Romaji reverse representation of the katakana ティ (from English -ty ending)?

[Shannon Jacobs]

Found a number of questions about how to enter it. Personally, I use the sequence t-e-l-i-F7 (or some other convert key), but I'm asking a rather screwier question, and so far I haven't been able to find a prior reference, outside in the big Web or inside this nearly morbid newsgroup...

How should the katakana form ティ be represented in Romaji? Yes, it probably came from "ty", but the letter "y" is not a legal character in Romaji, so how should I reverse the writing?

Longish story why I want it, but is anyone actually interested in those bizarre details?

[Ben Bullock]

Post by Shannon Jacobs
Found a number of questions about how to enter it. Personally, I use the sequence t-e-l-i-F7 (or some other convert key), but I'm asking a rather screwier question, and so far I haven't been able to find a prior reference, outside in the big Web or inside this nearly morbid newsgroup...
How should the katakana form ティ be represented in Romaji? Yes, it probably came from "ty", but the letter "y" is not a legal character in Romaji, so how should I reverse the writing?
Longish story why I want it, but is anyone actually interested in those bizarre details?

I have a frequently-asked question page about it here:

http://www.sljfaq.org/afaq/kana-without-romaji.html

The text is as follows:

------

What kana do not have romanized forms?

There is no standard form of romanization for some forms of kana. In particular there is no official form of romanization for full-sized kana combined with smaller versions of the vowel kana, 'ぁ', 'ぃ', 'ぅ', 'ぇ' and 'ぉ', the smaller versions of the ''y'' kana, 'ゃ', 'ゅ', and 'ょ', and the sokuon or small ''tsu'' kana 'っ'.

There is also no standard way of romanizing common combinations such as 'トゥ' of katakana ''to'' and small ''u'', used to represent sounds as in the English word 'too'. Some people write this pair as ''tu'', but this may be confused with the ''tu'' Nippon-shiki and Kunrei-shiki romanizations of the kana ツ, romanized as ''tsu'' in Hepburn romanization.

On a computer or word processor, these smaller kana may be produced in various ways. For example, an 'x' or an 'l' preceding the romanization of the full-sized kana produces a small version on some systems, thus ''xtu'' gives 'っ' on a Microsoft computer. See How can I input small kana like ぃ, っ and ょ, or づ on a computer? However this is not standardized, and these forms are restricted to use in input methods (see How is Japanese input on a computer?); they are not used to represent the smaller kana in romanized Japanese.

99式 is an extended romaji system from The Society for the Romanization of the Japanese Alphabet that is intended to formalize solutions to many of the above shortcomings.

------

There is also another page on how to input these characters on a computer:

http://www.sljfaq.org/afaq/small-kana.html

------

How can I input small kana like ぃ, っ and ょ, or づ on a computer?

Most input methods recognize several different ways of inputting these kana. As a group, ティ or ディ can usually be input as "thi" or "dhi". Small kana like ぃ, っ and ょ can usually be input by typing x or l before, hence "xi", "xtu", and "xyo".

The づ kana in つづける should correspond to "du" in most input systems, but if not then "dzu" is usually OK.

If you need to input ゐ (wi) or ゑ (we), the obsolete kana (see What is the use of the "we" and "wi" kana?), these can be obtained by typing "wi" or "we" and then using the henkan (kana to kanji) function.

See also What are the names of the Japanese non-kana, non-kanji symbols? for the names of various punctuation symbols, which can be used to input them.

------

[Jim Beard]

Post by Shannon Jacobs
Found a number of questions about how to enter it. Personally, I use the
sequence t-e-l-i-F7 (or some other convert key), but I'm asking a rather
screwier question, and so far I haven't been able to find a prior
reference, outside in the big Web or inside this nearly morbid
newsgroup...
How should the katakana form ティ be represented in Romaji? Yes, it
probably came from "ty", but the letter "y" is not a legal character in
Romaji, so how should I reverse the writing?
Longish story why I want it, but is anyone actually interested in those bizarre details?

Entry of Japanese symbols depends entirely on your software. Specify
your OS and software, and someone might be able to help.

