Lord of the Rings: The Animated Musical | Appendix E: Writing & Spelling, I Pronunciation
The first part of “Appendix E: Writing and Spelling”, focused on pronunciation. I am notoriously bad at pronouncing words that I only know by sight, some of my favorites being liaison, epitome, Yosemite, and mischievous. Because I learned most of my vocabulary from reading, I rarely heard more difficult words pronounced aloud until high school or college, when I learned the hard way that what I heard in my head was not the generally accepted way to say the word. I tended to think my linguistic snafus were pretty funny.
This difficulty with language learning and pronunciation extended to the fictional world of Middle-earth. Tolkien noted in a footnote that most hobbits had trouble pronouncing their only language of Westron compared to what the Rohirrim expected, and when it came to learning Elvish languages like Sindarin, most English speakers “will err little more than Bilbo, Meriadoc, or Peregrin. Frodo is said to have shown great ‘skill with foreign sounds’” (The Return of the King, 438). Throughout the text, translators of other Races were impressed by his ability to not only learn languages but to speak these like a native.
Because of this difficulty, Tolkien included an extensive pronunciation guide to aid readers in correctly saying the names of places and people. He even wrote, “I have tried to represent the original sounds (so far as they can be determined) with fair accuracy” (Return, 435), which seems to be Tolkien’s academic sense of humor, since he made up the languages and the stories in which they appear. Unfortunately, Tolkien did not use the modern pronunciation system of the International Phonetic Alphabet (IPA), which I will describe in greater detail later in this essay. He preferred instead to use common words and phrases from British English to explain the sounds; perfectly logical for a professor teaching undergraduates in the 1950s, but less practical for those of us unfamiliar with his accent, although plenty of recordings exist online. Fortunately, modern nerds have already put in the work and shared their results through wikis such as the “Sindarin” page on Tolkien Gateway.
The International Phonetic Alphabet and Its History
With so many languages and writing systems in the world, one would think comparing them would be impossible. Academics are drawn to the seemingly impossible, and thus came the invention of the IPA, a system to turn phonemes or single sounds into a standardized chart of characters. The International Phonetic Association currently governs this chart and includes helpful resources on its website for linguists of all levels.
The first version of a phonetic alphabet came onto the linguistics scene in 1888 in the academic journal Maître Phonétique [Master Phonetic]. This French-language publication compared sounds made in English, French, German, and sporadically in other languages including Italian, Spanish, Swedish, Danish, and Flemish. While this comparison was a novel concept, the chart was still incomplete, and it was only tested in Western European languages.
The first chart appeared two years later in 1900. Consonants stood on the top half of the chart, while vowels stood on the bottom. Row labels were based on whether the glottis, or the gaps between vocal cords, was open, closed, or somewhere in between. Column labels were based on where the sounds were located in the mouth, from Labiales or the front near the lips to Laryngales and Gutturales or the back of the throat. Tolkien used the colloquial term “guttural” to describe his fictional language of Khuzdul, which included many of the sounds found in Germanic, Semitic, and some Celtic languages. The IPA changed this term to “Bronchs” in the 1904 English edition and later sorted sounds from that column into different categories. Gutturales appeared in the 1907 French Edition, but the term was gone for good by the 1912 edition.
Tolkien continued to use the term throughout his writing and interviews when it was no longer considered academic, even though versions of the IPA without the term were available during his college career, which lasted from 1911 to 1915. However, his interest in linguistics predated that time, as did his vocabulary. In Tolkien: A Biography, author Humphrey Carpenter cited A Primer of the Gothic Language by Joseph Wright as a favorite grade school read; personally, I was in my B.F. Skinner phase at that time. The format of this text showed striking similarity to the format of Appendix E, and pronunciations in several Gothic languages were described as “guttural”. While other scholars not discussed further here have accused Tolkien of antisemitism due to using this terminology when describing his Semitic-inspired dwarfish language, he more likely used the word in reference to a favorite childhood book.
The Great Vowel Shift
The study of how languages change over time falls under the academic area of historical linguistics, a combination of two of my favorite things: history and words. The Great Vowel Shift in English was identified by Danish linguist Otto Jespersen who loved English grammar. The shift began in the 15th century and ended by the 18th century, turning the Middle English used by Geoffrey Chaucer, author of Canterbury Tales, into Early Modern English used by playwright William Shakespeare and translators for the original edition of the King James Bible. This change in the way vowels were pronounced is a reason why some of Shakespeare’s verses do not rhyme anymore.
Vowels in Middle English were pronounced the same way as vowels in most other Western European languages, including Latin. A, E, I, O, and U corresponded to the IPA values of / a:/, /e/, /i/, /o:/, and /u:/; the same as the sound in ‹father›, ‹date›, ‹meet›, ‹goat›, and ‹lute›. However, for unknown reasons, Middle English speakers collectively decided they wanted more vowel sounds, and the shift began.
