To evaluate the biomechanical properties of the upper thoracic spine in anterior–posterior shear loading at various displacement rates. These data broaden our understanding of thoracic spine biomechanics and inform efforts to model the spine and spinal cord injuries. Seven T1–T2 thoracic functional spinal units were loaded non‐destructively by a pure shear force up to 200 N, starting from a neutral posture. Tests were run in both posterior and anterior directions, at displacement rates of 1, 10, and 100 mm/s. The three‐dimensional motion of the specimen was recorded at 1000 Hz. Individual and averaged load–displacement curves were generated and specimen stiffnesses were calculated. Due to a nonlinear response of the specimens, stiffness was defined separately for both the lower half and the upper half of the specimen range of motion. Specimens were significantly stiffer in the anterior direction than in the posterior direction, across all rates. At low displacements, the anterior stiffness averaged 230 N/mm, 76% higher than the low displacement posterior stiffness of 131 N/mm. At high displacements, anterior stiffness averaged 258 N/mm, 51% stiffer than the high displacement posterior stiffness of 171 N/mm. Shear displacement rate had a small effect on the load response, with the 100 mm/s rate causing a mildly stiffer response at low displacements in the anterior direction. Overall, the load–displacement response exhibited pseudo‐quadratic behavior at 1 and 10 mm/s but became more linear at 100 mm/s. The shear stiffness in the upper thoracic spine is greatest in the anterior loading direction, being 51%–76% greater than posterior, most likely due to facet interactions. The effect of the shear displacement rate is low.