Abstract Modal sense classification (MSC) is aspecial WSD task that depends on themeaning of the proposition in the modal’s scope. We explore a CNN architecture for classifying modal sense in English and German. We show that CNNs are superior to manually designed feature-based classifiers and a standard NN classifier. We analyze the feature maps learned by the CNN and identify known and previously unattested linguistic features. We bench-mark the CNN on a standard WSD task,where it compares favorably to models using sense-disambiguated target vectors. (Marasović and Frank, 2016)