This is the first of two sections on Tengwar writing in Quenya, covering basic consonants and vowels. The remaining details are covered in Chapter 5, §5.1.
4.1.1.1 Simplified Tengwar Chart: Recall that the basic consonants of Quenya are:
Most of these can be represented by a single tengwa character, the exceptions being d, b, g which can only be part of consonant clusters and hr, hl which are represented by a tengwar pair. The tengwar for d, b, g; hr, hl will be discussed in Chapter 5, §5.1. The remaining consonants appear in the simplified tengwar chart below, along with two more sounds that are represented by a single tengwa: qu and ñw (later pronunciation nw). The numbers in the chart are the row and column numbers from the full tengwar chart in The Lord of the Rings Appendix E:
| I | II | III | IV | |
| 1. | t t | p p | c c | qu qu |
| 3. | þ s [þ] | f f | h -h- | hw hw |
| 5. | n n | m m | ñ n [ñ] | ñw nw [ñw] |
| 6. | r -r | v v | y y | w w |
| 7-8. | 7 r- | l l | s s̄ | i ṡ |
| 9. | 9 h- | hy hy | l -i | . -u |
Each column in the chart above is a “series” (téma) and each row is a “grade” (tyellë). Consonants in the same téma or tyellë have similar sounds, at least in the classical version of the language.
Two of the above tengwar are classical prononciations still represented in spelling: þ, ñ vs. modern pronounciations s, n (and classical ñw vs. modern nw). For example, nésa “sister” is spelled with þ because its classical pronunciation was †néþa or †nétha, so its modern spelling is still 5~V3D and not **5~ViD because its spelling reflects its older pronunciation (é = é and C = a, as will be discussed in section 4.1.2 Tengwar Vowel Tehta).
There are three sounds h, r, s that are represented by different tengwar depending on where in the word they appear, as indicated by h-, r-, ṡ vs. -h-, -r, s̄. The exact rules vary by character:
The l and . represent -i and -u only where they appear after another vowel in a diphthong, such as ai lD or au .D. This will be discussed in the section on vowel tehta.
Finally, two of the sounds in the chart above are not unitary tengwar: y = y and hy = hy. These have a pair of under-dots Ï indicating an added “y” sound: hÍ and 9Í. These underdots (y-tehta) will be discussed in Chapter 5, §5.1.
You can also map from Latin letters used for Quenya spelling (the English alphabet comes from Latin like many European languages) to the associated tengwa letter. The resulting list would be:
4.1.1.2 Advanced Topic: Featural Aspects of Tengwar: The tengwar are a featural writing system, where the shape of the letter indicates the sound of the word. Korean Hangul is a good example of a real-world featural writing system. To understand the featural aspect of tengwar, you need to know a bit about how linguists classify sounds. The tengwar chart is arranged into rows and columns. Each column (téma) represents the “place of articulation”, the location in the mouth where the sound is made. Each row (tyellë) represents the “method of articulation”, the way in which the sound is made.
For example, row 1 represents voiceless stops: p, t, k. Voiceless stops are made by briefly interrupting (stopping) the flow of breath and voiceless means the vocal cords do not resonate: compare voiceless p, t, k and voiced b, d, g. The columns represent the location where the tongue interrupts the sound: labial (near the lips) p, dental (near the teeth) t, velar (at the back of the mouth) k and labialized velar kw (qu), which is a velar with lip-rounding combined with a w sound.
Likewise, row 5 represents nasal sounds: m, n, ñ, ñw. Here the place of articulation is the same: labial, dental, velar and labialized velar. The method of articulation is different, though, since the sound passes through the nose. Compare the tengwar for these two rows and you should notice a pattern:
| 1. | t t | p p | c c | qu qu |
| 5. | n n | m m | ñ n [ñ] | ñw nw [ñw] |
Most tengwar are made up of two elements:
The voiceless stops all have a long-down stem (~) and a single bow (6), while the nasals all have a short stem (`) and a double bow (5). Meanwhile, all the dental sounds have a right-open bow (6), labials right-closed (y), velars left-open (h), and labialized velars left-closed (n). Thus the combinations of features tells you the sound of the tengwa:
The tengwar as they are used for modern Quenya are not a perfect featural system, however, since the features were assigned to the sounds Quenya had in the early classical period, when the tengwar were invented. The sounds of the language have changed over the centuries, and the features no longer match the letters perfectly. What was originally the velar nasal ñ (ñ) is now pronounced as a dental nasal (n) at the beginning of words. Row three tengwar (þ f h hw) used to be a set of fricatives or spirants, pronounced “þ, f, ch (as in Bach), hw”, but the pronounciation “þ” became “s” and the pronunciation of “ch” became a simple breath “h” except in the combination ht.