The shift in back vowels, or those pronounced near the back of the mouth, began with /ɔ/, called the “open-mid back rounded vowel”. This sound still appears in the General American English accent for some speakers in the word ‹caught›, at least for those whose accent has not experienced another phenomenon, cot-caught merger, where the similar /ɑ/ sound is used instead. The artificial intelligence voice that I use to proofread my articles, Microsoft English (United States) Female, apparently has cot-caught merger. I tested this on my other favorite voice, Microsoft English (United Kingdom Female), and sure enough, her ‹cot› and ‹caught› do not merger.
When the /ɔ/ sound moved into the /o:/, from ‹caught› to ‹coat›, it produced a chain reaction: /ɔ/ → /o:/ → /u:/ → /ɑu/, or ‹caught› → ‹coat› → ‹coot› → ‹cow›. The central and front vowels also shift: /ɑ:/ & /æ:/ → /e:/ → /i:/ → /ɑi/, or ‹father› & ‹bag› → ‹bait› → ‹beet› → ‹brine›. Some readers might ask why I used ‹father› instead of ‹bother›, and the two reasons are that I pronounce these vowels slightly differently, ‹father› as /ɑ:/ and ‹bother› as /ɑ/, but many people with General American English accents use /ɑ/ for both, which is the same sound they would use for ‹cot› and ‹caught›. The vowels continue to shift and cause confusion for anyone studying English, even native speakers.
Other Language Shifts
Shifts in pronunciation are not particular to English. A shift in sounds appeared in Canaanite dialects, which include modern Semitic languages like Hebrew and Arabic. Tracking this shift was complicated by the lack of characters for some or all vowels in the written version of these languages; instead, diacritics or marks around the characters are sometimes used to indicate these sounds, but not always. The original sound /ɑː/ as in ‹father› remained the same in Arabic but became the sound /oː/ as in ‹coat› for Hebrew. Many cognates, or words in two different languages with a similar sound and meaning, display this shift. Using common romanization of the languages, which means using the Latin alphabet to represent the sounds of the words, ‹priest› is kahin or kohen, ‹daughters› are banat or banot, ‹queens› are malikat or malakot, ‹peace› is salam or shalom, and head is ras or rosh. Similar shifts appeared in neighboring languages like Phoenician and Ancient Greek.
Hochdeutsch or High German languages underwent a consonant shift instead of a vowel shift. This was what made Old Saxon, an ancestor of English, sound different than Upper German, a dialect spoken by the upper class in the lower part of the country. This change seemed to have happened before surviving writing of German, which dates from the 8th century. For one common shift, the sound /t/ became /s/ or /z/. Tehan in Old Saxon became zëhan in Upper German and ‹ten› in English; herta became herza and ‹heart›. Another shift was /p/ to /f/: appul to apful and ‹apple›, helpan to helfan and ‹help›. Finally, /k/ shifted to /x/, the sound in Bach: korn to chorna and ‹corn›, ackar to acchar and ‹acre›.
For an ongoing vowel shift, I do not have to look far. North American English is currently undergoing its own changes. The United States is known as a country with a myriad of accents and dialects differentiated by location, class, race, and native language. While the internet is at war over whether the Boston accent is disappearing, among many other things, one thing actual linguists agree on is that the dialects have historically used a wide range of vowel pronunciation.
Remember cot-caught merger I mentioned during The Great Vowel Shift? This has long showed up in Boston but now moved over to the Midwest and gone a step further. A vowel shift for Americans living in the North likely began by the 1930s but has been studied since 1972, when linguist William Labov of the University of Pennsylvania and his team coined the term, Northern Cities Vowel Shift (NCS). Labov even wrote a book about it, Atlas of North American English Phonetics, Phonology and Sound Change, with the first edition in 2006 and the second in 2012.
The change in vowels was originally an urban phenomenon among White speakers, hence the word ‹cities› in its name, but has spread to small towns and rural areas. Black and Latino communities have not adopted this new pronunciation and continue to speak in distinct accents. The shift is not consistent across the region but generally flows /ɪ/ → /ɛ/ → /ʌ/ → /ɔ/ → /o/ → /a/ → /ɪ/, or ‹bit› → ‹bet› → ‹but› → ‹caught› → ‹cot› → ‹cat› → ‹bit›. In another linguistic change unrelated to the vowel shift, younger Americans no longer pronounce the ‹h› that appears after a ‹w›, so that ‹which› and ‹witch› sound identical. In my own travels, I have noted that adults over seventy living on Cape Cod pronounce the ‹h› in ‹white› and ‹what›.
