Re: Is "wa" a punctuation mark?

On 09/05/2009 06:52 PM, Sean wrote:

The Wanderer wrote:

On 09/05/2009 12:45 PM, Sean wrote:

When you are speaking Japanese, are you "pronouncing kana"? Of
course not. You seem to be speaking from the linguistically naive
position that a language is its written form, or that the written
form is somehow the basis.

I don't think I am, no. I think I'm simply focusing on the written
language because that's what is relevant to romaji vs. kana.

That would make you a very strange human. In normal humans, the
speech motor areas of the brain are active while they are reading (as
they are when listening to speech). Consider that written language
hasn't been around long enough for us to have evolved special brain
functions or brain parts for reading. The process has to make use of
existing functions, and that happens to be speech. No big surprise
since, for the most part, writing represents speech.

None of that is particularly surprising or strange to me. I'm not sure
(or, at least, not sure how to put into words) what you're expecting to
convey by this.

(It just occurred to me that I may have been unclear. By "I don't think
I am", I meant that I don't think I'm speaking from the position you
described, not that I don't think I'm pronouncing kana when I'm speaking
Japanese.)

When someone is reading, it doesn't matter whether they internally
pronounce the は as "ha" or as "wa", so long as they know what it
means.

What a horrible thought. Well, if it is just a meaningful graphic
symbol with no necessary connection to speech, then why not just
represent it with a picture of an avocado?

Any of a number of reasons, not least the fact that it's harder to draw
a recognizable avocado than it is to write an already-familiar letter.
(Even harder on a computer, if the picture isn't already available, or
if it can't be readily put into the text in some particular case.)

You're also taking this farther than I intended it... as I said, there
*is* a connection to speech, it just doesn't have the effect you seem to
say it does. Yes, the reader does (and probably should) mentally
pronounce the kana, or the romaji; however, for the purposes of the
reading it doesn't matter *what* they pronounce it as, as long as they
understand whatever pronunciation they do use (and the symbol which
represents it) as matching the appropriate meaning. It's only when they
go beyond that to actually speak based on what they read that the
pronunciation becomes relevant.

If it is written as "wa", it will be harder for them to tell what
it means - or, if it is not harder, then they will essentially be
learning to treat "wa" as the topic-marker particle, which is not
correct. On the other hand, if it's written as "ha", they can
correctly recognize it as the topic-marker particle and understand
the meaning - and until the point where they read it aloud is
reached, it makes no difference what their internal pronunciation
is.

Sorry, but this is just nonsense to me. When I *hear* someone say
"Boku wa nihonjin da" I know what they mean. When I *read* it, the
same thing applies. You claim somewhere that most readers in this ng
are beyond the beginner level. Do you seriously think they are going
to have trouble with "wa" in the context of a sentence?

Depends what you mean by "trouble". In the sense I think you mean it,
no, I don't. I don't think that's relevant, however, since this is about
a basic principle.

Gee, when people speak to me I always understand the meaning
because of the sounds they are uttering.

But we're not talking about when people are speaking - we're
talking about when people are reading and writing.

Neurologically very closely related activities.

True enough. I'm not sure it's close enough to be necessarily relevant
here,though.

It's easier to parse the misleadingly-written "wa" in that case
than it is in some others, because of the spaces around it. Those
aren't always present. When parsing romaji without spaces -
something I've needed to do on multiple occasions, because the
romaji was taken down as an exact "don't attempt to parse this,
just transcribe it" copy of the original Japanese, and/or was
machine-converted from a kana/kanji original - the additional
information provided by the "ha" instead of the "wa" is somewhere
in the range from useful to essential.

I have never had the need to read long passages of romaji Japanese.
If presented with the opportunity, I'd probably turn it down.

That might explain some of your position on this, then.

It's much more common to encounter romaji Japanese in the form of
single lexical items, phrases, or single example sentences.

In your experience, perhaps - but not in mine.

In those contexts, I get much more of a reading hiccup from topic
marker は written as "ha" than I ever would from seeing it written as
"wa."

This is something which puzzles me.

Why exactly do you get a 'hiccup' from that?

If it's because the "natural" pronunciation of that spelling doesn't
match the way that syllable would actually be pronounced when spoken,
then I would expect that you would get exactly the same hiccup when
reading the kana, because は is normally pronounced "ha" instead of "wa".