Row six (6 y h n) is even worse. It used to be a set of voiced fricatives or spirants pronounced “ð (the voiced counterpart of þ), v, ʒ (IPA [ɣ]), ʒw (IPA [ɣʷ])”, but only v survives in the modern form of Quenya. The sound “ð” came to be pronounced like “r” but only in the middle and end of words, which is why 6 is frequently used for r in the interior of words but r is represented by 7 at the beginning. The sound “ʒw” became “w” and the sound “ʒ” vanished entirely, leaving h with no sound at all. Since h came to represent “no sound”, the y-tehta Ï was added to represent a simple “y” sound: hÍ.
All of this indicates the depth at which Tolkien thought about his Elvish languages. Not only did he invent an elaborate script for writing his languages, he also decided how that script evolved over time as the sounds of the languages changed.
For more information on the history of individual tengwar letters, a good resource is the Amanyë Tenceli website: http://at.boktypografen.se/teng_quenya.htm.
4.1.2.1 Tehta for Short Vowels: Unlike the Latin alphabet, tengwar does not represent a vowel as an independent letter, but rather as a diacritic mark above the preceding consonant called a tehta. The tehtar for the five Quenya vowels are:
The tengwa ` is used with vowels, but is not the vowel itself. It is the “short vowel carrier”, used when no other consonant is available to carry the vowel: a. However, typically it is a consonant that “carries” the vowel: ta (ta). In Quenya, the tehta is put above the preceding consonant. For example, in the word man “who”, the a-tehta is put above the m (m) and not the n (n): man and not **mna. This works best for Quenya because it is common to have vowels at the end of words, and each vowel can be put over the consonant that precedes it: tanomë “there” = tanomë. To read Quenya tengwar, you go through an up-diagonal-down zig-zag through the consonants and vowels of the word.
The short carrier ` becomes required when there is no consonant to carry the vowel, which can happen when:
In this last example, note how the i-tehta appears above the t (t) but the e-tehta goes over the short carrier ` because there is no preceding consonant. This does not happen when the vowel pair is one of Quenya’s six diphthongs: ai, oi, ui; iu, eu, au. In this case one of two diphthong carriers is used: ai = lD, au = .D. See the next section for discussion.
4.1.2.2 Tehta for Long Vowels and Diphthongs: The short carrier ` is only used for short vowels. There is a separate long carrier ~ used for long vowels. Unlike short vowels, the vowel-tehta always goes on the long carrier even if there is a preceding consonant, to indicate the vowels length: nórë “land” = 5~N7R. Thus the five Quenya long vowels are:
As mentioned above, the six Quenya diphthongs get special treatment. The three i-diphthongs ai, oi, ui place a vowel tehta above the i-diphthong carrier l, that is: lD, lH, lJ. The three u-diphthongs iu, eu, au place a vowel tehta above the u-diphthong carrier ., that is: .G, .F, .D. Thus the six Quenya diphthongs are:
In the section on 4.1.1 Simple Tengwar Consonants, we said that the s-tengwa was i before short vowels that required a vowel-tehta and 8 otherwise. This means i is often used at the beginning of words, but not always. Compare the following:
Note how silë is spelled with i but síla and saila are spelled with 8 because the following vowel is either long or a diphthong. In such cases the vowel tehta is placed on the long or diphthong carrier, not on the s-tengwa.
Here are some spelling examples to demonstrate tengwa consonant use. Remember that sometimes different tengwar are used in different places in the word. Thus r uses 7 at the beginning of words and before vowels, but 6 at the end of words and before consonants. Meanwhile s uses i when tehta are added but 8 otherwise. Finally h uses 9 at the start of words and d in the interior of words.
Also note that some consonant clusters have special tengwar of their own, such as mb for mb and ld for ld; we will discuss those tengwar in Chapter 5, §5.1.
Here are some spelling examples to demonstrate vowel-tehta use:
The tengwar letters have names, which also happen to be Quenya words and usually (but not always) are spelled with the given tengwa. Here are the names of the tengwar discussed in this section, using the numbering scheme from The Lord of the Rings Appendix E. Some numbers are missing for the tengwar that we will be discussing in Chapter 5, §5.1.
Other new vocabulary mentioned in this section:
Much of the information above comes from The Lord of the Rings Appendix E, with additional information drawn from other documents, notably the Outline of Phonology from the early 1950s (PE19/68-107) and some notes Tolkien wrote on The Feanorian Alphabet from the 1930s and 40s (PE22/6-53). This last document reflects Tolkien’s earlier ideas of the Elvish languages, but nevertheless gives useful insights on how he thought the writing systems of Elvish evolved.
