Paralysis occurring in amyotrophic lateral sclerosis (ALS) results from denervation of skeletal muscle as a consequence of motor neuron degeneration. Interactions between motor neurons and glia contribute to motor neuron loss, but the spatiotemporal ordering of molecular events that drive these processes in intact spinal tissue remains poorly understood. Here, we use a spatially resolved view of disease-driven gene expression changes to stratify these events, reveal the relevant sub-populations of cells involved in each stage of disease progression, and characterize the underlying molecular mechanisms that trigger and drive the course of disease. Based on the well characterized cellular organization of the spinal cord and the importance of intercellular interactions in ALS disease progression, we applied spatial transcriptomics (ST) to obtain spatially and anatomically resolved quantitative gene expression measurements of mouse spinal cords over the course of disease, as well as in postmortem tissue from ALS patients. We developed a novel Bayesian generative model for assembling a spatiotemporal atlas of gene expression in ALS that integrates cell-type, anatomical region, space, and time. We identify novel pathways implicated in ALS progression, key differences between microglia and astrocyte populations at early time-points and in different anatomical regions, and discern several transcriptional pathways shared between murine models of ALS and human postmortem spinal cords. We provide a general experimental-computational design for mapping and understanding the transcriptional landscape of diseases in complex tissues. An interactive data exploration portal for our ST analysis is available at als-st.nygenome.org.