One accelerated development of the advent of the digital age as far as language is concerned is the diminishing role of orality in the transmission of linguistic norms. This was observed the other day when a young woman writer (no doubt, a member of the so-called “millennial generation”) was being interviewed on NPR about an article she had written for the National Geographic concerning the situation in present-day Syria. In describing life in Damascus, she mispronounced the words Damascene (adjective < Damascus) and sepia, rendering the penultimate consonant (for the second sound of the digraph –sc-) of the first word as [k] instead of the correct [s], and the initial stressed vowel of the second word as [e] instead of [i].
This sort of error arises because the speaker has obviously never been exposed to the words’ correct pronunciation. There are simply no oral milieux in which a youngish speaker of American English––doubtless, college-educated withal––can hear words such as the two at issue pronounced correctly. Knowledge of such vocabulary items now tends to come about solely from an acquaintance with them in written form, where the ambiguity of their orthographic representation gives rise to a phonetic choice that is exercised without benefit of an oral precedent from an authoritative source––hence incorrectly, as often as not.