Romanization is another matter. If all you want is a conversion, there
is a kunreishiki table at the bottom of WIKI page,

https://en.wikipedia.org/wiki/Kunrei-shiki_romanization

Choice of which form of romanization to use is yet another matter, and
depends on your purpose.

Kunreishiki matches nicely with Japanese orthography and grammar, and is
the one I prefer to see.

Until 5th Edition, Kenkyusha's Japanese-English dictionary had entries in
romanized form, and due to near-universal reliance on it in translation
to English its form of modified Hepburn romanization was adopted by most
as the standard, with some dropping the indication of long vowels as a
matter of style in English.

Placenames were of particular significance to some, and the usual
solution was to obtain and use the U.S. Board of Geographical Names index
of Japanese place names. The latest print version of the index I have
seen was from the 1950s or 1960s, but you can do a look-up or search for
Japanese place names at http://geonames.nga.mil/gns/html/

The current BGN romanization system for use by U.S. Government offices is
here.

http://geonames.nga.mil/gns/html/Romanization/Romanization_Japanese.pdf

Cheers!

jim b.

[Ben Bullock]

Post by Jim Beard

Post by Shannon Jacobs
Found a number of questions about how to enter it. Personally, I use the
sequence t-e-l-i-F7 (or some other convert key), but I'm asking a rather
screwier question, and so far I haven't been able to find a prior
reference, outside in the big Web or inside this nearly morbid
newsgroup...
How should the katakana form ティ be represented in Romaji? Yes, it
probably came from "ty", but the letter "y" is not a legal character in
Romaji, so how should I reverse the writing?
Longish story why I want it, but is anyone actually interested in those
bizarre details?

Entry of Japanese symbols depends entirely on your software. Specify
your OS and software, and someone might be able to help.
Romanization is another matter. If all you want is a conversion, there
is a kunreishiki table at the bottom of WIKI page,
https://en.wikipedia.org/wiki/Kunrei-shiki_romanization

This entire page and the table you specify do not contain any information
whatsoever which answers OP's question.

Post by Jim Beard
Choice of which form of romanization to use is yet another matter, and
depends on your purpose.
Kunreishiki matches nicely with Japanese orthography and grammar, and is
the one I prefer to see.

That has absolutely nothing to do with OP's question.

Post by Jim Beard
Until 5th Edition, Kenkyusha's Japanese-English dictionary had entries in
romanized form, and due to near-universal reliance on it in translation
to English its form of modified Hepburn romanization was adopted by most
as the standard, with some dropping the indication of long vowels as a
matter of style in English.
Placenames were of particular significance to some, and the usual
solution was to obtain and use the U.S. Board of Geographical Names index
of Japanese place names. The latest print version of the index I have
seen was from the 1950s or 1960s, but you can do a look-up or search for
Japanese place names at http://geonames.nga.mil/gns/html/
The current BGN romanization system for use by U.S. Government offices is
here.
http://geonames.nga.mil/gns/html/Romanization/Romanization_Japanese.pdf

Again, there is no information here at all which answers OP's question.

[Jim Beard]

Post by Ben Bullock

Post by Jim Beard

Post by Shannon Jacobs
Found a number of questions about how to enter it. Personally, I use
the sequence t-e-l-i-F7 (or some other convert key), but I'm asking a
rather screwier question, and so far I haven't been able to find a
prior reference, outside in the big Web or inside this nearly morbid
newsgroup...
How should the katakana form ティ be represented in Romaji? Yes, it
probably came from "ty", but the letter "y" is not a legal character
in Romaji, so how should I reverse the writing?
Longish story why I want it, but is anyone actually interested in
those bizarre details?

Entry of Japanese symbols depends entirely on your software. Specify
your OS and software, and someone might be able to help.
Romanization is another matter. If all you want is a conversion, there
is a kunreishiki table at the bottom of WIKI page,
https://en.wikipedia.org/wiki/Kunrei-shiki_romanization

This entire page and the table you specify do not contain any
information whatsoever which answers OP's question.

Post by Jim Beard
Choice of which form of romanization to use is yet another matter, and
depends on your purpose.
Kunreishiki matches nicely with Japanese orthography and grammar, and
is the one I prefer to see.

