Dutch IJ with dots

135

Comments

  • John SavardJohn Savard Posts: 1,088

    People can learn to read things as complex as Chinese;
    Yes, they can.

    The Chinese writing system is not as complex as is popularly believed by outsiders. A native speaker of Chinese - particularly the Mandarin dialect, but this is true of other Sinitic languages as well - does not need to memorize even 3,000 characters, let alone 40,000, in order to read Chinese.

    Instead, about 500 characters - individual symbols, and ideographic compounds, such as sun + moon for "bright" - need to be memorized. Other characters can then be read on the basis of their component parts: a radical (boushou, literally "classifier") and a phonetic - a previously-known character having the same pronunciation.

    However, to write Chinese, one does have to memorize exactly which phonetic to use, as there are multiple possibilities, for each character of this type. If one uses the wrong phonetic, one's handwritten Chinese will still be understood, but, as with English, spelling mistakes give one away as illiterate. (The good news, of course, is that if one does enough reading, one will eventually learn correct spelling without too much effort.)

    The complexity of Chinese writing, therefore, has to an even greater extent the same destructive result as the complexity of English spelling: reinforcing social stratification, increasing the amount of time children must spend in school.
  • @John Savard
    Now there's an interesting hypothesis. Complexity of spelling's correlation with social inequality. Question is: how could we ever establish what is consequence and what is cause?


    Regarding the fij-problem:
    I may be able to speak for the younger generation. An fi ligature followed by a j, as in fijn, does look a little awkward. However, in many typefaces an unligated fi sequence will also look awkward, so the choice is one of two evils. Personally, I would opt for the ligature, since it stands out less than the excess whitespace that may be caused by an unligated fi. The best option (for this particular problem) remains an f that doesn't require a ligature. One could make such an f to be used only in the fij instance, but I suspect that would look awkward again.
  • Ben BlomBen Blom Posts: 250

    In the course of history, the “ij” has been dropped from Dutch keyboards, and the “y” in the spelling of most Dutch words has been replaced by “ij”. Now one cannot use one’s little finger anymore to type an “ij”. In some Dutch names, the old “y” coexists with the new “ij”, which may be confusing (Feyenoord/Feijenoord; Cruyff/Cruijff; Ypma/IJpma). One might consider turning back history to be progress.

    Hrant H. Papazian: There’s no progress without effort.

    Indeed. But not all ideas for change are practical. The keyboard we have today, is quite similar to the original typewriter keyboard. Many attempts to modernize the keyboard have failed. When people are used to drive on the left side of the road, some might say it would be an improvement to drive on the right side of the road. A change like that has happened in some countries—but it’s an endeavor most people would like to avoid.

  • Kent LewKent Lew Posts: 905
    One could make such an f to be used only in the fij instance, but I suspect that would look awkward again.

    As long as {calt} comes before {liga}, this should work for localized interception:
    feature calt { <br>    script latn;<br>        language NLD; <br>            sub f' i j by f.short;<br>} calt;<br>

  • Ben BlomBen Blom Posts: 250
    edited November 2017

    To me, “fiets” with ligature, looks weird directly after “fijne” without ligature. It looks inconsistent. I would prefer this:

    Perhaps it looks weird because both “ij” and “ie” represent a single vowel—and with the ligature in “fiets”, half of the vowel is connected with the “f”. In “fiets”, “i” and “e” belong together, not “f” and “i”. Compare with “mies” in the image below.

  • Agreed that fi ligature should not be used with ij. But I think in headline sizes the ligatures draw attention to themselves anyhow. In text size I'm not having really an issue with the fi ligature in 'fiets'.



  • Since f-ligatures are mostly a solution to a visual not phonological or linguistic problem, I never understood why it matters whether the i is a half vowel or belong to a different syllable, since the original visual problem is still there regardless? Do people in cursive handwriting also have an issue with cursive connections in these same scenarios?
  • Ben Blom said:

    In the course of history, the “ij” has been dropped from Dutch keyboards

    Yes, things change... :-)  Especially now with touch-screens. It's never too late to improve. Which necessarily comes with some sacrifice.
  • Since f-ligatures are mostly a solution to a visual not phonological or linguistic problem, I never understood why it matters whether the i is a half vowel or belong to a different syllable, since the original visual problem is still there regardless? Do people in cursive handwriting also have an issue with cursive connections in these same scenarios?
    Ligatures can be a solution to way more than that.
    Making "fi" a unit where "ij" is a more meaningful one is bad.