Some of the information above is also drawn from actual writing samples of tengwar, the most notable being Tolkien’s rendering of the Namárië poem in tengwar in The Road Goes Ever On, a book Tolkien published in 1967 in collaboration with Donald Swann putting many of Tolkien’s poems to music. Tolkien experimented with various ways to write Quenya using tengwar, and the system described in this course is largely based on the system he used for the Namárië poem.
There are other ways to write tengwar, though, far too many for an introductory course like this. Måns Björkman’s web site Amanyë Tenceli (http://at.boktypografen.se/teng_quenya.htm) is the best available resource on tengwar usage, so I recommend looking there if you want to learn more.
This chart shows the tengwar we’ve learned so far:
| I | II | III | IV | |
| 1. | t t | p p | c c | qu qu |
| 3. | þ s [þ] | f f | h -h- | hw hw |
| 5. | n n | m m | ñ n [ñ] | ñw nw [ñw] |
| 6. | r -r | v v | y y | w w |
| 7-8. | 7 r- | l l | s s̄ | i ṡ |
| 9. | 9 h- | hy hy | l -i | . -u |
Here are the vowel tehtar, on short (`) and long (~) carriers:
The six Quenya diphthongs are:
The tengwar exercises in this course are optional. You can understand and use Quenya as a language without learning its writing system. In fact, most people who write Quenya do so with Latin letters, simply because they are so much easier to type. This course doesn’t teach you how to type tengwar, which is also quite complicated and depends on which tengwar fonts you use. It assumes if you are going to do tengwar exercises, you will do them by hand.
Many people do enjoy using tengwar for Elvish writing, though, especially when writing by hand, so the course includes some exercises to let you practice it if you want. You could, for example, skip all the tengwar exercises your first time through the course, then come back and redo those exercises later for extra practice.
For this exercise, transcribe the following Quenya phrases to tengwar, and then translate them to English. Pay special attention to (a) words that use letters from their archaic pronunciation and (b) letters for sounds that are different based on their position in the word.
Note ¹: You might wonder why this sentence uses mi taurë rather than the more usual tauressë. That’s because we haven’t learned how to write ss in tengwar yet.
Answers are in Answer Key 4.1 at the end of this chapter.
This section discusses the pluralization rules for nouns, adjectives and verbs.
4.2.1.1 Plurals and Noun Classes: Quenya nouns may be singular (one) or plural (multiple). Nouns form plurals based on their noun class. There are three major noun classes in Quenya based on how they form plurals:
The class of consonantal nouns has more variety in its plural forms, since these nouns often have stem forms different from the base noun, for example: nér (ner-) “man [male]” and nís (niss-) “woman”. Their plurals would be: nér “man [male]” → neri “men [males]”, nís “woman” → nissi “women”. Some of the variations between singular and plural forms can be quite dramatic, as in nelet “tooth” → nelci “teeth”.
Note that there are complexities hidden behind this system of three noun classes (vocalic, e-noun, consonantal) in the same way that there are complexities hidden behind our system of three major verb classes (basic, a-stem, u-stem). We will talk more about noun classes in Chapter 11, Section §11.3, as well as touching on them here-and-there as “alternate” formations throughout the course, starting with the next subsection on wi-plurals.
Also note that it is frequently useful to group nouns by which plural form they use: the i-plural or the r-plural. The plurals of most noun cases depend only on which plural form is used, not on the base noun class.
4.2.1.2 Alternate wi-plurals: There are a few nouns ending in o with stems ending in -u which form alternate plural endings using -wi:
These nouns retain their ancient plural forms rather than forming plurals with -r, so angwi “snakes” not **angor. In fact, there are a few nouns without u-stems which properly form plurals using -r that can instead use -wi, such as orco “orc” which has a regular plural form orcor but also an alternate form orqui by analogy with words like ranco → ranqui (WJ/390). Note that these wi-plurals do not appear for nouns whose singular forms end in u: thus Ainu “angel” → Ainur “angels”, not **Ainwi (Ety/AYAN). It also doesn’t apply to o-nouns whose stem also ends in o: axo “bone” → axor “bones” (MC/222).
4.2.1.3 Alternate er-plurals: There are a few nouns ending in e whose plurals use -er instead of -i. Many of these nouns have singular forms ending in -ie or -le:
The ie-nouns use er-plurals because changing the final e to i would result in a double-i, which would be hard to pronounce. The le-nouns use er-plurals because otherwise their plural form would look too much like the partial or “partitive” plural -li discussed in the next section. There are a few other e-nouns that use er-plurals, but there seems to be no pattern to these; likely they are examples of the beginning of a transition for all e-nouns to er-plurals, interrupted by Quenya effectively becoming a dead language at the end of the First Age.