That has absolutely nothing to do with OP's question.

Post by Jim Beard
Until 5th Edition, Kenkyusha's Japanese-English dictionary had entries
in romanized form, and due to near-universal reliance on it in
translation to English its form of modified Hepburn romanization was
adopted by most as the standard, with some dropping the indication of
long vowels as a matter of style in English.
Placenames were of particular significance to some, and the usual
solution was to obtain and use the U.S. Board of Geographical Names
index of Japanese place names. The latest print version of the index I
have seen was from the 1950s or 1960s, but you can do a look-up or
search for Japanese place names at http://geonames.nga.mil/gns/html/
The current BGN romanization system for use by U.S. Government offices
is here.
http://geonames.nga.mil/gns/html/Romanization/Romanization_Japanese.pdf

Again, there is no information here at all which answers OP's question.

Not knowing his OS and software, I cannot address input key sequences,
and I note that all you did was provide possibilities. Better to respond
in a manner that may lead to a definitive response.

I find it useful to elaborate beyond that required to answer a question
now and then. Better to teach someone where to find answers than dole
them out one at a time, unless your ego is gratified by someone coming
back and back and back, wasting the time of all concerned.

While the table did not present the ティ romanization, in the immediate
vicinity on the page there is a sentence answers OP's question how to
romanize it (in Kunreishiki) precisely.

"For some Japanese speakers, however, the sounds ティ "ti" and チ "chi" are
the same phoneme; they are both represented in Kunrei-shiki as tîmu."

Thank you for reminding me why long ago I concluded you were a pompous
egotistical asshole.

Cheers!

jim b.

[Ben Bullock]

Post by Jim Beard
Not knowing his OS and software, I cannot address input key sequences,
and I note that all you did was provide possibilities. Better to respond
in a manner that may lead to a definitive response.

No, I answered OP's question with what I believe to be correct information.

You did not answer OP's question, and you provided incorrect information.

Post by Jim Beard
I find it useful to elaborate beyond that required to answer a question
now and then.

You have neither elaborated beyond that required to answer a question
nor answered the question.

Post by Jim Beard
Better to teach someone where to find answers than dole
them out one at a time, unless your ego is gratified by someone coming
back and back and back, wasting the time of all concerned.

You have posted three pieces of wrong information so far. You're not in
a strong position to accuse others of timewasting.

Post by Jim Beard
While the table did not present the ティ romanization, in the immediate
vicinity on the page there is a sentence answers OP's question how to
romanize it (in Kunreishiki) precisely.

No there is not.

Post by Jim Beard
"For some Japanese speakers, however, the sounds ティ "ti" and チ "chi" are
the same phoneme; they are both represented in Kunrei-shiki as tîmu."
In Hepburn, these would be distinguished as different sounds and
represented tīmu and chīmu respectively, giving better indications
of the English pronunciations.

There is no representation of these sounds in Hepburn romanisation at all.

"The encyclopedia any idiot can edit."

Please refer to the frequently asked questions document I posted as
an original response.

[r***@gmail.com]

On Thursday, June 2, 2016 at 10:45:07 PM UTC-4, Shannon Jacobs wrote:
so far I haven't been able to find a prior reference, outside in the big Web or inside this nearly morbid newsgroup...
I hope you meant to say "this nearly moribund newsgroup". I don't think things have strayed quite that far afield . . . .

[Shannon Jacobs]

Post by Shannon Jacobs
so far I haven't been able to find a prior reference, outside in the big Web or inside this nearly morbid newsgroup...
I hope you meant to say "this nearly moribund newsgroup". I don't think things have strayed quite that far afield . . .

Hmm... I guess your word choice of "moribund" is more precise, though I feel bad about my mistake and I'm tempted to rationalize it by considering the tone of the tiff I seem to have triggered.

The funny part is that I still can't figure out the answer to my original question... I do think that Bullock-san tried to address it, but it seems his answer is that there is no answer, so let me try to reword it in a more answerable form:

Within the rules of Romaji transcription of the Japanese syllabalaries, which (as far as I know) do not allow for "ty" as a proper character sequence, how can I unambiguously represent the katakana string ティ that was derived from a foreign word ending in "ty"?

