Tengwar



The Tengwar are an artificial script which was invented by J. R. R. Tolkien. In his works, the Tengwar script, supposedly invented by Fëanor, was used to write a number of the languages of Middle-earth, including Quenya and Sindarin. However, it can also be used to write other languages, such as English (most of Tolkien's tengwar samples we know of are actually in English). The word tengwar is Quenya for "letters". The corresponding singular is tengwa, "letter".

Internal history and terminology
According to HoME 11 (Appendix D to Quendi and Eldar), Fëanor when he created his script introduced a change in terminology. He called a letter, i.e. a written representation of a spoken phoneme (tengwë) a tengwa. Previously, any letter or symbol had been called a sarat (from *sar "incise"), especially the alphabet of Rúmil of Valinor on which Fëanor supposedly based his was known as Sarati, but became later also known as "Tengwar of Rúmil". The plural of tengwa was tengwar, and this is the name by which Fëanor's system became known. Since, however, in commonly used modes, an individual tengwa was equivalent to a consonant, the term tengwar in popular use became equivalent to "consonant sign", and the vowel signs were known as ómatehtar. By loan-translation, the tengwar became known as tîw (singular têw) in Sindarin, when they were introduced to Beleriand. The letters of the earlier alphabet native to Sindarin were called cirth (singular certh, probably from *kirte "cutting", and thus semantically analogous to Quenya sarat). This term was loaned into exilic Quenya as certa, plural certar.

Precursors
The sarati, described in Parma Eldalamberon 13, a script developed by J. R. R. Tolkien in the late 1910s, anticipates many features of the tengwar, especially the vowel representation by diacritics (which is found in many tengwar varieties), different tengwar shapes and a few correspondances between sound features and letter shape features (though inconsistent).

Even closer to the tengwar is the Valmaric script, described in Parma Eldalamberon 14, which J. R. R. Tolkien used from about 1922 to 1925. It features many tengwar shapes, the inherent vowel found in some tengwar varieties, and the tables in the samples V12 and V13 show an arrangement that is very similar to the one of the primary tengwar in the classical Quenya "mode".

Jim Allan (An introduction to Elvish, ISBN 0-905220-10-2) compared the tengwar with the Universal Alphabet of Francis Lodwick of 1686, both on grounds of the correspondance between shape features and sound features, and of the actual letter shapes. A corresponance between shape features and sound features is also found in the Korean Hangul alphabet. It is not known whether Tolkien was aware of these previous scripts. However, considering the sarati and the valmaric script, it is conceivable that Tolkien developed the idea of a general correspondance between shape features and sound features by himself.

The tengwar
The tengwar were probably developed in the late 1920s or in the early 1930s. The Lonely Mountain Jar Inscription, the first published tengwar sample, dates to 1937 (The Hobbit, most editions). The full explanation of the tengwar was published in Appendix E of The Lord of the Rings in 1955.

The Mellonath Daeron Index of Tengwar Specimina (DTS) lists 67 known samples of tengwar by Tolkien.

There are only few known samples predating publication of The Lord of the Rings (many of them published posthumously):
 * DTS1 The Lonely Mountain Jar Inscription, published 1937
 * DTS 22 - Ilbereth's Greeting from The Father Christmas Letters, dating to 1937
 * DTS 50/51 - Edwin Lowdham's Manuscript from The Notion Club Papers has Old English language text written in tengwar (with a few Adûnaic and Quenya words), dating to 1945/6.
 * DTS 10 - The Brogan Tengwa-greetings, appearing in The Letters of J. R. R. Tolkien, No. 118, tentatively dated to 1948.
 * The following samples presumably predate the Lord of the Rings, but they were not explicitly dated: DTS 16, DTS 17, DTS 18 - Elvish Script Sample I, II, III, with parts of the English poems Errantry and Bombadil, first published in the Silmarillion Calendar 1978, later in Pictures by J. R. R. Tolkien, as well as DTS 23 - So Lúthien, a page of the English Lay of Leithan text facsimiled in HoME 3:299.

A few other samples, e.g. a tengwar mode for Gothic are known to exist, but remain unpublished to date.

Modes
Just as with any alphabetic writing system, every specific language written in tengwar requires a specific orthography, depending on the phonology of that language. These tengwar orthographies are usually called modes.

Some modes, called ómatehtar (or vowel tehtar) modes, represent vowels with diacritics called tehtar ("signs"; corresponding singular: tehta, "sign"), while other modes, called full writing modes, represent vowels by normal letters. Some modes map the basic consonants to, , and , while others use them to represent , ,  and. Some modes follow pronunciation, while others rather follow traditional orthography. The "full writing" modes are sometimes called Beleriandic modes because a well-known "full writing" mode is called the "mode of Beleriand".

