We performed an in vitro study to investigate the stabilization (i.e. motion reduction) provided by the external spinal fixator (ESF), and to compare the three configurations of the ESF with two internal fixation techniques. Six human cadaveric lumbar spine specimens (L3-S1) were subjected to multidirectional flexibility testing in six configurations: (1) intact, (2) ESF in neutral, (3) ESF in distraction, (4) ESF in compression, (5) translaminar facet screw fixation, and (6) internal transpedicular fixation. Both the ESF and the internal fixation systems stabilized the specimens from L4 to S1. In each testing configuration, pure bending moments of flexion-extension, bilateral axial rotation, and bilateral lateral bending were applied to the uppermost vertebra stepwise to a maximum of 10 Nm. The rigid body motion between the vertebrae was measured using an optoelectronic camera system, and custom software was used to calculate the intervertebral rotations. For each applied motion in all testing configurations, the total range of motion (ROM) of L4-S1 is reported. All three ESF configurations stabilized the spine significantly when compared to the intact specimen. The ESF in compression provided significantly more stabilization in flexion-extension than the two other ESF configurations, but no other significant differences were found between the three ESF modes. In flexion-extension the ESF stabilized the spine significantly when compared with the two internal fixation devices. Only in bilateral lateral bending was the ESF inferior to internal transpedicular fixation in providing stabilization. The results of the present study suggest that the ESF provides a high degree of stabilization for preoperative assessment of selected low back pain patients. Whether other non-mechanical factors affect the pain relief experienced by the patients remains unknown.