One solution might be to ignore the rules of Romaji transcription and simply appeal to the original language. I don't like that as a general solution because the same problem might arise with Japanese words borrowed from languages that are not written with the Latin alphabet.

Another solution would be to use "tei" and ignore the reduction of the ィ character. I don't like that solution because the Japanese pronunciation is different (though I had to ask a Japanese person about the difference).

That same Japanese person recommended writing it as "thi", and I think that is unambiguous but looks a bit bizarre. There are already enough problems with seeing Romaji through my non-Japanese eyeballs.

[Ben Bullock]

Post by Shannon Jacobs
The funny part is that I still can't figure out the answer to my
original question... I do think that Bullock-san tried to address
it, but it seems his answer is that there is no answer, so let me

If you say "てぃ" is "ti" you run in to the problem where "ti" has a
different meaning in Kunrei romanization, which is ち. You can't
specify these without specifying a particular romanization
system. Apart from the 99 shiki mentioned in the article I posted, he
only one I could find which addresses these issues is this American
standard from 1972:

http://www.ab.cyberhome.ne.jp/~kaizu/roomazi/doc/ansiz3911.html

Scroll down to "表3. 外来語の表記につかわれる特殊なカタカナ".

Post by Shannon Jacobs
Within the rules of Romaji transcription of the Japanese
syllabalaries, which (as far as I know) do not allow for "ty" as a
proper character sequence, how can I unambiguously represent the
katakana string ティ that was derived from a foreign word ending in
"ty"?

You can't do that without specifying the romanization system.

Post by Shannon Jacobs
One solution might be to ignore the rules of Romaji transcription
and simply appeal to the original language. I don't like that as a
general solution because the same problem might arise with Japanese
words borrowed from languages that are not written with the Latin
alphabet.
Another solution would be to use "tei" and ignore the reduction of
the ィ character. I don't like that solution because the Japanese
pronunciation is different (though I had to ask a Japanese person
about the difference).

Or use texi or teli or something.

Post by Shannon Jacobs
That same Japanese person recommended writing it as "thi", and I
think that is unambiguous but looks a bit bizarre. There are already
enough problems with seeing Romaji through my non-Japanese eyeballs.

The question you asked doesn't have an answer, there is no way of
romanizing these things unambiguously without specifying a
(non-standard) romanization system.

[Jim Beard]

Post by Shannon Jacobs

Post by Shannon Jacobs
so far I haven't been able to find a prior reference, outside in the
big Web or inside this nearly morbid newsgroup...
I hope you meant to say "this nearly moribund newsgroup". I don't think
things have strayed quite that far afield . . .

Hmm... I guess your word choice of "moribund" is more precise, though I
feel bad about my mistake and I'm tempted to rationalize it by
considering the tone of the tiff I seem to have triggered.
The funny part is that I still can't figure out the answer to my
original question... I do think that Bullock-san tried to address it,
but it seems his answer is that there is no answer, so let me try to
Within the rules of Romaji transcription of the Japanese syllabalaries,
which (as far as I know) do not allow for "ty" as a proper character
sequence, how can I unambiguously represent the katakana string ティ that
was derived from a foreign word ending in "ty"?
One solution might be to ignore the rules of Romaji transcription and
simply appeal to the original language. I don't like that as a general
solution because the same problem might arise with Japanese words
borrowed from languages that are not written with the Latin alphabet.
Another solution would be to use "tei" and ignore the reduction of the ィ
character. I don't like that solution because the Japanese pronunciation
is different (though I had to ask a Japanese person about the
difference).
That same Japanese person recommended writing it as "thi", and I think
that is unambiguous but looks a bit bizarre. There are already enough
problems with seeing Romaji through my non-Japanese eyeballs.

To repeat from one of my earlier posts,

"For some Japanese speakers, however, the sounds ティ "ti" and チ "chi" are
the same phoneme; they are both represented in Kunrei-shiki as tîmu."

This came from https://en.wikipedia.org/wiki/Kunrei and you need to
scroll down to Usage and look at the fourth paragraph below it. That
example dealt with ティーム(チーム) team and in Kunreishiki transliteration
both are rendered as "timu." Your ティ becomes ti, exactly like チ.