Since the publication of the first official description of the Tengwar at the end of The Lord of the Rings, others have created modes for other languages such as English, Spanish, German, Esperanto and Lojban.

Tengwar letters
The most notable characteristic of the tengwar script is that the shapes of the letters correspond to the features of the sounds they represent.

The letters are constructed by a combination of two basic shapes: a vertical stem (either long or short) and either one or two rounded bows (which may or may not be underlined, and may be on the left or right of the stem).

The principal letters are divided into four series ("témar") that correspond to the main places of articulation and into six rows ("tyeller") that correspond to the main manners of articulation. Both vary among modes.

Each series is headed by the basic signs composed of a vertical stem descending below the line, and a single bow. These basic signs represent the voiceless stop consonants for that series. For the classical Quenya mode, they are, , and , and the series are named tincotéma, parmatéma, calmatéma, and quessetéma, respectively; téma means "series" in Quenya.

In rows of the general use, there are the following correspondences between letter shapes and manners of articulation:
 * Doubling the bow turns the voiceless consonant into a voiced one.
 * Raising the stem above the line turns it into the corresponding fricative.
 * Shortening it (so it is only the height of the bow) creates the corresponding nasal. It must be noted though that in most modes, the signs with shortened stem and single bow don't correspond to the voiceless nasals, but to the approximants.

Here is an example from the parmatéma (the signs with a closed bow on the right side) in the general use:
 * The basic sign (with descending stem) represents (it happens to look much like the letter p).
 * With the bow doubled, it represents.
 * With a raised stem, it represents.
 * With a raised stem and a doubled bow, it represents.
 * With a short stem and double bow, it represents.
 * With short stem and single bow, it represents.

In the classical Quenya mode, the rows are used differently:
 * With descending stem and double bow, we have.
 * With raised stem and double bow, we have.

There are additional letters that don't have regular shapes. They may represent e.g., , and. Their use varies considerably from mode to mode. Some aficionados have added more letters not found in Tolkien's writings for use in their modes.

Non-Unicode
The contemporary de facto standard in the tengwar user community maps the tengwar characters onto the ISO 8859-1 character encoding following the example of the tengwar typefaces by Dan Smith. This implies a major flaw: If no corresponding tengwar font is installed, an awful string of nonsense characters appears.

Since there is not enough place for all the signs, certain signs are included in a "tengwar A" font which also maps its characters on ISO 8859-1, overlapping with the first font.

For each tengwar diacritic, there are four different codepoints that are used depending on the width of the character which bears it.

Other tengwar typefaces with this encoding include Johan Winge's Tengwar Annatar, Måns Björkman's Tengwar Parmaite, Enrique Mombello's Tengwar Élfica or Michal Nowakowski's Tengwar Formal (note that most of these differ in details).

The following sample shows the first article of the Universal Declaration of Human Rights written in English, according to the traditional English orthography. If no tengwar font is installed, it will look awful since the corresponding ISO 8859-1 characters will appear. It should look similar to the picture at the top of the page.

j#¸ 9t&5# w`Vb%_ 6EO w6Y5 e7~V 2{( zèVj# 5% 2x%51T`Û 2{( 7v%1+º 4hR 7EO 2{$yYO2 y4% 7]F85^ 2{( z5^8i`B5$i( 2{( dyYj2 zE1 1`N ]Fa 4^(6 5% `C 8q7T1T W w74^(69~N2º

Unicode
A proposal has been made to include the Tengwar in the Unicode standard.

The following Unicode sample is meaningful when viewed under a typeface supporting Tengwar glyphs in the area defined in the Tengwar proposal for the ConScript Unicode Registry (U+E000–U+E006F; see External links).

At the moment, the only typefaces that support this proposal are fonts by James Kass such as: Code2000, Code2001 and Code2002.

The following sample repeats the above one according to the Unicode proposal. It will only display correctly if either of James Kass's fonts is installed.

                             

Modes

 * Tengwar mode for English
 * ESR's Tengwar modes for Esperanto and Lojban languages
 * Tengwar mode for Spanish
 * Tengwar mode(s) for German
 * Tengwar mode for Hebrew (PostScript format)
 * Tengwar mode for Arabic

Technical

 * WikiTeX supports editing Tengwar directly in Wiki articles.
 * Official proposal to encode Tengwar in Unicode
 * Tengwar proposal for ConScript Unicode Registry
 * Critique on the official proposal to encode Tengwar in Unicode