Drawing from the LEGA-C dataset, we present the spectroscopic view of the stellar population across a large volume- and mass-selected sample of galaxies at large lookback time. We measure the 4000\AA\ break (D$_n$4000) and Balmer absorption line strengths (probed by H$\delta$) from 1019 high-quality spectra of $z=0.6 - 1.0$ galaxies with $M_\ast = 2 \times 10^{10} M_\odot - 3 \times 10^{11} M_\odot$. Our analysis serves as a first illustration of the power of high-resolution, high-S/N continuum spectroscopy at intermediate redshifts as a qualitatively new tool to constrain galaxy formation models. The observed D$_n$4000-EW(H$\delta$) distribution of our sample overlaps with the distribution traced by present-day galaxies, but $z\sim 0.8$ galaxies populate that locus in a fundamentally different manner. While old galaxies dominate the present-day population at all stellar masses $> 2\times10^{10} M_\odot$, we see a bimodal D$_n$4000-EW(H$\delta$) distribution at $z\sim0.8$, implying a bimodal light-weighted age distribution. The light-weighted age depends strongly on stellar mass, with the most massive galaxies $>1\times10^{11}M_\odot$ being almost all older than 2 Gyr. At the same time we estimate that galaxies in this high mass range are only $\sim3$ Gyr younger than their $z\sim0.1$ counterparts, at odd with pure passive evolution given a difference in lookback time of $>5$ Gyr; younger galaxies must grow to $>10^{11}M_\odot$ in the meantime, and/or small amounts of young stars must keep the light-weighted ages young. Star-forming galaxies at $z\sim0.8$ have stronger H$\delta$ absorption than present-day galaxies with the same D$_n$4000, implying larger short-term variations in star-formation activity.
This is an author-created, un-copyedited version of an article accepted for publication/published in The Astrophysical Journal. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at http://iopscience.iop.org/article/10.3847/1538-4357/aab0a6