Philosophy and Psychology: Special issue on number and language
The question of how language and conceptual thought are related is unresolved in both philosophy and psychology. Many recent tests of the so-called 'Whorfian hypothesis', the idea that the structure of a particular language influences the way its speakers conceptualize the world, have focused on number. As has been noted earlier on this blog, the results of these investigations do not present a unified picture.
For example, some studies, such as Everett et al's investigation of Piraha number words, suggest a strong connection between the way the language allows for natural numbers discrimination and numerical cognition. On the other hand, a recent paper by Butterworth et al. in PNAS shows the opposite: in this study, Australian aboriginal monolingual children who only knew words for 'one, two many' performed as well as English speaking aboriginal children on classical tests of numerical cognition.
A special issue that recently appeared in Philosophical Psychology, edited by Pierre Pica and myself, contains several experimental and philosophical papers on this topic. Again, this special issue shows a mixture of papers that support or disconfirm the Whorfian hypothesis. For example, Dowker et al. show that children who speak Welsh or Tamil (two languages with very regular counting words) have an advantage in arithmetic compared to culturally similar control subjects who speak languages with irregular counting words (e.g., English). On the other hand, Pica and Lecomte argue that the limited counting abilities of the Munduruku may not be due to just an absence of counting words, but to the general structure of the Munduruku language (which has a very low compositionality power) and even to the way they conceptualize the human body. I therefore wonder whether the Whorfian hypothesis is something that can be empirically investigated, and whether we are asking the right questions.