Relevance with empty domains. Submited.
Abstract: We propose a modification of Magri’s relevance procedure in blind scalar implicatures.
We argue that mismatching scalar implicatures involving contextually empty domains are evidence
for a Meinongian understanding of natural language quantification which allows computations with
merely subsisting entities. We discuss that a recent alternative account of such cases based on
presuppositional exhaustification is unable to explain the observations as it is bound to overgenerate
Abstract: Breaking with traditional antipragmatic assumptions, recent discussion inthis journal submits evidence for pragmatic inferences in mathematical language. The ob-servation is that the interpretation of crucial expressions, e.g., positive quantifiers, resultsfrom a truth-conditional enrichment of their literal meaning. While providing general in-sights into mathematical language, the authors unfortunately do not provide a clear andtestable linguistic analysis of the observed inference. However, in this contribution weprovide support to their informal proposal by presenting a formal reconstruction of the ob-served inference as a free choice implicature. In particular, we submit an algorithm basedon recursive exhaustification limited to domain alternatives that generates the desired read-ing under a natural assumption of homogeneity. Our argument is also based on indepen-dent and, to our knowledge, hitherto unnoticed evidence for free choice interpretations inmathematics