Concern over global environmental change and associated uncertainty has given rise to greater emphasis on fostering resilience through forest management. We examined the impact of standard silvicultural systems (including clearcutting, shelterwood, and selection) compared with unharvested controls on tree functional identity and functional diversity in three forest types distributed across the northeastern United States. Sites included the Argonne, Bartlett, and Penobscot Experimental Forests located in Wisconsin, New Hampshire, and Maine, respectively. We quantified functional trait means for leaf mass per area, specific gravity, maximum height, height achieved at 20 years, seed mass, drought tolerance, shade tolerance, and flood tolerance as well as standard functional diversity measures from standing biomass reconstructed at the beginning and end of a 20-year study period using increment cores and historic inventory data. As expected, functional identity differed between harvest methods with means for plant traits associated with later stages of succession (e.g. shade tolerance) increasing in stands managed with selection systems. Opposite trends occurred with greater canopy disturbance, and functional diversity indices remained stable over time in the absence of disturbance. Estimates of functional diversity and functional identity hold promise as important approaches for evaluating outcomes of forest management, particularly as the connections among functional diversity, delivery of ecosystem services, and ecosystem resilience are further developed.