Concord and Doubling Phenomena: an Introduction†

Formal semanticists have difficulty facing the fact that the law of double negation (DN) does not appear to hold in a vast majority of the languages of the world. In many languages, multiple negations may, semantically, yield one negation, and this is difficult to understand if each of the negations involves a semantic negation. One can assert, even in English, ‘If you ain’t got nothing to hide, then you don't need no privacy’—while meaning ‘If you got nothing to hide, you don't need privacy’. A primary school teacher could say that this is bad English, and that it reflects illogical thinking. However, for a lot of languages, and for a lot of speakers of American English, something like this is just the right way to say these things. So we have to live with the linguistic fact that, for instance, a DN may imply a single negation. Or, at least, that what seems to be a DN turns out to be a single negation after all. Once this fact is acknowledged, there appear to be many more similar facts about language. Multiple modality expressions can be seen to contribute to a single modality. Multiple number- or person-markers in a sentence contribute to one number- or person-related interpretation. And then the following questions readily pops up: are these all instances of the same phenomenon? And how should they be analyzed?