Which extended Latin characters never occur at the beginning of a word?

Vasil StanevVasil Stanev Posts: 230
edited September 10 in Technique and Theory
The ones like Eszett. Excluding rare orthographical cases like accented letters that are only found in dictionaries.
I am aware a text can be all-caps.

This is relevant to a font of mine made up from initials and intended for drop caps.

Comments

  • Kent LewKent Lew Posts: 800
    edited September 10
    I believe that Ŋ, in its Non–African language form, is only used in the Sami languages, where it never begins a word.
    However, I am not certain about which form is preferred in some of the indigenous North American languages that use this character, nor whether it would begin a word.

  • Is that Latin-1 Supplement, Latin Extended-A, Latin Extended-B, Latin Extended-C, Latin Extended-D, or Latin Extended Additional?

    AFAIK Ṃ Ṇ and Ṅ never occur at the beginning of a word. 
  • In all those cases, they're using ß as a substitute for β.
  • We can't care about that.
    We have to be tight because they're sloppy.
  • I’ve met several people (presumably non-German speakers) who are convinced that β is one of the few greek characters directly accessible from the US Macintosh keyboard.
  • Vasil StanevVasil Stanev Posts: 230
    edited September 11
    I think my question is quite exact and so far only @Kent Lew and @Bhikkhu Pesala gave exact answers (no offence :) ). I am not commenting on the feasability of drawing or not drawing this or the other glyph. But I would never, e.g. design a Cyrillic ь for a drop cap, because it is never used at the beginning of words. Maybe the same is true about Ñ? It takes a native speaker to illuminate that and I think it would be most useful to all type professionals to be aware of these situations while they design their products. I have touched on this issue in previous topics of mine. I have a strict "no corners cut" policy for my fonts, but in this particular case the Beziers of the Basic Latin are pretty detailed and I have to take font size into consideration if I extend the Latin for the pro version.


  • Ñ is used at the beginning of Pāḷi words such as Ñāṇa = knowledge
  • Frode Frode Posts: 54
    edited September 11
    ĸutjuk
  • Frode Frode Posts: 54
    Another thing to consider re. the ß is that it may be doubled: “Äwwer et eßß en ripoarisch.”
  • André G. IsaakAndré G. Isaak Posts: 239
    edited September 11
    Sorry — I mispoke with respect to kra, which isn’t actually unicase. However, the unicode rendering of the uppercase is two separate characters, so kra isn’t needed as a drop-cap. We’d get KʼUTJUK with a plain vanilla K (or, more likely, QUTJUK unless you’re reproducing pre-reforming spelling).
  • But drop caps, depending on the language, may need to include punctuation. Opening quotes are often styled as ‘A or «B. Further, what about L’ (French/Italian) and ’s (Dutch)?

  • AFAIK Ṃ Ṇ and Ṅ never occur at the beginning of a word. 
    But you just used them at the beginnings of words :) Every letter can appear on its own in a text that talks about this letter.
  • But drop caps, depending on the language, may need to include punctuation. Opening quotes are often styled as ‘A or «B. Further, what about L’ (French/Italian) and ’s (Dutch)?
    Тhat is a very valid point. :)
  • Bhikkhu PesalaBhikkhu Pesala Posts: 171
    edited September 11
    Quotation marks do not need to be included as Drop Capitals. Below is my Pali font with Stylistic Alternates (salt) used as Drop Capitals in Serif PagePlus. The font supports colour, but PagePlus does not, so that was added manually.


  • Hrant H. PapazianHrant H. Papazian Posts: 1,143
    edited September 11
    Figuring out what not to include here is not unlike the case with kerning. Like if you believe lc-UC shouldn't be kerned because it's grammatically incorrect...
    http://typedrawers.com/discussion/comment/35275/#Comment_35275

    Do prioritize "correct" usage, since time/effort is always a limited resource. But always remember what lurking things (often quite common, if "incorrect") might be worth supporting after all.
  • Frode Frode Posts: 54
    Kʼutjuk / ĸutjuk is the current day Labrador Inuttitut spelling. Greenlandic swapped ĸ for q in the 70s.
  • Interesting... I was unaware that kra had ever made its way to Canada. I’d just automatically assumed that was Kalaallisut.
  • Ŋ is never used initially in words in Sami languages, where its uppercase has the N-form, but certainly does in some Australian languages where its uppercase also most often has the N-form. For example in Ŋaymil (the name of the language itself). The n-form is definitely used word initially in several African languages like Bambara or Songhai.

    Ṅ is used word initially in Igbo and Venda.

    Ṇ is used word initially in Angas.

    Ṃ isn’t used word initially as far as I can tell, neither in Indic script transliteration nor in Heiltsuk orthograhpy.

    Ñ is used word initially in Wolof.

  • Kent LewKent Lew Posts: 800
    edited September 13
    Although Ṃ is not part of the official orthography of Marshallese, isn’t it sometimes found as a substitute for M̧ (due to the latter not having an encoded precomposed form)? This is the case for the Marshallese-English Online Dictionary (MOD).
    As such M̧/Ṃ would be found word initially in Marshallese. For example, m̧ōñā/ṃōñā "food".
    (Not that the MOD is going to have much use for a decorative initials font. ;-)
  • edited September 13
    Ah yes, nice find! Ṃ is used word initially in Marshallese. The orthography of the MOD is what should be used, ie. the dot below instead of the cedilla, the official document was being in worked on last time I heard.
  • Vasil StanevVasil Stanev Posts: 230
    edited September 14
    So basically there is always a language or need for retro-proofing that nullifies copouts. Got that now.
  • Kent LewKent Lew Posts: 800
    You can never say never. So each designer has to decide where the point of diminishing returns is for him/herself in cases like this.
Sign In or Register to comment.