Hacker News new | past | comments | ask | show | jobs | submit login

The phoneme-grapheme correspondence in Spanish is better than English, but let's not pretend it is 1:1. Does it account for assimilation in rapid speech? Does it account for coarticulation of adjacent consonants? Does it account for regional/dialectal variation? Does it account for secondary articulation?

Even ignoring all of these, its clearly not bijective. For example:

C --> /k/, /θ/

Z --> /θ/ [0]

K --> /k/

Q --> /k/

G --> /ɡ/, /x/

J --> /x/

N --> /n/ (with several distinct secondary articulations), /m/ (rarely)

M --> /m/

R --> Can be tapped or trilled.

Etc. You can go here and see many bijection-failures here: [1]

I am being intentionally unfair to Spanish (which truly does have a much, much better phoneme-grapheme correspondence than English[2]), mostly to illustrate the point that there aren't really any languages which have a 1:1 mapping between spellings and pronunciations. Even if you decide to use the IPA to write your language, non-standard dialects end up needing to read words that don't match their pronunciations. What happens when inevitably the language undergoes change - do we update all of the books to use the 'new' spellings of words?

The ideal orthography shouldn't be completely 1:1, but it should be relatively shallow. From that perspective, Spanish orthography is a fairly attractive option.

[0] The non-1:1 situation with /θ/ gets much worse in most dialects of Spanish, where it is not distinguished from /s/. See: https://en.wikipedia.org/wiki/Phonological_history_of_Spanis...

[1] https://en.wikipedia.org/wiki/Spanish_orthography#Alphabet_i...

[2] Look at how effective Spanish-speakers are at reading without "decoding" compared with Portuguese, which also has a good p-g correspondence. In particular, look how much faster the Spanish students are at pseudowords, on page 141: https://www.academia.edu/17872463/Differences_in_reading_acq...




Most of these examples are unambiguous in context. For example, C is always pronounced /k/ when preceding A, O, U, and always pronounced /s/ or /θ/ depending on the dialect when preceding E and I.

The function from written Spanish to spoken Spanish (provided we are talking about a single dialect) is surjective, but darn close to bijective, especially if we exclude words of recent foreign origin.


From Mark Rosenfelder (https://www.zompist.com/spell.html):

> Many people expect … to predict the spelling from the pronunciations-- not realizing that few orthographies meet this goal. It's far from true of Spanish, for instance, which is often held up as an example of a good orthography. I stopped fervently admiring Spanish orthography when I saw a sign in a Mexican bakery with about one spelling mistake every third word.

So, no, hardly bijective!


Well, rules can be simple and there can be few exceptions and people will still screw up, so I don't think that anecdote proves anything. In any case, my claim is that there is exactly one possible pronunciation per correctly spelled Spanish word. The opposite direction is not quite 1:1, but again, it's very close, and anyway it's far closer than English.


Sure, that I can agree with.


Laughs in chilean or andalucian rap god dialect




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: