Stellar populations and dynamics of nearby early-type galaxies

M.A.K. Halliday's metafunctional functionalism, which forms the basis for the semantics of his grammar, is based upon a revision of Bühler's triadic functions of language. However, the more philosophical, classical American branch of the pragmatic tradition developed a likewise triadic understanding of experience, which also differs from Bühler's. Whitehead's pragmatic metaphysics provides the most comprehensive example of such a non-reductive philosophical system, crowned by an aesthetically-based general theory of value. In this thesis, I use stellar population and dynamical modelling techniques to study the star formation histories and mass distributions of early-type galaxies. First, I make a detailed comparison of four commonly used stellar population synthesis codes, which have been shown to give discrepant results when applied to near-infrared data. This has commonly been attributed to different prescriptions for the thermally-pulsing asymptotic giant branch phase in the various models. In this work, however, I show that the discrepancies extend beyond this: applying the four models to high quality optical and near-infrared spectroscopy of 12 fast-rotating early-type galaxies, I find the largest differences between models is due instead to the choice of stellar spectral library. Models including updated, high resolution stellar libraries are the most self-consistent when comparing optically derived properties with near-infrared ones. Second, I study the effects of incorporating stellar population information into dynamical models, using high signal-to-noise MUSE spectroscopy of three fast-rotating early type galaxies. Specifically, I include spatial variation in the stellar mass to-light ratio (M=L) due to gradients in age, metallicity and stellar initial mass function (IMF). I show that age and metallicity gradients have little effect on dynamically derived quantities, however inclusion of IMF gradients of the strength observed by van Dokkum et al. (2017) increase the derived dark matter fraction within an effective radius by 50%. Finally, I perform a completely self-consistent stellar population and dynamical modelling analysis using high quality MUSE data. I measure gradients in age, metallicity and IMF directly from the same data from which I extract kinematics. I find similar M=L gradients to other stellar population-based IMF studies, however dynamical models which include these measured M=L values cannot reproduce the observed kinematics. If the stellar mass is allowed to be rescaled, it is consistently scaled downward, and the fits to the kinematics become acceptable, indicating that the M=L gradient is not incompatible with the data, however the absolute normalisation of the M=L from the stellar population models is higher than allowed by the kinematics. Overall, this thesis highlights the importance of selfconsistent studies, which accurately account for systematic errors with a thorough exploration of the techniques and models used.