Pronunciation Highlights
In the interest of not repeating the entire appendix, I have selected a few of my favorite pronunciation guidelines to share, as some choices in romanization made by Tolkien were counterintuitive. For example, in Elvish languages like Sindarin and Quenya, the letter ‹c› was always pronounced as hard ‹c› or /k/, not soft ‹c› or /s/. In all Middle-earth languages, the letter ‹r› was trilled, meaning that some part of the mouth vibrated as it was spoken. In Elvish and Mannish languages, the trill was at the tip of the tongue, similar to the ‹r› in Spanish words like ‹rojo› meaning “red”. Some speakers of Khuzdul, the language of the Dwarves, used a uvular ‹r› pronounced in the back of the throat, as found in the Yiddish word ‹brik› meaning “bridge”. The Elves apparently found this “distasteful” (Return, 436). With this information in mind, the names Celeborn and Celebrían were pronounced /kelebɒrn/ and /kelebrian/ each with a trilled ‹r›.
Tolkien made another interesting romanization choice when deciding how to spell the /x/ sound as in the Scottish word ‹loch›. For Elvish spellings, he used ‹ch› as seen in ‹yrch› meaning “orcs”, while Dwarfish words used ‹kh›, including Khazad-dûm, Khuzdul, and Kheled-zâram. This is similar to romanization practices used for Hebrew and Arabic, as Hebrew words are traditionally spelled with ‹ch› as in ‹chaim› meaning “life”, as seen in the name of the author Chaim Potok, but Arabic words are spelled with a ‹kh›, like the name of polymath Omar Khayyam. Mannish languages had “weakened” the sound, so the country name Rohan used /h/ instead of /x/. The past pronunciation was apparently Rochann, and before that Rochand (Return, 437). Like me, Tolkien must have had a lot of time on his hands to think of this.
Finishing up with pronunciations, the letter combination ‹dh› make the sound /ð/ found in “these clothes”, which sounds like a Bilbo phrase with his “whole rooms devoted to clothes” (The Hobbit, 1), and making Caradhras fairly difficult to pronounce for English speakers. ‹G› is always hard /g/ even in front of an ‹i› as in Gildor and Gimli. The ‹i› in the front of words becomes a /j/ as heard in the English word ‹your› and the name ‹Ioreth›, the eldest of the healers in Minas Tirith. I also learned that a little hat over a letter, like /û/, is called a circumflex and is intended to make the vowels pronounced longer.
The only other comment I had was on stress, or what syllable of the words were emphasized. Tolkien wrote: “it falls on the last syllable but one, where that contains a long vowel, a diphthong, or a vowel followed by two (or more) consonants. Where the last syllable but one contains (as often) a short vowel followed by only one (or no) consonant, the stress falls on the syllable before it, the third from the end.” Got all that? Me either. Some examples given were the names Denethor (DE-ne-thor) and Ecthelion (ecth-EL-i-on) along with the word periannath (pe-ri-AN-nath) meaning halfling (Return, 439).
Conclusion
This section is both a boon and a bane for voice actors, and yet it rarely seems to be used. Past versions of The Lord of the Rings have leaned towards the accents of the British Isles in their vocal coaching despite clear evidence that the characters sounded different. One technical way to create a diverse array of character voices and accurately portray each would be to determine their sounds in IPA and then write practice sentences in IPA until the accent was mastered. The easiest character to learn might be Pippin, as he spoke one dialect of one language and never fully spoke another. The most difficult, and one with some of the most dialogue, would be Frodo with his chameleon-like ability to perfectly match other accents, molding his voice into whatever the others in the conversation might want to hear, in addition to his “good voice” trained for singing (The Fellowship of the Ring, 181).
Also difficult might be Aragorn with his multiple names and matching personas. His code-switching, or changing between multiple dialects and languages, was gradual towards the beginning of the text. He used modern-sounding common language as Strider with occasional breaks into more academic language when describing literary concepts, but he used heightened language as the radiant Aragorn son of Arathorn. His code-switching grew more severe later in the text as he incorporated his past persona of the warrior Thorongil and his new persona of the king Elessar II Telcontar into the mix. Among his most complex switches came as he found a seedling of the White Tree of Gondor and said, “Yé! Utúvienyes! I have found it! Lo! here is a scion of the Eldest of Trees!” (Return, 270), jumping from Númenórean Sindarin to common Westron to heightened Westron in three phrases. At least he felt comfortable being himself in his many varieties.
Future posts will focus on the visual and audio design of individual characters, using historical costumes, as previously seen in posts on the Races, and historical linguistics to make characters who resemble figures from real-world history. While no creator can exactly replicate what Tolkien had in his head, I hope by studying the original text along with supplemental information from historical records will create fully rounded character designs that are respectful to the original intention of the author.