Abstract
In this chapter I provide empirical and theoretical considerations in favor of a non-inferential view of speech comprehension. On the view defended, we typically comprehend speech by perceiving or grasping apparently conveyed meanings directly rather than by inferring them from, say, linguistic principles and perceived phonemes. ‘Speech’ is here used in the broad sense to refer to not only verbal expression but also written messages, including Braille, and conventional signs and symbols, like emojis, a stop sign, or a swastika. Along the way I define what I mean by ‘inference’ and provide an account of what it means to say that we perceive apparently conveyed meanings.