Morse code signals were presented dichotically to Morse code operators and to subjects who did not know Morse code. Morse code operators showed right ear superiority, indicating left hemisphere dominance, for the perception of dichotically presented Morse code letters. Naive subjects (i.e., subjects who did not know Morse code) showed right ear superiority, indicating left hemisphere dominance, (n.b.: same as Morse code operators) when presented with a set of dot‐dash patterns which was restricted to pairs including seven or fewer elements, counting dots and dashes each as elements. The list restricted to pairs with seven or fewer elements in all was presented to the naive subjects at two levels of intensity, but even though the overall accuracy changed, the direction and degree of asymmetry as measured by the left ear contribution to the total errors remained constant. When presented with a list that included longer stimuli, naive subjects showed left ear superiority, indicating right hemisphere dominance, the opposite of their result with the shorter stimuli. We hypothesize that pairs consisting of the Magical Number Seven or fewer elements are perceived with reference to the subparts of which they are composed, but that longer stimuli force naive subjects to adopt strategies involving the holistic qualifies of the stimuli. Therefore we speculate that the left hemisphere is specialized for processing the sequential parts of which a stimulus is composed. Analysis of the substitutions made by Morse code operators does not reveal a phonetic basis for the lateralization of the stimuli as perceived by them. Consideration of our findings in the light of other literature on lateralization suggests that language is lateralized to the left hemisphere because of its dependence on segmental subparts, and that this dependence characterizes language perception as distinct from most other human perception.