This week’s syntax-semantics reading group meeting will take place Friday, February 5th at 2:30pm. Bernhard Schwarz will be presenting his recent work with Alan Bale (Concordia) and David Shanks (McGill), ‘Measurement monotonicity and comparisons of concentration’. A precursor of this work can be found at: https://archive.illc.uva.nl/AC/AC2019/uploaded_files/inlineitem/Bale_Schwarz_and_Shanks_Monotonicity_restored_more_.pdf
Abstract: Grammar does not fully determine a measure function for much to invoke when combining with a mass noun, as in more watermore string, or more travel. But certain measure functions are invariably excluded. Schwarzschild (2006) argued that permissible measure functions must be monotonic—they must track the relevant parthood relation between entities in the mass noun’s extension. We draw attention to potential counterexamples to this constraint: comparisons of concentration, comparatives about the concentration of substances in substances that contain them. Such comparatives seemingly reference measure functions that are non-monotonic, as they seem to output degrees of purity or density (Bale and Barner 2009). Extending Bale and Schwarz (2020), however, we argue that they must not in fact be so analyzed. What such cases actually teach us, we propose, is that much can reference Skolemized measure functions—functions that map any individual, their Skolem argument, to a measure function.  Comparisons of concentration can arise when different values serve as Skolem arguments in the main clause and in the comparative clause. Schwarzschild’s constraint, now extended to also apply to the outputs of Skolemized measure functions, can and must be upheld.