Pronunciation of W

Question

I often see the alphabet W being transcribed or pronounced as "ダブル" rather than "ダブリュー" in Japanese, and I think that in most cases, that was not what was meant. Indeed, there seems to be a practice of abbreviating "ダブリュー" as W, most often seen in a menu in a fast food restaurant, (whose meaning intentionally or unintentionally goes along with the origin of W being "double 'U' "), but even out of that context, I observe it being transcribed/pronounced as "ダブル", and I don't know if people are doing it because of the origin of the alphabet character W. Why do you think that happens?

Answer 1

I don't believe a definitive answer to your question is possible. That said, you asked what people think, so a couple of my own thoughts:

When reading out URLs (arguably now the most common use of W in modern Japanese), ダブリューダブリューダブリュードット would be a real mouthful, so it's often pronounced ダブルダブルダブルドット or even ダブダブダブドット. It's possible that this has contributed to the abbreviated pronunciation.

I don't know if you'll agree with this, but I find that the transliteration of borrow-words is very prone to change and evolution over time for convenience or what have you. The meaning "double" on fast food menus etc is probably the second most common use of W in modern Japanese. Presuming this use came first, this could also have influenced a change in the general pronunciation.

Alternatively, as you say, the English pronunciation of W comes from double-U, so you could say the most "correct" transliteration would be not ダブリュー but ダブル・ユー. ダブル would be a sane abbreviation of this, just like many 和製英語 are informally abbreviated, purely for the sake of brevity.

Answer 2

I believe this may an example of what is called a volatile pronunciation. Take a look at what is happening phonemically:

1. ダブリュー → ダブル
2. /dabuɾjuː/ → /dabuɾu/

Using the description in Vance(2008) of long vowels:

Japanese vowel length distinctions don't have a very high functional load; at any given point in a conversation mistakenly pronouncing a long vowel as a short vowel or vice versa is very unlikely to result in something that a listener could interpret as a plausible alternative utterance.

And most relevant to the above alternation:

Word-final long vowels are conspicuously susceptible to shortening, although the explanation for this tendency isn't clear.

Vance gives two good examples of this:

1. コンピューター　→　コンピュータ
2. 本当　→　ほんと

These facts would explain the long vowel drop of ダブリュー but doesn't yet account for why the palatalization /j/ is also dropped in /ɾjuː/ → /ɾu/ since an alternation of the form /ɾjuː/ → /ɾju/ would not result in a phonotactic infringement.

I think one way to explain the /ɾjuː/ → /ɾju/ is by taking a risk and generalizing or extrapolating Vance's implicit suspicions above to stipulate that:

1. Functional loads are a criteria in phonological processes. Namely, the lower the functional load the more susceptible to a (non-mandatory) phonological alternation
2. Word-final vowel shortening process can extend to certain semi-vowels. (Note that the palatalized /j/ of /ɾjuː/ is a semi-vowel)

Which, taking these as assumptions, the admissibility of /ɾjuː/ → /ɾju/ might be predicated on:

1. The alternation /ɾju/ → /ɾu/ has a low functional load
2. Palatalized semi-vowels in word-final mora are in the class of semi-vowels referenced above

I would love for someone to comment with more insight on what Vance might be alluding to. I suspect I just haven't read deep enough into Japanese phonology.

As for orthographic discrepancies, I truly cannot understand what you are asking. Native orthographic transcriptions can be notoriously volatile, and is of rather specialized concern. If you absolutely need an explicit correspondence, I would just advice the following function:
f({ダブル,ダブリュー})={W} with f^-1({U})⊈{ダブル,ダブリュー}