[OT] musings about orthography (was: Re: [Ffmpeg-devel] Re: Semantic Error?)

Diego Biurrun diego
Wed Feb 1 00:21:04 CET 2006


On Tue, Jan 31, 2006 at 01:28:21PM +0100, Erik Slagter wrote:
> 
> Anyway, I think spelling of English (either American or British) is
> ehrm... suboptimal anyway. The relation between how a word is written
> and how it is pronounced is very ehrm... weak, which makes it very
> difficult for non-native speakers to learn the language.

This has nothing to do with how difficult it is to learn the language.
English grammar is - compared to most other languages - trivial, which
makes the language easy to learn.

English orthography is mostly random, but curiously this is less of a
problem for non-native speakers since vocabulary is something that you
have to learn by heart anyway and people learn the spoken and written
forms of the words simultaneously.  This leads to the peculiar situation
that native English speakers - knowing the sounds - try to map sounds to
letters in a meaningful way and fail because this strategy does not get
you very far with English.

Here is a small table comparing the relationship between pronounciation
and writing for a few languages where I'm familiar with the orthography:

English:	arbitrary
French:		arbitrary
German:		medium
Spanish:	strong
Turkish:	1:1

Germans often (mistakenly) think that German spelling is consistent,
probably because they compare with French or English, which are even
worse.  Deciding which of the two is worst is pointless.

German needs multi-letter combinations for a few sounds (sch, ch, ck),
some combinations of consonants sound different depending on their
position in the word (sp, st, ch), some vowel combinations produce
illogical sounds (ei, eu), some letters represent different sounds (s
like in "so" or in "Gas", e like in "Regel" or in "Bitte"), vowels can
be long or short, sometimes they get doubled, sometimes they get an h
appended, sometimes neither.

Spanish has a few idiosyncracies regarding the letters g/j and c/qu/k,
but apart from that it's remarkably consistent.

Turkish is wonderful.  Each sound is mapped to exactly one letter, no
exceptions.  They do have a g with a half circle above it that lengthens
the vowel before it, but it's a really small complication.

OK, back to multimedia programming :-)

Diego





More information about the ffmpeg-devel mailing list