Re: [edict-jmdict] Editoriel Policy

On Jun 21, 2010, at 11:04 AM, René Malenfant wrote:

I don't believe that we are always going to be able to know that a species name is of foreign origin... although that is always the case when there really isn't a Japanese name for a species (especially if the species is not near to nor commercially relevant to Japan) and they simply katakanize some foreign name (sometimes the latin, sometimes the scientist to first describe the species, or the first Japanese to describe it, or sometimes the german, sometimes the english etc etc).
I fail to see the problem. In the rare case that you're unable to determine the origin of the word, submit it as you see fit. Personally, I would recommend that it be submitted as katakana:hiragana since that's probably in accordance with frequency of use. I would hope that another editor w uld be able to determine the etymology and change the entry if necessary.

The problem, as I see it, is for example... I don't speak Portuguese, Dutch, nor German... and yet any of these could be the origin of an otherwise native sounding word. So you didn't quote what I was responding to, but the idea that you always know a "native" Japanese katakana word from a "borrowed" katakana word is unlikely... especially when it comes to the many diverse and random names of species that could be based on latin or an author or a location on the planet which is then katakana-ized... is that a foreign word if the location where a species was found was written in the latin and then turned to Japanese ??? Gets confusing.

However, you will find almost all publications that deal with lists of species of some type deal primarily in katakana. That's an important aspect of the language that says loud and clear that this is taxonomy. I would argue that ignoring this without grounds is incorrect.
All Japanese and English dictionaries ignore many taxonomic conventions. For example, Japanese dictionaries do not list Japanese species names as katakana (nor do they fail to omit the kanji headword even though it may only rarely be used); English dictionaries do not use capital letters in standardized species names. Both may choose to represent a species under a more common (but non-standard or even incorrect) name rather than the official name. A dictionary is not a taxonomic database.

ALL? www.oceandictionary.net lists thousands of organisms by latin, and provides (almost) nothing but the katakana name. Does that fail to meet the standard of a dictionary? Or does it have to fail to use standard taxonomic conventions to be considered a dictionary?

I use about 4 or 5 sources which attempt to list a large number of the fish species of the world to make many of my entries. None use hiragana nor kanji compounds. The only place that consistently provides kanji compounds if and when they are available is wikipedia, and it lists them as subordinate entries and never in my experience as the primary entry. So in contrast is the current policy as you have written it:
The other place that consistently provides kanji for species names (even though they may never be seen "in the real world") is... Every kokugo dictionary I've ever seen (and most J-E dictionaries except for Eijiro). And we're building a dictionary.

In my opinion, the reason that your sources do not include kanji headwords is because: (1) they're irrelevant---katakana is sufficient for taxonomic purposes, although insufficient for the purposes of a proper dictionary, (2) laziness in the face of "too much effort".

The reason katakana is used is as I described earlier... because that's the convention for species names. I really don't think "laziness" was the motive. The overwhelming factor is that the number of species which happen to have a kanji based name is extremely small in comparison to the number of species who are ONLY named via katakana... which you then ascribe to laziness, and invent kanji compounds which nobody has ever used... except you... that's the danger here.

I focus on the random dictate: "should include the headword written entirely in kanji, even though it may be only rarely used in practice" which is tantamount to an entire policy which has no basis in actual Japanese usage, and even states that it's basically only a random convention of JMDICT... with no relationship to "practice".
A "random dictate" that seems to be in effect by the publishers of all Japanese dictionaries.

Again ALL? Except none of the dictionaries devoted to actual living creatures...

And I consider this to be dangerous practice for a dictionary: "When unsure of a kanji headword, it is often easy to determine based on the English translation or the appearance of the species." Basically I might interpret this to say that even though its not used, or you can't prove its being used, make it up and make it the leading entry.
As I wrote, there is a difference between being able to determine something with absolute certainty and guessing. If you're not sure, don't submit it. But Japanese words aren't just random collections of sounds. A native Japanese speaker KNOWS that each part of ホオジロオナガガモ has a meaning and that it means 頬白尾長鴨. Almost all Japanese species names can be ~determined~ (emphatically not "guessed") in this manner.

And it's important to include the kanji, because to someone ~learning~ the language (our target audience), something like ホオジロオナガガモ seems like a mere nonsense word and is nearly impossible to remember until it's displayed as what it means---頬白尾長鴨 obviously means "white-cheeked pintail" and by giving the kanji you allow people to learn this and remember it with minimal effort.

Re: [edict-jmdict] Editoriel Policy - kana order