If you would get that hiccup from the kana, then you *should* get that
hiccup from the romaji.

If that's not the reason why you get a 'hiccup' from the apparent
mismatch, then I don't understand what the reason actually is.

It's entirely possible that there is ambiguity in the
pronunciation - but if so, there is the exact same ambiguity in
the kana as there is in the one-to-one romaji. You would have
to figure out from context that は is pronounced "wa"; why
should you not have to figure out from context that "ha" is
pronounced "wa"?

I never said that the one-to-one romanization scheme was
completely unambiguous in all aspects. I said (or would have
said) that it unambiguously represents the kana. If there's
ambiguity in the kana - which there is - then there will be the
exact same ambiguity in the corresponding romaji, and that's
precisely the way it should be.

These last two paragraphs (the latter more than the former) were
the main point of my response, and indeed I think they sum up the
main point of my entire argument; is there any particular reason
you didn't reply to them?

None in particular that I can think of. I'm mostly uninterested in
the arguments about representing kana.

But that's exactly what romaji are for...

If you're not talking about that, then I don't think we're having the
same conversation, and we probably can't get anywhere useful out of
this.

I would, however, like to hear your opinion (if indeed you have one -
since, contrary to popular the impression, opinions are optional) on the
point in the last sentence from the second paragraph, above: that any
ambiguity in the kana should be matched by exactly the same ambiguity in
the romaji.

I'm more interested in using romaji or kana or pictures of avocados
to represent *speech*.

Kana (and kanji) normally do represent speech, in Japanese.

Romaji are, purely and exclusively, a means of making written Japanese
accessible to people who have not learned the kana and/or the kanji.

Romaji are not for representing spoken Japanese, and should not be
treated as if they were. Any use of romaji to represent spoken Japanese
should be taken through the (potentially implicit) intermediate stage of
representing it as kana and/or kanji first.

I'm very interested in the idea of a human who reads without any
phonological representation in the brain. Maybe you're a mutation
that will result in a new stage of evolution.

I never said I read without phonological representation. I just said
that, when considering reading by itself, it doesn't matter how someone
mentally pronounces what they read as long as they understand it.

(Also, is there any particular reason you haven't been snipping out
these sections that you haven't been replying to?)

You are quite a demanding fellow, aren't you? Any particular reason
for that?

None that I could name. This wasn't particularly intended as a demand,
however, though of course I'd like it if you did start doing that; it
was intended as the question it appears to be.

I just think of snipping as part of standard Netiquette, and it
disconcerts me to see a quoted segment without a reply, unless there's
some specific reason why the segment would need to be left in despite
the lack of a response (context for a part which did get a response, for
example). This is particularly the case when the unsnipped part was in
some way important, and the important part didn't get touched on in the
actual replies.

--
The Wanderer

Warning: Simply because I argue an issue does not mean I agree with any
side of it.

Secrecy is the beginning of tyranny.
.

Relevant Pages

  • Re: Megami71
    ... such thing as Romaji. ... Learn Kana." ... which is always pronounced with the slurred sh' sound. ... And that would make the Nanoha A's manga a derivative work of the ...
    (rec.arts.anime.misc)
  • Re: Is "wa" a punctuation mark?
    ... If it's because the "natural" pronunciation of that spelling doesn't ... If you would get that hiccup from the kana, ... But that's exactly what romaji are for... ... consequent purpose of an attempt to aid the reader in knowing what ...
    (sci.lang.japan)
  • Re: Beginner Question about Kana/Kanji/Romaji
    ... That I should study Kana & Kanji first. ... I do think you should be learning to read kana from the first ... I think there's a good reason *against* learning romaji in the early ...
    (sci.lang.japan)
  • Re: Megami71
    ... no such thing as Romaji. ... Learn Kana." ... of representing all such small characters. ... and allows you to pick from any other kanji that sound the same. ...
    (rec.arts.anime.misc)
  • Re: amusing quodlibet on YouTube
    ... Can you by any chance tell me how to pronounce the word ... My English-> Japanese dictionary gives an animal's den or lair as su-ana ... or in Nelson's kanji dictionary, and ana (hole, pit, cave, lair, den, ... If you could send me a jpg or gif or the ideographs or kana that ...
    (rec.music.filk)