4.2.1.4 Total versus Partial Plurals: For definite nouns (preceded by i “the”), Quenya plurals have the meaning that you expect them to have: singular for one individual (i elda “the elf”) and plural for a group of individuals (i eldar “the elves”). For indefinite nouns (without i “the”), the implication of the plural form is that of a total plural, referring to all possible members of that class. So the phrase eldar marir tauressen “elves dwell in forests” implies all elves live in forests, not just some unspecified group of elves.
Quenya has a separate plural suffix for a partial or “partitive” plural: -li. It can be translated into English with the word “some”. Thus “(some) elves live in the forest” would be eldali marir i tauressë. The ordinary plural isn’t proper here: eldar marir i tauressë would imply the entirety of Elvenkind lives in this one forest, which is unlikely. We will talk about this partitive plural in more detail in Chapter 6, Section §6.2.
4.2.1.5 How do we know this? Tolkien described the general process of Quenya plural formation in a number of places. The i-plurals were the more ancient plural form: “Plurality ... In nouns the most used element was [ī]. This was added to the stem direct, and since it preceded the addition of any other affixes, as those for ‘case’, it was probably the oldest element” (PE22/73). The use of r-plurals for vocalic nouns was a later development: “r (more rarely l) was originally, it seems, employed chiefly in verbs to mark the plural subject ... The invasion of noun-inflexion by r as a (nominative) plural sign is peculiar to Q. and an event that occurs within the period of the oldest records” (PE22/74).
The big exception to this “invasion” of r-plurals into vocalic nouns were those nouns ending in e: “The sg. quendë (not much used) was made in Quenya from Quendi, on the model of other nouns in -e, the majority of which formed their plurals in -i” (WJ/361). These e-nouns mostly retained their more ancient i-plurals. There are, however, various exceptions to this r- vs. i-plural split among nouns, as noted above. Tolkien didn’t describe every exception in detail, so many of them must be deduced from the example plurals he gave, as indicated by the page references in the descriptions of the alternate plural forms above.
4.2.2.1 Plurals and Adjective Classes: Quenya adjectives may also be singular or plural, and the plurality of the adjective matches the plurality of the noun it modifies. Adjective plurals are formed based on the adjective’s ending:
This last group of o-adjectives is very small and has no attested plurals, but probably follows a pattern similar to the wi-plurals for nouns from the previous section. They would change o to wi after g (-gwi) or c (-qui) and to ui otherwise. As a general rule, adjective plurals are similar to noun plurals, the main exception being a-adjectives and ëa-adjectives which form plurals with -ë and -ië.
4.2.2.2 Adjective-Noun Agreement: Plural adjectives are used when the noun they modify is plural. This can happen in several ways:
Note that there is no special “partitive plural” adjective form, so if the noun is in the partitive plural, the adjective still uses the regular plural: eldali norir lintë “some elves run swift[ly]”.
4.2.2.3 Unpluralizing Adjectives: Reading plural adjectives can be a bit tricky, since you have to mentally “unpluralize” them to determine their base form:
If the noun is plural (which is usually obvious), you need to “unpluralize” the adjective to determine its meaning.
4.2.2.4 How do we know this? The pluralization rules for adjectives are based on attested examples, as indicated by the page references above. Tolkien did briefly describe the origin of the e-plural for a-adjectives in a 1955 letter to David Masson: “final -ai in plural of adjectives > e, lintë, unótimë etc.” (PE17/76). Here Tolkien indicates the ancient plurals of adjectives used the same ancient i-plural as nouns: ✶lintai. But in the case of adjectives, this formation survived and went through historical sound changes to become the e-plural lintë, whereas with nouns the ancient plural was replaced by an r-plural: elda → pl. eldar.
As indicated above, we don’t have an example of a plural of adjectives like lungo (lungu-) “heavy”, but it is reasonable to assume they followed the same pattern as similar nouns, which likewise retained their ancient i-plurals.
Verbs form plurals by adding an -r: nori-r “run-(pl.)”, norne-r “ran-(pl.)”, nóra-r “(are running)-(pl.)”, nor-uva-r “run-(future)-(pl.)”. As usual with the aorist, the addition of a suffix changes the final e to i: norë “runs” vs. norir “run-(pl.)”.