This may be ambiguous, but you have a consistent standard.

Ben seems to be wandering mentally around issues of computer input with
his texi or teli (part of your concern) but best to let him wander.

As a parting note, in language as in life, everything including meaning
understood is shaped by context (circumstances) and clear-cut unambiguity
is rare. As Einstein put it,

"As far as the laws of mathematics refer to reality, they are not
certain, and as far as they are certain, they do not refer to reality."

If math of a useful nature is not unambiguous, what hope have we for mere
human language? Learn to live with it.

Cheers!

jim b.

[Ben Bullock]

Post by Jim Beard
"For some Japanese speakers, however, the sounds ティ "ti" and チ "chi" are
the same phoneme; they are both represented in Kunrei-shiki as tîmu."
This came from https://en.wikipedia.org/wiki/Kunrei and you need to
scroll down to Usage and look at the fourth paragraph below it. That
example dealt with ティーム(チーム) team and in Kunreishiki transliteration
both are rendered as "timu." Your ティ becomes ti, exactly like チ.

This is not supported by the references in the Wikipedia article, and
it doesn't answer the question.

[TJ Edmister]

On Mon, 06 Jun 2016 01:53:39 -0400, Shannon Jacobs

Post by Shannon Jacobs
Within the rules of Romaji transcription of the Japanese syllabalaries,
which (as far as I know) do not allow for "ty" as a proper character
sequence, how can I unambiguously represent the katakana string $B%F%#(B
that was derived from a foreign word ending in "ty"?

Are you using "ti" elsewhere in your transcription for another purpose? If
you use "chi" for $B$A(B, and use "ti" only for $B%F%#(B, then there should be no
ambiguity.

Might as well make up your own rules. That's what everyone else does.
(Don't you love it when something is transcribed as "you" and it's not
clear whether the original was $B$h$&(B or $B$f$&(B?)

[Shannon Jacobs]

Well, mostly I feel like I've caused a spat, but I can point my finger and say "Japanese started it." Oh wait, I can't point at Japanese.

I think the real solution is to replace my software with a new program that supports proper Japanese and completely rejects the use of Romaji, which appeals to me on esthetic grounds, too. Even if it provided a field for a title in English letters, it should be a translated title with some kind of flag field to indicate its degree of official-ness. (Unfortunately that already seems to be a new can of worms.)

In the specific example that led to this question, the book サイバーセキュリティのひみつ would have no Romaji title, but I could add my unofficial translation of the title in a translated title field. (Good read, by the way. Surprisingly good coverage considering the target audience.)

Oh wait. What if the publisher or author included an official Romaji version of the title. There is NO escape. It's worse than Godel's Incompleteness Theorem... (Just joking. Nothing is worse than that.)

[Ben Bullock]

Post by Shannon Jacobs
Well, mostly I feel like I've caused a spat, but I can point my finger
and say "Japanese started it." Oh wait, I can't point at Japanese.

I think this discussion was useful in parts.

Post by Shannon Jacobs
I think the real solution is to replace my software with a new
program that supports proper Japanese and completely rejects the use
of Romaji, which appeals to me on esthetic grounds, too. Even if it
provided a field for a title in English letters, it should be a
translated title with some kind of flag field to indicate its degree
of official-ness. (Unfortunately that already seems to be a new can
of worms.)
In the specific example that led to this question, the book サイバー
セキュリティのひみつ would have no Romaji title, but I could add my
unofficial translation of the title in a translated title
field. (Good read, by the way. Surprisingly good coverage
considering the target audience.)

A lot of the time, what people do is to back-translate the original
word into English, so they write it something like "Cybersecurity no
Himitsu".

Post by Shannon Jacobs
Oh wait. What if the publisher or author included an official Romaji
version of the title. There is NO escape. It's worse than Godel's
Incompleteness Theorem... (Just joking. Nothing is worse than that.)

Maybe you need to define more clearly the problem you're trying to solve.

[r***@gmail.com]

Post by Shannon Jacobs
In the specific example that led to this question, the book サイバーセキュリティのひみつ would have no Romaji title, but I could add my unofficial translation of the title in a translated title field. (Good read, by the way. Surprisingly good coverage considering the target audience.)

