Motivations and Methodology

This lexicon is my attempt to organize information about Tolkien’s Elvish languages, Quenya and Sindarin, into a coherent system. It examines Tolkien’s linguistic works as an integrated whole, linking Quenya and Sindarin both to each other and to their ancient past. It also considers how these languages developed over Tolkien’s lifetime, as well as how those languages have been interpreted by later students of his languages, both casual and scholarly. The lexicon has two primary goals:

The portions of this lexicon addressing these two goals are carefully sequestered so that fan-invented words are clearly designated and separated from words created by Tolkien. The academic-oriented sections of the lexicon are limited exclusively to words created by Tolkien himself, or straight-forward extrapolations of those words to help illustrate the conceptual development of those words (such as hypothetical but probable primitive forms). The fan-oriented (Neo) sections are less rigorous and omit some details in order to make the languages more accessible to beginners.

Organizational Scheme

As much as possible, this lexicon is based on primary sources, those written by Tolkien himself. Very little of Tolkien’s linguistic work was published during his lifetime. Fortunately, there is a large body of posthumous publications to work from, without which the creation of this lexicon would be impossible. The authors of these works, most notably Christopher Tolkien and the scholars responsible for Parma Eldalamberon and Vinyar Tengwar (Christopher Gilson, Carl Hostetter, Arden Smith, Bill Welden, Patrick Wynne and others) have already done the extremely difficult work of deciphering much of Tolkien’s unpublished writings and presenting them in a format useful to the general public.

The linguistic elements of Tolkien’s writings form an elaborate web of relationships to one another. To manage these relationships, I’ve organized the material into an XML data model. This data model lets me correlate the information in various ways to examine different linguistic features. Though the data model itself is based on primary sources, I also consulted a number of secondary sources, as indicated in the Reference Index, to help improve my understanding of the primary material.

The atomic elements of this lexicon’s data model are what I call references. A reference is single linguistic form appearing in Tolkien’s writing, each with a unique identifier, generally in the form [book]/[page].[word position], such as: LotR/0305.0604. These references stand in various relationships to one another, including grammatical inflection, etymological derivation, and cognates from different languages. These references are grouped into words or entries, each of which is a candidate for a valid word in one of Tolkien’s languages at some point in time.

In ordinary dictionaries, a word is a concept collecting a set of possible expressions for different grammatical contexts. For example, the English word “woman” conceptually includes both its singular form (woman), its plural form (women) and its possessive form (woman’s). Most dictionaries use words as their organizing principle, listing different forms as variations in the entry for each word. This lexicon must also address the word variations from different conceptual periods of Tolkien’s writings. For example, Tolkien considered several different forms for the word “finger”, including lepsë, lepta, leper in Quenya and lhebed, leber in Sindarin. It is useful to group the entries for these conceptual variations together so that we can evaluate how Tolkien’s ideas about his languages evolved. In general, entries of related words are organized in order of conceptual development, with latest forms first and then working backwards to earlier precursors.

Although I’ve attempted to be rigorous in recording Tolkien’s words, the nature of this organization is necessarily subjective. The boundaries between linguistic elements is rarely clear even in real-world languages, much less the ever-evolving conceptions of Tolkien’s fictional languages. I have done my best to group the various elements together in a useful way. The XML metadata and relationships are critical for this purpose. For example, this metadata allows us to extract the attested past tenses of verbs to compare them with each another, which can be used to formulate theories on how Tolkien intended the past tense of Elvish words to be formed as his ideas on the Elvish verb conjugation changed over time.

Language Sections and Navigation

As discussed in the section on the Conceptual History of Elvish, this lexicon divides Elvish languages into three broad periods: Early (1910-1930), Middle (1930-1950) and Late (1950-1973). Word created by Tolkien are grouped into languages categories by time period, such as Early, Middle and Late Quenya, and Gnomish (Early), Noldorin (Middle) and Sindarin (Late). Noldorin is a bit of a special case, since some of it spans into the Early and Late periods.

As noted above, these time-oriented language sections are reserved almost exclusively for words created by Tolkien himself. Fan-invented word lists are grouped into separate “Neo-Eldarin” languages: Neo-Quenya, Neo-Sindarin and Neo-Primitive Elvish. This is to separate the more rigorous, research-oriented content of the lexicon from the less rigorous fan-oriented content.

To better support this separation, the lexicon has two “navigation modes”:

You don’t need to do anything special to “activate” a particular navigation mode: it depends entirely on which section of the lexicon you are exploring. If you start in the Tolkien-specific language sections, you will be using the default navigation mode, whereas if you start in the Neo-Eldarin sections, you will be using the “neo” navigation mode. The two navigation modes determine how entries are displayed and linked together, and control which results show up in searches. To switch modes, go back to the home page of the lexicon and enter the appropriate section.

Neo-Eldarin in Eldamo

Many of those who use Eldamo data are fans who want to try their hand at writing or speaking in Tolkien’s invented languages. Since these composition are not, strictly speaking, consistent with Tolkien’s own ideas, they are usually designated “Neo” or post-Tolkien compositions. I use the term “Neo-Eldarin” to collectively describe such Neo-Quenya and Neo-Sindarin writing.

