'Articleless' languages are not created equal

We adopt a translation corpus approach based on the first chapter of 'Harry Potter and the Philosopher’s Stone' to evaluate Dayal’s updated version of the neo-Carlsonian framework and the predictions it makes for bare nouns in Hindi, Russian and Mandarin (Dayal 2004). Our Hindi data turn out to be overall in line with Dayal’s predictions but the same does not hold for our Russian and Mandarin data, leading us to explore a number of extensions and modifications of Dayal’s analysis. For Mandarin, our data lead us to hypothesize a role for the numeral 'yi' (‘one’) as an indefinite article and for demonstratives as definite articles. For our Hindi and Russian bare noun data, we argue that the only way to account for them is to reverse at least part of Dayal’s updates to the neo-Carlsonian framework and to hypothesize that Hindi – unlike Russian – is developing an indefinite article.