Sorry, I'm confused. In an earlier thread, you stated that you couldn't read a web page because of limited Japanese skills. Despite this, you're translating a book?

[TJ Edmister]

Post by r***@gmail.com

In the specific example that led to this question, the book $B%5%$%P!<%;(B
$B%-%e%j%F%#$N$R$_$D(B would have no Romaji title, but I could add my
unofficial translation of the title in a translated title field. (Good
read, by the way. Surprisingly good coverage considering the target
audience.)

Sorry, I'm confused. In an earlier thread, you stated that you couldn't
read a web page because of limited Japanese skills. Despite this, you're
translating a book?

I believe you've mixed up two different posters.

[r***@gmail.com]

Post by TJ Edmister
I believe you've mixed up two different posters.

So I have. Sorry about that.

[TJ Edmister]

On Fri, 10 Jun 2016 17:13:26 -0400, Shannon Jacobs

Post by Shannon Jacobs
I think the real solution is to replace my software with a new program
that supports proper Japanese and completely rejects the use of Romaji,

Are you doing an automated (via software) transcription? Or do you mean
that the only reason you were using romaji is for the sake of software
which lacks Japanese support?

[Shannon Jacobs]

Sorry if I have said something to suggest that my Japanese is at a level that supports translation. My reading is at the level of the ひみつ series. However, I do use online translation tools sometimes. They can handle many of the obvious word-for-word replacements and usually provide hints for the grammar. From that rough form I may be able to figure out enough, and in other cases the rough translation guides me to the Japanese words that are keys to the remaining ambiguities.

Another way to describe my language skills is from the output side. When I am going to say something in English, I can immediately think of many ways to say variations of that thing and the best ways are immediately apparent, along with the different nuances and distinctions. In Japanese, I'm often desperately struggling for any way to express a meaning that might be close to my intention, and there is almost no chance that I will say it in a way that approximates natural Japanese.

This morning's example involved wanting to say "brick wall" in Japanese, as in the expression "like talking to a brick wall". It took me several minutes to decide that 煉瓦の壁 might be close enough--but I bet it isn't the most natural way a Japanese person would express the idea.

My original concern in this thread involved a perceived shift in the priorities of public librarians in Japan. My original example was with regards to their willingness or even eagerness to issue library cards, but yesterday I had an even better example of their new priorities. I returned a book called うめパワーのひみつ and the librarian actually ran after me to ask if I were responsible for a mark in the book.

I think the old priorities of old librarians would have been to encourage lots of people to read the books. Some of them would even think it was not such a terrible thing if so many patrons read the books that they fell apart and had to be replaced with new books. Checking for every possible mark in a returned book? Seems to be a new priority.

(Do I need a disclaimer that my very first job was in a public library? LONG time ago, and have no idea what I'll do next...)

[Shannon Jacobs]

Sorry if I have said something to suggest that my Japanese is at a level that supports translation. My reading is at the level of the ひみつ series, which is where the original question arose. I read them mostly for pleasure because they aren't too difficult, but they do manage to include some interesting information.

However, I do use online translation tools sometimes. They can handle many of the obvious word-for-word replacements and usually provide hints for the grammar. From that rough form I may be able to figure out enough, and in other cases the rough translation guides me to the Japanese words that are keys to the remaining ambiguities.

Another way to describe my language skills is from the output side. When I am going to say something in English, I can immediately think of many ways to say variations of that thing and the best ways are immediately apparent, along with the different nuances and distinctions. In Japanese, I'm often desperately struggling for any way to express a meaning that might be close to my intention, and there is almost no chance that I will say it in a way that approximates natural Japanese.

This morning's example involved wanting to say "brick wall" in Japanese, as in the expression "like talking to a brick wall". It took me several minutes to decide that 煉瓦の壁 might be close enough--but I bet it isn't the most natural way a Japanese person would express the idea.

[Crystal]

Post by TJ Edmister
Might as well make up your own rules. That's what everyone else
does. (Don't you love it when something is transcribed as "you"
and it's not clear whether the original was よう or ゆう?)

Even better: YouTube becomes ようつべ :-)