Non-Latin Scripts

Introduction

An adequate representation of information in non-Latin languages has been one of the strengths of Brill in the “paper period”. The use of online platforms to access our publications still increases. Using PDF as a format for onscreen presentation ensures the same quality that our customers are used to, based on the print. Besides PDF most of the publications are also available in a full text HTML-format. A correct display of non-Latin scripts in HTML format can only be ensured by “pushing” the necessary font(s) together with the webpage. This starts with a proper coding of language and script information in the source xml, as explained hereafter.

The xml:lang-attribute

Both the JATS- and BITS-standard support an attribute @xml:lang. Some important sections from the description of this attribute in the JATS Tag Library:

1. The language of the intellectual content of the element for which this is an attribute.
2. The value of this attribute must conform to IETF RFC 5646 (https://tools.ietf.org/html/rfc5646). For most languages, a primary-language subtag such as “fr” (French), “en” (English), “de” (German), or “zh” (Chinese) is sufficient.
3. In some languages, script codes are also critically important; for example, in Japanese, there is the need to express whether a name is in Kanji as opposed to in Kana (Hiragana or Katakana) to determine sort keys. Best practice is to use the full language-code-plus-script-code as the value for @xml:lang. In our use of both language and script tagging as values for @xml:lang, we are following the IETF (Internet Engineering Task Force) best practice guideline: Network Working Group Request for Comments: 5646 [Tags for Identifying Languages, A. Phillips and M. Davis, Editors, September 2009]. That document defines a language tag as composed of (in part):
   1. A language code Language (typically using the shortest ISO 639)
   2. Potentially followed by a hyphen and then a script code script (using the ISO 15924 code)
4. Thus, for example, the following are among the expected values of @xml:lang for Japanese, incorporating both a language (“ja”) and a script type:
   1. xml:lang="ja-Hira" (Japanese written in Hiragana)
   2. xml:lang="ja-Hrkt" (Japanese written in Hiragana + Katakana)
   3. xml:lang="ja-Jpan" (Japanese written in Han + Hiragana + Katakana)
   4. xml:lang="ja-Hani" (Japanese written in Kanji (Hanzi, Hanja, Han))
   5. xml:lang="ja-Kana" (Japanese written in Katakana)

The @xml:lang can be assigned to most elements of the JATS and BITS standard. A listing of the elements is available in the JATS and BITS documentation.  Thus, if the contents of the element is completely in a particular language, the attribute can be assigned to the element. For example:

1. <abstract xml:lang=”zh-Hans”> for a Chinese language abstract in a simplified Chinese script.
2. <article-title xml:lang=”cop-Copt”> for an article title in Coptic

The <styled-content>-element

Often in Brill publications a few words in a non-Latin script are used in an English- or other language sentence. In this case, the language and script information should only be assigned to these few words and not to the complete element. For this situation, the element <styled-content> should be used(See https://jats.nlm.nih.gov/publishing/tag-library/1.2/element/styled-content.html) . The @xml:lang can be assigned to the <styled-content> element, for example:

1. … instead of <styled-content xml:lang="mid-Mand">&#x0848;&#x0840;&#x0841;&#x0845;&#x0855;&#x0840;</styled-content> &#x1E6D;abuta … For a word in Mandaic in a sentence. Note that the Mandaic characters are coded using their Unicode values.
2. … themselves “<styled-content xml:lang="txg-Tang">&#x18736;&#x17D32;&#x170A7;</styled-content>” [tha dźjwij lhji.j] — the “State of Great Xia”… For a word in Tangut. Here also, the characters are coded using their Unicode values to avoid squares if a font to display the values in the xml-editor is not available.
3. … sage (<styled-content xml:lang="he-Hebr">&#x05DC;&#x05DE;&#x05E9;&#x05DB;&#x05D9;&#x05DC;</styled-content>) to instruct … For a word in Hebrew.

Overview of supported languages / scripts / web fonts

The following table gives an overview of the non-Latin scripts of which we know that they are used in Brill publications. For most of these we have web fonts available, including the instruction for size in relation to the Brill typeface. Please note that because Brill uses language-script tags exclusively to trigger web fonts at this time, language tags may be artificial, as in the case of Aramaic text written in the ‘Hebrew’ square script: in order to simplify the tagging, such text is tagged as ‘he-Hebr’, even though the language tag ‘he’ does not apply to Aramaic, which was and is a language distinct from Hebrew.

brill.com

At brill.com the xml:lang attribute and <styled-content> element is converted to the corresponding html encoding. Furthermore, the CSS includes the declarations of web fonts used. Web font packages for the scripts listed in the table above are available and will thus ensure a proper display of the characters in html, similar to the pdf-format.