Biologists and the public increasingly appreciate the diverse and pervasive effects of humans on natural systems, but data are difficult to acquire for more than a few select species and over the past few decades, centuries, and millennia needed to recognize change, discriminate between natural and human drivers, and establish natural baseline conditions, all critical to ecological assessment and restoration. Death assemblages — the actively accumulating organic remains encountered in present-day seabeds and landscapes, as distinct from permanently buried fossil assemblages — are an underexploited source of ecological historical information at precisely these scales. Field work, experiments, and statistical exploration of ‘live-dead agreement’ in modern environments, originally motivated to better understand the formation of ancient fossil assemblages, reveal that dead shell and bone assemblages differ from the local living community primarily because they are temporally coarse, time-averaged samples, contrary to concerns that postmortem bias dominates. Time-averaging damps the ability of the dead to detect small-scale variability, but promotes their ability to inventory regional diversity, evaluate historic habitat use, and identify now-absent species, community states, and anthropogenically shifted baselines.