Like adjectives, verbs are put into the plural when its subject noun is in the plural (including partitive plural): i eldar norir lintë “the elves run swift[ly]”, eldali norir lintë “some elves run swift[ly]”. This is also the case when the subject reference is indirect via a relative pronoun: tíran i eldar i nórar “I am watching the elves that are running”. This is not the case when the verb uses a plural pronominal suffix: norintë “they run”: there is no need to double-mark the verb as plural since the pronoun is suffixed directly to the verb. However, a verb plural is required if the subject pronoun is separated from the verb, such as with an indirect imperative ántë norir “let them run”, or an emphatic pronoun intë norir “even they run”; emphatic pronouns will be discussed in Chapter 7, Section §7.1.1.
Here are some vocabulary words having to do with the body:
Here are some vocabulary words having to do with the colours, using the British spelling of colour because that’s what Tolkien did:
Genitive Summary (Refresher): The genitive is a Quenya noun case roughly equivalent to English “of”.
Since the genitive is used to describe the relationships of body parts to people (i eldo telco = “the elf’s leg”; telco i eldo = “leg of the elf”), the genitive is useful for the following exercise.
Translate the following from Quenya to English. This should also give you some practice in “unpluralizing” adjectives.
Translate the following to Quenya:
Answers are in Answer Key 4.2 at the end of this chapter.
This section discusses the basic rules for Quenya word order, and how they apply to direct objects.
The basic word order of Quenya is similar to that of English. In most Quenya sentences, the order is subject-verb-object, as in atan matë massa “a man eats bread”. Likewise, Quenya adjectives generally precede the noun they modify: linta elda “a swift elf”, and prepositions precede the phrase they govern: nu i oron “under the mountain”, all like in English.
There are exceptions to these similarities, however. The biggest exception is with pronominal subject suffixes for verb, where the subject follows (and is suffixed to) the verb: mati-n massa “eat-I bread”. This is a remnant of a more ancient word order, because ancient Quenya sentences used the order verb-subject-object (PE22/93 note #4; PE22/132). Thus the ancient ✶mati ni massā became over time the modern inflectional Quenya matin massa, with the pronoun that followed the verb evolving into a suffix.
Another big exception is with Quenya noun cases. Because the noun case indicates the role of a word in a sentence, there is more freedom in where to place the declined noun in Quenya. In English, the sentence order typically dictates the function of each word: “The man gave the elf a pen” is subject (the man), verb (gave), indirect object (the elf), direct object (a pen). The equivalent Quenya phrase is i atan antanë i eldan tecil, where the indirect object elda-n is marked by the dative suffix -n. This marker allows the indirect object to be moved around in the phrase without introducing any ambiguity: i eldan i atan antanë tecil “to the elf the man gave a pen”, i atan antanë tecil i eldan “the man gave a pen to the elf”. English can accomplish something similar by marking the indirect object with the preposition “to”.
Quenya has some freedoms that English doesn’t. In English you can say “the man’s finger” but not “finger the man’s”. But the order of the equivalent Quenya phrase (using the genitive case) is flexible: both i atano leper and leper i atano are acceptable. In English the second example would need to be translated “finger of the man” to retain the same order. These noun cases will be discussed further in Chapter 5, §5.3 and Chapter 6, Section §6.3.
4.3.2.1 Direct Object Word Order: One area where Quenya word order is more restrictive is with direct objects. The direct object of a verb is the thing on which the action is performed. Modern Quenya does not have a case marker for the direct object (though classical Quenya did) and the direct object is generally marked by position alone, by following the verb: atan matë massa “a man (subject) eats (verb) bread (object).”
Even here there are exceptions, though. When the direct object is a pronoun, there is a tendency to displace the pronoun before the verb, so instead of melin tye “love-I you (object)”, it is also possible to say tye melin “you (object) love-I” (LR/61). In Chapter 3, Section §3.4.3 we had the example wish Nai Eru mantëa tye “may God bless you”, but what Tolkien actually wrote placed the pronoun before the verb: Nai Eru tye mantëa “may God you (object) bless” (PE17/75) [and the actual verb form was mánata, but we won’t discuss that alternative until Chapter 8, Section §8.1.4].
As another example, in a 1963 letter Tolkien wrote nai lye hiruva airëa Amanar, meaning “may holy Yule (Amanar) find thee”, reversing the order of subject and object. Here lye cannot be the subject because if it were, it would be suffixed to the verb: hiruvalyë. This reversal of subject and object can also be seen in the Namárië poem, where Tolkien wrote ilyë tiër undulávë lumbulë “all (ilya+ë) paths (tië-r) are drowned (undulávë) in shadow (lumbulë)”. Tolkien used the passive voice in the English translation to keep the word order similar, but the actual meaning is “shadow drowned all paths”. In notes on the poem from 1955 Tolkien wrote:
lumbulë is the subject, as is shown by singular verb (not unduláver), otherwise the poetic order metri gratia would be impossible. The normal [ancient] order in Quenya had been verb first, subject, direct object, indirect object. This survives in cases of “persons” inflexionally expressed [subject suffixes], but the classical and normal order was expressed subject, verb, object (PE17/72).