    And handwriting is not reading.
  • Kent LewKent Lew Posts: 905
    Perhaps it looks weird because both “ij” and “ie” represent a single vowel
    Hmm. Interesting. Is that a constant? That is, does the sequence “ie” always represent a single vowel in Dutch? Could the Dutch localization be expanded to include
        sub f' i e by f.short;
    leaving fi to ligate except when followed by j or e? Seems like such things are usually bound to run afoul of loanwords or other exceptions.

    Yet another reason why trying to address such linguistic things from within the font may not be the ideal approach.
  • So what about a ligature of f with ij-acute?  :-Þ
  • John HudsonJohn Hudson Posts: 2,955
    For the record, Brill favour the ligated f+ij in their typography (for which I provide a contextual variant of the /ij glyph, since Brill also favour a single glyph for the ij vowel).

  • For the record, Brill favour the ligated f+ij in their typography (for which I provide a contextual variant of the /ij glyph, since Brill also favour a single glyph for the ij vowel).
    Nice! Is there a visual difference between /ij glyph and the /i/j combination?
  • John HudsonJohn Hudson Posts: 2,955
    No, they look the same, but Brill actually make use of the precomposed IJ/ij characters when typesetting Dutch text.
  • Ben BlomBen Blom Posts: 250
    edited November 2017

    Except for “ij” and “ie”, there are no two-letter vowels in Dutch that start with “i”.

    I searched for Dutch words that contain “fij”, in which “ij” is not a vowel. I did not find such a word—so I believe in the Dutch letter combination “fij”, “ij” is always a vowel. (I did find “Fiji” and “Fijisch” in which “ij” is not a vowel, but these words always start with a capital.)

    I searched for Dutch words that contain “fie”, in which “ie” is not a vowel. I did not find such a word—so I believe in the Dutch letter combination “fie”, “ie” is always a vowel. (I did find “fiësta”, but this word doesn’t qualify because it always contains a dieresis.)

    Perhaps I missed some unusual loanwords in which “ij” or “ie” is not a vowel.

    Kent Lew: Could the Dutch localization be expanded to include ... leaving fi to ligate except when followed by j or e?

    Yes, this would make sense to me.

    When searching for words containing “fij” or “fie”, I consulted the dictionary for which Bram de Does’s Lexicon font has been designed. I noticed that in this dictionary, in all occurrences of “fij” and “fie”, “f” and “i” were ligated—and this didn’t look weird to me. So Artur has a point: in small Dutch text, the “fi” ligature may look OK.

    I don’t like a “fi” ligature in “fijn” and “fiets”. Perhaps there are other Dutch people who don’t have the same gut feeling about this.

    Hrant H. Papazian: So what about a ligature of f with ij-acute?

    According to the European rules for the use of the IJ in public records, in “ij”, for indicating a stressed syllable, an acute accent is placed on the “i”, not on the “j”. So a stressed “fijn” would look like “fíjn”, and the question would be how to make f+í+j look good (or f+íj, if the ligature “íj” would exist in the font involved; or fí+j, if the ligature “fí” would exist).

  • @Ben Blom Good research. I would caution against one particular word in these two passages: "these words always start with a capital" and "it always contains a dieresis"...  :-)
    Ben Blom said:
    According to the European rules for the use of the IJ in public records, in “ij”, for indicating a stressed syllable, an acute accent is placed on the “i”, not on the “j”.
    Yes, this seems to be a rule. That needs to be fixed.
  • John HudsonJohn Hudson Posts: 2,955
    According to the European rules for the use of the IJ in public records, in “ij”, for indicating a stressed syllable, an acute accent is placed on the “i”, not on the “j”. So a stressed “fijn” would look like “fíjn”, and the question would be how to make f+í+j look good (or f+íj, if the ligature “íj” would exist in the font involved; or fí+j, if the ligature “fí” would exist).

