Lossy Encoding Across Mismatched Alphabets
When a rich sound system is written with an alphabet that marks fewer distinctions, transcription becomes lossy encoding: distinct sounds must share symbols, or spelling must improvise new signals. The resulting oddity belongs to the mapping between systems, not necessarily to the speaker.
Suppose a sound sits between what English readers recognize as d and r. The alphabet offers no faithful symbol, so the transcriber must pick one—and make half the audience think the result is wrong.
E1The bottleneck is the alphabet
Speech begins with distinctions made by the source sound system. Writing then passes them through a target alphabet whose available symbols divide sound differently. When the target has no one-to-one match, the encoder must collapse distinct sounds into one letter, choose the nearest misleading letter, or build a workaround such as a digraph. Information disappears or moves into conventions the reader must already know.
That is why unfamiliar spellings and pronunciations can look irrational from inside the target language. Readers decode them using their own letter-to-sound rules, although those letters were only approximations for another system.
E1Not every oddity is an encoding failure
This model applies when the source language makes a sound distinction the target alphabet cannot represent cleanly. It does not explain every irregular spelling or pronunciation, and with only the letter choice between d and r here, it cannot establish which wider conventions arose from the same mismatch.
E1Locate the missing distinction
When a borrowed name looks misspelled or sounds “mispronounced,” identify the original sound before correcting anyone. Compare how the two sound systems divide that phonetic space, then ask which distinction the written form was forced to discard—or whether a device like a borrowed phoneme or digraph preserves it elsewhere.
E1Episodes that teach this
-
Indian English is Weird & Britishers are to Blame - Schwa Deletion & Sound Pronunciations - FutureIQ
· explained at 6:57
257,831 views
"English is just... doesn't have enough letters to capture the sound, so you had to pick — you either pick a d or you pick an r — you had to make half the people unhappy anyway."