In the prose version of the poem Tolkien did indeed write the phrase in the more normal order: lumbulë undulávë ilyë tier, with the object last (RGEO/59). For beginners, I recommend sticking to the normal subject-verb-object order for Quenya, but you should be aware that other orders are possible, especially in poetry.
4.3.2.2 Object Suffixes: When the object of the verb is a pronoun, then normally Quenya independent pronouns are used: melin lye “I love you” (or lye melin). However, in the case of 3rd person singular (him/her/it) -s and plural (them) -t, special object suffixes can be appended directly to the verb. This can be done in two ways:
The 3rd person plural object suffix is used similarly: laituvalmet “we (exclusive) will praise them” (Let/448). Speculative: In the case where the object suffix is appended directly to an aorist verb without any subject suffix (hire-s “find it”), it seems the vowel remains e rather than becoming i as usual with inflected aorists (hiri-nyë “I find”). Likely this is because the object suffix was separated longer from the verb, and was only appended after the ancient final i had become e.
It has long been speculated that the other short subject suffixes (-n, -l, -t “me, you polite, you familiar”) can also be used as object suffixes, and there is some evidence of this in earlier writing: cestallen “you ask me” (PE22/138), using an earlier suffix -lle for “you”. But in a note from 1955 Tolkien said:
The inflexions are subjective but -s (singular), -t (plural, dual) may be added as objectives of 3rd person, utuvienye-s “I have found it” (PE17/110).
Tolkien’s omission of any object suffixes other than -s and -t implies that by 1955, these were the only two that remained valid. I recommend sticking to those two and avoiding any other object suffixes.
As for the notion that the aorist retains e when an object suffix is added directly to a verb, this is based on two examples: híres “find it” from 1969 (PE22/151) and apsenet “forgive them” from the 1950s (VT43/20). This makes it somewhat speculative. There is also a counterexample from the 1950s: caris “do it” (VT43/26), but that form was deleted.
Syntax is the linguistic term for the rules a language uses to assemble words into sentences. Unfortunately, Tolkien only rarely discussed Elvish syntax. He did describe the basic word order of Quenya sentences in a few places (PE17/72; PE22/93 note #4 and #5), but he changed in mind several times on how basic Quenya syntax worked. For example, he kept flipping between subject suffixes (which he used in The Lord of the Rings) and subject prefixes (as used in the 1948 document on the Quenya Verbal System: PE22/99-127).
It’s also clear from written examples that Quenya syntax is more complicate than just subject-verb-object. Since Tolkien didn’t directly explain the more complex cases, we are forced to deduce how things actually worked based on the examples he gave us. Since it isn’t always clear which examples are part of an internally consistent paradigm, multiple interpretations are possible.
Here are some vocabulary words having to do with food and drink:
Word Order Summary: Quenya word order is similar to English, but not identical:
Object Suffix Summary: There are two object suffixes which are added to verbs: -s “him/her” and -t “them”.
Instrumental Summary (Refresher): Here is a quick refresher on the instrumental for the next exercise:
Translate the following to English:
Translate the following to Quenya:
Answers are in Answer Key 4.3 at the end of this chapter.
This section discusses past tense formation, mostly for basic verbs.
This section is considerably more technical than the other information we have covered so far, and would better be classified as an “advanced topic” if the material wasn’t necessary for understanding the Quenya past tense. However, a beginning student doesn’t need a deep understanding of this material, so you can skim through it for now and just focus on memorizing the past forms that appear in this course. You can then reread this section more carefully when you find yourself in a position of needing to craft your own Quenya past tense forms.
Quenya past tenses are very irregular, and my general recommendation is that beginners simply memorize the past tense forms of verbs. The short-coming of this approach is that you will eventually encounter Quenya verbs whose past tenses are not known. To handle such verbs, you need a general idea of what the past tense of a verb might be, so that you can at least make an informed guess as to the verb’s past tense.
4.4.1.1 Ancient Past Formations: To understand the Quenya past tense, you need to understand its ancient origins. Quenya basic verbs originate from ancient biconsonantal verbal roots: modern tul- “to come” is derived from the ancient root √TUL. Ancient past forms were derived from the past suffix ✶nē, but this could be applied to ancient verbs in two ways: by nasal suffixion ✶tul-nē or nasal infixion ✶tu-n-l-ē. Different ancient past forms then underwent various phonological developments over Quenya’s history, producing differing modern forms: ✶tulnē > tuldë but ✶tunlē > tullë.
