Extensive neuron loss occurs in Alzheimer disease (AD) brain and some authors have speculated that dysregulation of apoptotic death pathways is etiologically responsible for the disease. Apoptosis is regulated in mammalian cells by a family of cysteine proteases called caspases. At least 7 different caspases (caspases 1, 2, 3, 6, 8, 9, and 12) have been implicated in regulating neuronal cell death in response to amyloid beta (A beta) exposure in vitro, in animal models of neurodegenerative diseases, and in AD brain itself. Despite this seemingly impressive array of data implicating caspases and apoptosis as etiologic factors in AD, the direct involvement of caspase-dependent neuronal apoptosis in AD pathogenesis remains uncertain. Alternative explanations for some findings, contradictory experimental observations, and lack of morphologically convincing apoptotic neurons in the vast majority of AD brains has led to the revised hypothesis that apoptosis-associated molecular events cause neuronal dysfunction in the absence of, or prior to, neuronal death. Unfortunately, this new view renders the term "apoptosis-associated" functionally meaningless since it bears no relationship with apoptotic death and fails to focus scientific investigation on the molecular insults that trigger the "apoptosis-associated" response in AD neurons. On balance, an etiologic role for caspases in AD is far from proven. It remains possible, however, that caspase-dependent neuronal death contributes to AD neuron loss and thus, caspase inhibition offers some hope for extending AD neuron survival so that other agents, targeting upstream events, may delay or reverse primary AD pathology.