It is a well-known fact that the verb dare can be used with either modal or main verb characteristics both in its inflection and in its syntax. When used as a modal, it drops the -s ending in the third person singular present indicative (She dare not mention it in his presence), has no imperative, infinitive or participial forms, takes direct negation by not, AUX-inverts in questions (Dare I ask you another question?) and is followed by the bare infinitive. In main verb use, on the other hand, it has all the normal forms of the verb, occurs with do auxiliary in negatives and interrogatives, and is construed with the to infinitive (She doesn’t dare to mention it in his presence).

The motivations of sound change are almost exclusively treated as articulatory phenomena, even when the source and the target sounds are generated by very different articulatory mechanisms and no transitional stages can be found. What will be argued here is that seemingly regular place shifts in the articulation of consonants, such as the ones known under the term of “Rhenish Velarization” or “Rhenish Gutturalization” are actually manifestations of acoustically induced variants, very similar to those known as “slips of the ear”, which differ from one another only by the presence or absence of a single acoustic feature. The data assembled from a larger Rhenish corpus show that the operative process entails variations between labial, dental and velar place of articulation and involves stops, fricatives and nasals; its directionality is basically undetermined. Data from other sources and languages suggest that such place variations may be much more general than previously thought.