The result was often modern past forms that bear little resemblance to their aorist, present and future forms. The variable past tenses were sometimes reformed to align with other similar verbs, or were reformed to align with other tenses of the same verb. For example, it is fairly common for past tenses to be reformed to match the perfect tense, such as past túlë “came” based on its perfect utúlië “had come” (túlë is the actual past tense of tul-, as opposed to hypothetical and archaic tuldë/tullë). Different verbs underwent different reformation processes, and the net result is a modern past tense system full of irregularities.
That is not to say there is no regularity at all, however. Some patterns can be observed, and the past forms for some categories of basic verbs can be very predictable. The remainder of this section discusses the possible past forms of basic verbs, organized by the final consonants of their verb stem.
4.4.1.2 Pasts with p, t, c: Past tenses of verbs ending in voiceless stops p, t, c [k] are very predictable. The ancient nasal-infixed and nasal-suffixed forms produced the same results, because the combinations pn, tn, cn almost always invert to mp, nt, nc. Thus the past tenses of basic verbs ending in p, t, c almost always have mp, nt, nc in their past forms:
The first past tense tumpë is updated from the past tompë of an earlier version of this verb top- “to cover” from √TOP, as it appeared in The Etymologies of the 1930s (Ety/TOP). The last example is also based on the past tense of a different verb nac- “hew” from √NDAK (PE22/156), but that verb is a bit too obscure for this course and won’t appear, though there is nothing wrong with it.
4.4.1.3 Pasts with f, h: Past tenses of verbs ending in voiceless spirants f, h are also fairly predictable for reasons similar to that of voiceless stops: both nasal-infixion and nasal-suffixion produced the same result, a pair of voiceless stops. Thus verbs with f, h have pasts with pp, cc. These verbs are rare, because their ancient precursors ph, kh rarely appeared at the end of roots in ancient Elvish:
4.4.1.4 Pasts with Modern s (Classical þ): Ancient th also produces past forms similar to those of f, h, with pasts having tt. In this case, however, the resulting voiceless spirant þ further evolved into s in modern Quenya, which is quite far removed from its past form. The older tt-past forms tended to be modernized to nasal-infixed past forms with ns.
In one document Tolkien said that this ns further evolved into ss in late Tarquesta pronunciation (PE19/89), but there are no examples elsewhere of ss pasts from TH-roots.
4.4.1.5 Pasts with Ancient s (Modern r): Verbs with ancient s also had similar complications. In most verb tenses, the ancient s became r, but its historical past tense produced ss. This sometimes survived, but other times reformed to rn by analogy with verbs having both ancient and modern r (see next).
This last example is a strong past of a derived verb, something that will be explained later in this chapter.
4.4.1.6 Pasts with Ancient and Modern r: Verbs derived from roots with ancient r were also quite regular in their past tense. They consistently had rn in their past because this is another example where both nasal-infixion and suffixion produced the same result:
The main challenge with these verbs is that not every modern r arose from ancient r; some arose from ancient s (see previous) or ancient d (see next) instead.
4.4.1.7 Pasts with Ancient d (Modern r): Verbs with ancient d also had similar complications. The historical development for these verbs produced pasts with nd, but other verb tenses had r. The nd pasts sometimes survived, were sometimes reformed to rn (see previous) and sometimes reformed to a long vowel after the perfect tense (such as sére from esérië):
4.4.1.8 Pasts with v: Verbs whose modern forms have v are derived from roots with ancient b, which produced mb from infixion (common) and mn from suffixion (rarer). The resulting past forms sometimes survived, but these were more frequently reformed to long-vowel past tenses based on the perfect (lávë from alávië).
4.4.1.9 Pasts with l: Verbs with modern l were derived from ancient l, and as noted above could have past tenses with ll or ld depending on whether they resulted from ancient nasal-infixion or suffixion. Of these, ll-pasts were more common, perhaps because it was closer to l and the ld-pasts tended to reform in that direction. There are also examples of reformation after the perfect (túlë from utúlië):
Good examples of ld-pasts are hard to find, and the example aldë above is a (probably archaic) strong past of a derived verb ala-.
4.4.1.10 Pasts with m, n: Verbs with modern m, n were derived from ancient m, n, and almost always have nasal-suffixed pasts with mn, nn. Even within this stable group there are occasional irregularities, such as sámë “had”, past tense of sam- “to have” (PE17/173).
