We generated low-temperature thermochronological data on crystalline rocks from the Chiapas Massif in southern Mexico to constrain the complex relationship among tectonics, exhumation, and sedimentation in the region. Our data show that the first recorded cooling event occurred at ca. 40–25 Ma due to denudation of the sedimentary cover of the Chiapas Massif at slow rates of ~0.1 km/m.y. This was followed by a period of tectonic quiescence from ca. 25 to 14 Ma. Between ca. 14 and 7 Ma, cooling implying exhumation of the massif at rates of up to ~0.7 km/m.y. was renewed, and this was associated with, and possibly responsible for, the Miocene “Chiapanecan” deformational event observed in the Chiapas fold-and-thrust belt to the northeast of the massif. This younger uplift was also accompanied by the onset of arc-related magmatism beneath the massif, between ca. 13 and 9 Ma, along the Tonalá shear zone at the Pacific coast. Since ca. 7 Ma, additional but slower cooling and exhumation are indicated along the length of the Chiapas Massif, and arc magmatism has jumped north by ~125 km from the Tonalá shear zone into the Chiapas fold-and-thrust belt. Concurrently, subsidence and sedimentation have persisted along the offshore Tehuantepec Shelf to the south, suggesting that the Tonalá shear zone has been recently active (despite no magnitude 4 or larger earthquakes), with up-to-the-north vertical displacement. We interpret the exhumation at ca. 40–25 Ma to pertain to displacement of the Chortis block along the paleo–Motagua fault zone, either as a northward propagation of a basement thrust beneath the massif within a regional transpressional setting, or as a deep, ductile crustal thickening and attendant isostatic uplift of the southern flank of the massif during the transpressional passage of the Chortis block. The ensuing quiescence (25–14 Ma) coincided, we believe, with the passage of the “western tail” of Chortis, which is internally deformed and perhaps transferred compressive stress less effectively than had the central, continental core of the Chortis block earlier. Renewed uplift and exhumation of the region began by ca. 14–10 Ma. An onset at ca. 10 Ma is probably the best estimate for the beginning of exhumation of the northwestern and central portions of the Chiapas Massif, whereas the present-day southeastern tip of the massif (potentially an allochthonous sliver belonging to the Chortis block) started to exhume earlier, at ca. 14 Ma. By ca. 13 Ma, arc magmatism had moved into the western Tehuantepec area, marking the onset of subduction of the Cocos plate beneath the Chiapas Massif. Hence, we interpret the main period of uplift of the Chiapas Massif and primary shortening of the Chiapas fold-and-thrust belt (ca. 14–7 Ma) as being driven by the establishment of Cocos subduction beneath the area.