As anyone who has carefully studied Tolkien’s languages knows, Tolkien himself never finished developing his languages. This was not his goal; he engaged in language creation as an artistic endeavor, and was more interested in linguistic aesthetics than in producing a “finished” language. Tolkien left us a large collection of linguistic ideas, often contradictory, as his own thoughts on his languages evolved over time. Some fundamental questions (such as how negation was expressed) were never fully resolved. This makes it impossible to fully describe the Elvish languages as Tolkien intended them, because Tolkien himself never decided what they would ultimately be.

Nevertheless, there is a large community of Tolkien fans who wish that he had finished his languages, and strive to find ways to reconstruct or extend Tolkien writings to produce more complete and coherent set of Elvish languages. Given the incomplete and often-contradictory nature of Tolkien’s own writings, there is no universal approach to defining Neo-Eldarin: well-meaning and knowledgeable people may prioritize on different aspects of Tolkien’s languages to produce different but equally-valid Neo-Eldarin dialects.

Eldamo includes a slowly increasing number of fan-invented words for the Elvish language, both my own inventions and the inventions of others. These words are carefully designated as Neo-Quenya (ᴺQ.) and Neo-Sindarin (ᴺS.) and are attributed to the original source or creator if possible. Because there are many possible approaches to creating Neo-Eldarin, these entries reflect my own biases on how Neo-Eldarin should work, though I try to discuss popular alternatives where appropriate.

Eldamo’s Approach to Extending Elvish

To make my biases completely clear, here are the factors I consider in defining Neo-Eldarin.

First and most importantly, I consider it vital to consider the Eldarin languages as a whole: Quenya, Sindarin and its primitive predecessors together. Tolkien didn’t work on the individual languages separately, and the diachronic aspects (language changes over time) were extremely important to him. I think that Neo-Eldarin definitions should consider the whole Elvish language-family, just as Tolkien did.

Second, because Tolkien’s ideas on his languages were constantly evolving, it isn’t possible to incorporate all of his Elvish writings into Neo-Eldarin. Crafting Neo-Eldarin is as much about figuring out which words to exclude as it is about figuring out how to add new words. A big portion of my work on Neo-Eldarin is focused on figuring out what subset of Tolkien’s words can be used as part of an internally consistent language family: to produce a coherent Neo-Eldarin, we are forced to cherry-pick from the words that Tolkien created.

When deciding which words of Tolkien’s words to include and exclude, I generally prefer his later words over his earlier words, but I don’t universally use Tolkien’s latest ideas. Even in his later writings, Tolkien constantly changed his mind and contradicted himself, so determining his “final” thoughts on any particular topic is often impossible. Furthermore, his later ideas often produce a less complete picture than some of his earlier ideas. Tolkien sometimes altered the development of Quenya words without considering or describing its affect on Sindarin (or vice versa): using those late Quenya words uncritically might force us to abandon important Sindarin words if we want to maintain an internally coherent Eldarin language family.

As example of this, see the various words related to marriage derived from the roots √BER, √BES or √BED at various points in Tolkien’s life. Using the latest root √BER would require us to abandon Neo-Sindarin marriage-related words, such as herven “husband”, herves “wife”. Those are only attested as derivatives of √BES. To preserve the integrity of Neo-Eldarin, I think it is better to continue to use the earlier root. Assembling a coherent Neo-Eldarin language-family is a process of thousands of small decisions like this, all of them necessarily subjective.

My own biases make more me comfortable with adapting early words than is the case with some other Neo-Quenya or Neo-Sindarin writers. Some concepts were only described by Tolkien in the early versions of his languages from the 1910-1930 (Early Period), and I think excluding words from this period is too limiting. However, when I reuse such earlier words (1) I only do so if they do not conflict with later words and (2) they can be re-etymologized in a way that makes them consistent with the phonetic character of later versions of the languages. Furthermore, to better reflect the level of acceptance such words have within the Elvish-speaking community, I always mark such as early adaptation as neologisms (ᴺQ. or ᴺS.), even though strictly speaking some of them were created by Tolkien himself.

As is the case with most writers, I am less strict with words adapted from 1930-1950 (Middle Period) in the Neo-Eldarin vocabulary lists. I generally use and include those without further comment, though I still strive ensure they are compatible with the languages as Tolkien conceived of them when The Lord of the Rings was published. In the more rigorous language lists, I always make it clear which time-period of Tolkien’s words originate from.

While I can’t help but make my presentation of Neo-Eldarin a reflection of my own views, I am more than happy to discuss the words I use and the approaches I take with others. I firmly believe that a working language can only emerge through a consensus from its community of speakers. Where there are other popular and well-reasoned ideas about what words should appear within Neo-Eldarin, I am willing to include such words within the Eldamo data set even if I personally don’t agree they are appropriate, though I won’t necessarily recommend their use. The Eldamo data is available under a very open license, so that its data model can conceivably be used to build Neo-Eldarin dialects different from my own.