Is prediction necessary to understand language? Probably not

Presents a critical review of arguments in favor of and against the view that prediction is necessary for understanding language. First, potential arguments in favor of the view that prediction provides a unified theoretical principle of the human mind and that it pervades cortical function are reviewed. It is discussed whether evidence of human abilities to detect statistical regularities is necessarily evidence for predictive processing, and suggestions that prediction is necessary for language learning are evaluated. Next, arguments in support of the contrasting viewpoint are reviewed: that prediction lends a ``helping hand'', but is not strictly needed for language processing. It is pointed out that not all language users appear to predict language and that suboptimal input makes prediction often very challenging. Prediction, moreover, is argued to be strongly context-dependent and impeded by resource limitations. Furthermore, it is argued that it may be problematic that most experimental evidence for predictive language processing comes from prediction-encouraging experimental set-ups. It is concluded that languages can be learned and understood in the absence of prediction. Claims that all language processing is predictive in nature are considered to be premature.