    That is contrary to what I am familiar with from the history of Dutch typography and from Brill's editorial style guide, which is that the acute should be on both i and j when that vowel is stressed.
  • What does the LC /ij look like Paul?
  • @Artur Schmal: non-connected, just a plain i-j combination in all IBM Plex fonts.

    Lowercase ligated /ij is of little practical use in text typefaces in my opinion.
    But for playful display typefaces, such as the custom typeface we produced for VPRO, I was happy to drawn one along with special fij versions.

  • Lowercase ligated /ij is of little practical use in text typefaces in my opinion.
    The lowercase "ij" is far more useful than the uppercase, it does't have to be ligated [so literally], and should not look like an accented "y".

  • It would look (and arguably work) better with one spanning accent.
  • Ben BlomBen Blom Posts: 250
    Paul van der Laan: In contrast to some of the anecdotal evidence or obscure sources that are cited in this thread, there is only one official source for spelling of the Dutch language which is de Nederlandse Taalunie (Dutch Language Union). Their rules are followed by the government, schools, newspapers, publishing houses, etc.

    When I consult my Stijlgids Financieele Dagblad [Manual of Style of the Dutch Financial Daily Paper, 1999], I read this about the rules concerning emphasis on two-letter vowels (page 64):

    My translation:

    Two-letter vowels with twice the same letter, get two accents: -éé, -áá, -óó. Two-letter vowels with different letters, get one accent on the first letter: -íe, -éu, -áu, -úi, -óe, -éi, -íj, -óu. (6)
    (6) We deviate from the Green Book, which put an accent on both letters of a two-letter vowel with different letters.

    So the Financieele Dagblad deliberately deviates from the “Green Book”. The Green Book contains the official spelling of the Dutch Language Union. The spelling of the Green Book is only mandatory for government and educational institutions; everyone else may deviate.

    My conclusion: (1) The official rules are described by Paul van der Laan. (2) There is, however, no consensus ‘on the ground’ about using one or two acutes on two-letter vowels with different letters.

  • Should «ie» and «oe» even count as a tweeklank, though? (I guess «eu» is diphthongized enough in practice to count as one...)

    «Dubbele klanken», I love that. :grimace:
  • Ben Blom said:

    So the Financieele Dagblad deliberately deviates from the “Green Book”. The Green Book contains the official spelling of the Dutch Language Union. The spelling of the Green Book is only mandatory for government and educational institutions; everyone else may deviate.

    There's also a "White Book". It's an alternative guide, published by het Genootschap Onze Taal (Google translate: Society Our Language). Their site is – as far as my knowlegde goes – very populair by writers, journalists and teachers. A lot of professionals are using the site as reference.

    https://nl.wikipedia.org/wiki/Spellingwijzer_Onze_Taal
    https://onzetaal.nl
  • John HudsonJohn Hudson Posts: 2,955
    [Seeing all this Dutch has produced a sudden craving for double-salt licorice.]
  • Mmmm. Salmiakki (not a Dutch word, I know...)
  • Paul van der LaanPaul van der Laan Posts: 242
    edited November 2017
    Hrant H. Papazian said:
    It would look (and arguably work) better with one spanning accent.
    If you don’t provide visual examples then I find it hard to take such a claim serious.
  • John SavardJohn Savard Posts: 1,088
    Mmmm. Salmiakki (not a Dutch word, I know...)
    Dubble Zoute Drop is what the Dutch call it... but, yes, no doubt it tastes as sweet by its Finnish name.
  • edited November 2017
    @Paul van der Laan The best test would be to take those three fonts and put a somewhat wide (and probably slightly taller) acute* [visually] centered between "I" and "J" (and "i" and "j"). One interesting resultant question: could the tittles provide an avenue of unifying a [non-accented] "ij"? Not least to avoid the "y" structure.

    * For more adventurous attempts, see this link I posted earlier: http://www.typophile.com/node/34111
Sign In or Register to comment.