Diaphanitas

Several members over at the Aquastar Anime/Manga Forums have been debating about whether ナツメグ approximates Nutmeg or not. (ナツメグ, in Negima, is used as a name of a character.) It is distressing that there is a group of people who reject the Nutmeg spelling in favor of the otherwise meaningless transliteration Nutsumegu. I have been wondering about the skepticism toward the real word, so I looked around to see the arguments that the Natsumegu proponents were presenting. Not surprisingly, I wasn't convinced by any of them. Here is something that someone named Esperman said in response to the idea that ナツメグ approximates Nutmeg:

Are you sure on that?

I mean, it would seem believeable that her name could be Nutmeg, but the "tsu" part of the name kinda makes me feel a little skeptical...

Wouldn't it be more believeable if it was Nattomegu, Na-T-To-Me-Gu, to be Nutmeg?

Na-Tsu-Me-Gu would be more romanized as Nutsmeg instead of Nutmeg...

But eh, Japanese have their own style of using katakana to romanize non-Japanese words, so it coud be Nutmeg...

Well, until it's revealed, I think we'll just have to wait on how Natsumegu's name is romanized, if it even is romanized at all... (For all we know, it might not even translate into an english word...)

That looks like an example of skepticism that comes from invented restrictions.

First of all, we can show easily that ナツメグ is used very often to approximate nutmeg even if the katakana-character combination seems odd to us: source/source/source/source. If we were simply hypothesizing the idea that ナツメグ approximated Nutmeg, and we did not have the evidence to back up the idea, then perhaps it would be more believable if ナットメグ (Nattomegu) approximates Nutmeg.

There is also the fact that ナツ approximates nut in various words: ドーナツ for donut, ピーナツ for peanut, and ココナツ for coconut. There is also フルーツ, which approximates fruit. An anti-Nutmeg person could counter that by saying that the tis being approximated by ツ when that t is at the end of a word. And then, if you listen carefully, you might hear the sound of goalposts being moved. (This is like how some have said that エリオス does not approximate Helios.) At this point, it seems that we have to show that ツ can approximate t in places other than at the ends of words. However, it has not been shown that such a restriction even should be considered in the first place. One explanation for the use of ツ in approximations such as ドーナツ is that whoever rendered donut (and the others) into katakana characters, the person had the tu transliteration in mind for ツ. In that case, ツ is not tsu, but tu, so ナツメグ is not Natsumegu, but Natumegu, a transliteration that is closer to Nutmeg than Nattomegu.

As far as I can tell, the skepticism toward Nutmeg is unjustified because an actual legitimate case against it has not been made. I can see plenty of good reasons to use Nutmeg for ナツメグ, but I have not been shown any good reasons to use Natsumegu.

Another person said this:

Let me ask one question: is Nutmeg a valid first name in english?
Moreover we still don't know her last name. Also that wiki link, prooves only that "natsumegi" translates to "nutmeg" only if it means thing not person.

In other words, we have shown that ナツメグ refers to nutmeg as in the tree, the seed, or the spice, but we have not shown that ナツメグ refers to Nutmeg the person. Yes, somebody actually argued that. But katakana-character combinations do not make such a distinction, so the restriction is artificial and unjustified.

And finally there is the old canard...

How do we know that real words/widely-accepted words are better than words that are not? Especially when we're dealing with puns?

This seems too much like a last-ditch tactic for those who want to find something wrong with a particular spelling. There really is not anything that keeps us from applying this principle to any name that that does not have a Roman-letter spelling in the relevant sources. Sure, we could say that we are more certain about the spelling of a certain name because the spelling is a real word or a widely-accepted word, but that says nothing about whether we should use that instead of a word that is not widely-accepted.

But how can we show that widely-accepted words (or real words) are more appropriate, better, more favorable than words that are not?

1. We should use a system of spelling that subordinates itself to something outside of its own tenets, which is the relevant source that we have accepted as canonical. We use spelling systems and spelling patterns that are observed most often in the source. It can be shown that the source more often than not uses real words. By analogy, we should use a system of spelling that does the same thing. Newly-coined words in the source that have a traceable etymology can be more linguistically and etymologically congruent than others (e.g. Librar is more congruent than Rye-bral to library.). The canonical source leads and we follow along with it.
   
   
2. The existence of a pun does not necessarily imply a) that there is an unusual spelling or b) that all of the relevant spellings and words are interchangeable. (In the case of ナツメグ, there does not seem to be a pun involved.)
   

Perhaps it would be a great idea if I wrote an article like this for DIES GAUDII. My poor site needs to be updated! There is so much I want to do, but I just do not have enough time anymore!

LWotP: myristica fragrans "nutmeg."