governed by high-angle faults to the north and south of the plateau, and it suggested that the 1897 earthquake occurred on a south dipping reverse fault, coinciding with the northern plateau margin. In this paper, we explore the available database, together with the coseismic observations from the region, to further understand the nature of faulting. The geophysical and geological data examined in this paper conform to a south dipping structure, but its location is inferred to be in the Brahmaputra basin, further north of the present plateau front. Our analyses of paleoseismic data suggest a 1200-year interval between the 1897 event and its predecessor, and we identify the northern boundary fault as a major seismic source. The Shillong Plateau bounded by major faults behaves as an independent tectonic entity, with its own style of faulting, seismic productivity, and hazard potential, distinct from the Himalayan thrust front, a point that provides fresh insight into the regional geodynamics.