We’ve all had the experience of listening to someone speak an unfamiliar language and perceiving their speech as being particularly rapid. It seems like a stereotype — depending on the language, that is — so we’re naturally skeptical. Maybe it’s all an illusion. I remember a professor of mine saying that the reason unfamiliar languages sound so rapid is that we don’t notice the spaces that aren’t there. The point, if that claim isn’t clear, is that in writing we put spaces between words but in speech one word runs seamlessly into the next.
But that doesn’t explain why some languages sound more rapid than others.
My first conscious experience with this phenomenon came when I was a teenager with a good language background but (oddly) no knowledge of Spanish. By that point I had learned a lot of Latin and Greek, some French, and a little German and Hebrew — but no Spanish. Of course my Latin and French gave me a good hook into figuring out a fair amount of Spanish, especially when I knew some of the patterns of sound relationships in Romance languages, but that worked only for writing: when listening to speech, how could I detect the spaces that aren’t there? How could I isolate distinct words? My perception, which might or might not have been real, was that Guatemalan speakers spoke significantly more slowly than Puerto Rican speakers, making it much easier for me to extract specific words from the former than the latter. But was this perception real?
An interesting short article about this topic was published recently in The Atlantic — interesting yes, but flawed and annoying as well. The article talks about the efficiency of different languages, where the efficiency concept comes from information theory: if language X conveys more bits of information per second than language Y, then it is deemed more efficient. And the conclusion is that although language X might have more phonemes per second, or more syllables per second, than language Y, both languages will still have equal efficiency. Then the article takes an odd excursion into whether there are linguistic universals and what the purpose of language is; those should be topics for another day. But nothing here addresses the question of whether my perception of the speeds of Guatemalan vs. Puerto Rican Spanish was real. Surely the information theory hypothesis isn’t explanatory here, since it’s unlikely that these two similar dialects would convey different amounts of information. What’s going on here?
And why was there no mention of sound spectrograms, which should clear up the basic question?