We present the results of a study on the properties and evolution of massive (M* > 1011Msolar) galaxies at z ~ 0.4-2 utilizing Keck spectroscopy, near-infrared Palomar imaging, and Hubble, Chandra and Spitzer data covering fields targeted by the DEEP2 galaxy spectroscopic survey. Our sample is K-band selected and stellar mass limited, based on wide-area near-infrared imaging from the Palomar Observatory Wide-Field Infrared Survey, which covers 1.53 deg2 to a 5σ depth of Ks,vega ~ 20.5. Our primary goal is to obtain a broad census of massive galaxies through measuring how their number and mass densities, morphology, as well as their star formation and active galactic nucleus content evolve from z ~ 0.4-2. Our major findings include: (i) statistically the mass and number densities of M* > 1011Msolar galaxies show little evolution between z = 0 and 1 and from z ~ 0 to 2 for M* > 1011.5Msolar galaxies. We however find significant evolution within 1 < z < 1.5 for 1011 Msolar < M* < 1011.5Msolar galaxies. (ii) After examining the structures of our galaxies using Hubble ACS imaging, we find that M* > 1011Msolar selected galaxies show a nearly constant elliptical fraction of ~70-90 per cent at all redshifts. The remaining objects tend to be peculiars possibly undergoing mergers at z > 0.8, while spirals dominate the remainder at lower redshifts. A significant fraction (~25 per cent) of these early-types contain minor structural anomalies. (iii) We find that only a fraction (~60 per cent) of massive galaxies with M* > 1011Msolar are on the red sequence at z ~ 1.4, while nearly 100 per cent evolve on to it by z ~ 0.4. (iv) By utilizing Spitzer MIPS imaging and [OII] line fluxes we argue that M* > 1011.5Msolar galaxies have a steeply declining star formation rate (SFR) density ~ (1 + z)6. By examining the contribution of star formation to the evolution of the mass function, as well as the merger history through the CAS parameters, we determine that M* > 1011Msolar galaxies undergo on average 0.9+0.7-0.5 major mergers at 0.4 < z < 1.4. (v) We find that a high (5 per cent) fraction of all M* > 1011Msolar galaxies are X-ray emitters. Roughly half of these are morphologically distorted ellipticals or peculiars. Finally, we compare our mass growth with semi-analytical models from the Millennium Simulation, finding relative good agreement at z < 2 for the M* < 1011.5Msolar systems, but that the number and mass densities of M* > 1011.5Msolar galaxies are underpredicted by a factor of >100.