[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [edict-jmdict] Inclusion of gikun readings in JMdict/Edict



My own thought is that gikun readings can be manufactured willy-nilly by applying the reading of one synonym to the headword for another. So there should be a relatively high bar of entry for gikun readings that make it into the dictionary; including idiosyncratic ones would just make a mess of things, and possibly cause more confusion than clarity.

I think part of the problem with the “Google it” method (in addition to Jim’s comment) is the idea that--to some extent--if a gikun reading is uncommon enough that the reading consistently has to be supplied alongside the kanji, then that may also kind of suggest that it the gikun is not common enough that it needs to be recorded in the dictionary. The actual proof that a gikun is worthwhile is if the gikun reading is widely known by native speakers even ~without~ the reading supplied. That’s something that’s much more difficult to show just from Google hits, I think.

After all, this is a Japanese–English dictionary (not English–Japanese), so: 1) helping our users to select artistic ways of writing things when translating from English to Japanese should not really be a concern, and 2) if the gikun reading is so uncommon that it has to be given to the reader anyway, then they can just look up either the headword or the real entry for the reading.




On Nov 13, 2014, at 1:44 AM, Matt Bloedel matt_bloedel@********* [edict-jmdict] <edict-jmdict@***************> wrote:


Thanks for the reply, Nils. I understand everything you are saying and don't really disagree with any of it. The gikun I am proposing adding are from Meiji period translation of The Bible: 
明治元訳 in which gikun is extensively and thoughtfully used throughout.  Here is an example found in this translation of the bible: 

Google "元始(はじめ)" (First word in Genesis 1 and also in John 1)  and you will get 1,320 search results, most quoting one of these two references I'm quite sure. This is a nice example of gikun using multiple Kanji (for origin and beginning) to convey more meaning than simply はじめに、初めに  (as found in other translations). These days you can even find electronic versions of this Meiji version. I see no problem in including gikun found in literature such as this translation of The Bible which is still used and quoted from to this day. I would however include a note explaining that it is a gikun reading found in Meiji period literature. 

Matt
 


On Wednesday, November 12, 2014 10:13 PM, Nils Roland Barth <jmdict.nbarth@********> wrote:


Hi Matt,

Do you have some examples in mind?
I suspect that most cases wouldn’t be suitable to include.

I’m keenly interested in the variety of readings, so more
gikun are appealing, but I’m concerned that they’re often
simply synonyms, and better treated as such.

Concretely, look at ja ’pedia gets us:
https://ja.wikipedia.org/wiki/義訓
(“Term mostly used for ancient texts like Man'yōshū, also for manga.”)

An example they give is glossing 本気 as マジ – how would we
feel about this? マジで?ガチ??

As a proud and obnoxious Kansai-ben speaker (and are there any other kind?),
I& x2019;d enjoy texts that glossed 本気 as ホンマ but I wouldn’t
want to include this as a reading in a dictionary!

In both these cases, listing the separate *words* as (near) synonyms
(casual form, regional form) is certainly appropriate
(though the semantic web on JMDICT is rather sparse), but
listing these are *readings* feels a serious stretch.

What benefit would a reading give to a user of the
dictionary that a semantic relation/synonym mention wouldn’t?
To understand these often requires knowing 2 words and
drawing the connection: it’s not usefully an independent
reading. (These are fundamentally puns.)

Against that there’s a serious risk of muddying waters with
minor idiosyncratic readings and confusing typical readers.

Lastly, I’d be very careful about using explicit glosses as evidence.
While these are useful for disambiguation in names (例:にほん・にっぽん)
they are also widely used exactly for idiosyncratic readings
(because this is *not* a normal reading), and are liable to
result in false positives for a joke that got spread around
but isn’t worthy of dictionary mention.

So while I’m supportive in principle, I’m deeply skeptical
in practice.

On a lighter note: some of my favorite puns along these lines ar :
四月一日 わたぬき(綿抜き)
  change of seasons, so unpadded kimono
小鳥遊・雀遊 たかなし(鷹無し) instead of 高梨
  no hawks, so the birdies play

There are in the names dictionary, and perhaps could use
some tagging as puns, but otherwise this seems the right
place for them.

  ~nils