4.4.1.11 Pasts with y, w: Roots with Y and W only rarely form basic verbs, but there are some nasal-suffixed past tenses from Y- and W-roots appearing as the strong pasts of derived verbs (see later in this chapter for a discussion of strong pasts).
The previous discussion only addresses basic verbs: verb based directly on some ancient verbal root such as √TUL > tul- “to come”. Quenya also has derived verbs, which are the result of some suffixal addition to an ancient root, verbal or otherwise. For example, the verb tulta- “to summon” is a derived verb based on √TUL, developed from the ancient verb ✶tultā- with the added suffix ✶-tā and literally meaning “to make come”. All the a-stem and u-stem verbs (and other verb classes we haven’t discussed yet) are examples of derived verbs.
Individual derived verbs frequently have one of two different kinds of past tenses: a “weak past” using the suffix -ne (from ancient ✶nē) or a “strong past”. Weak pasts are the most common and easiest to understand: just add -ne to the verb stem: tultane “summoned”, the past tense of tulta-. The weak past suffix -ne is roughly equivalent to the English past tense suffix “-ed” seen in many (but not all) English past tenses. But some derived verbs have strong pasts instead. A strong past is one based on the verb’s root rather than the verb’s stem. A good example of this is the verb coita- “to live”, which has a strong past coinë “lived” based directly its root √KOY rather than a weak past **coitanë.
It is even possible for a derived verb to have two past tenses, one strong and one weak, used in different circumstances. An example of this is the verb ulya- “to pour”, which has a strong past ullë based on its root √UL which is used when the verb has no direct object: nén ullë i ailinessë “water poured into the lake”. It also has a weak past tense ulyanë used when the verb has a direct object: i elda ulyanë limpë “the elf poured wine”. Another example is caita-, which so far we have used only in the sense “to lie (down)”, but can also be used with a direct object to mean “to lay, place, put”. It likewise has both a strong past cainë (cainen i cöassë “I lay in the house”) and a weak past caitanë (caitanen i massa i cöassë “I laid/put the bread in the house”). For verbs with two pasts, this is the usual split: a weak past used with direct objects and a strong (or half-strong) past without a direct object.
Knowing when to use a weak or a strong past with derived verbs requires a deeper discussion of Quenya verb classes, a topic we will cover in Chapter 8, Section §8.1. For now, just know that if you see an a-stem or u-stem verb, you can guess that it has a weak past using the suffix -ne and you will be right a lot of time. You won’t be right 100% of the time, though, the same way English-speaking children sometime guess incorrectly at past tenses like “speaked” rather than the correct “spoke” (an example of an English weak vs. strong past).
Tolkien discussed the origin of the Quenya past tense on numerous occasions (PE18/95; PE22/131; VT49/30 and elsewhere). The most detailed breakdown of the Quenya past tense of basic verbs appears in the Quenya Verbal System written in 1948 (PE22/102-103). It is consistent with most attested examples of Quenya past tenses. The only major change was in the pasts of verbs with modern s but classical þ, where Tolkien decided in the 1950s that the normal past tenses had ns (PE19/89); this matches attested pasts from the 1950s and 60s.
The community of Neo-Quenya authors has been slow to accept the extreme irregularity of the Quenya past tense (myself included). Many older Quenya courses present a simplified version of the above as if there were only one choice for how to form the past tense of various groups of basic verbs. The unfortunate reality, though, is that the past form of a Quenya verb cannot be predicted from its stem alone, making memorization of past tenses necessary. Memorization is the approach recommended by this course, but even that approach isn’t perfect, since we don’t know the pasts of all Quenya verbs and sometimes guessing is required.
Hypothetical past tenses in this course are marked with a “*” or a “?” if the guess is particularly uncertain. This discussion of basic pasts should help experienced students make educated guesses of their own.
Here are the example verbs from the discussion above that have not been mentioned in previous vocabulary lists.
Here are a couple of additional vocabulary words for the story in the following exercise, some from earlier chapters.
This summary outlines common past forms of basic verbs, but not every verb follows these patterns.
Translate the following to English:
Translate the following to Quenya:
Answers are in Answer Key 4.4 at the end of this chapter.
This section discusses the types of name an Elf might have, as well as how those names are formed.
Elvish names are not the same as modern English names. As a rule, Elves do not use a family names like “Jane Smith”. Rather they generally receive two names at birth, one from their father and one from their mother. Their first name is given by their father, their ataressë:
Soon after birth the child was named. It was the right of the father to devise this first name, and he it was that announced it to the child’s kindred upon either side. It was called, therefore, the father-name, and it stood first, if other names were afterwards added. It remained unaltered, for it lay not in the choice of the child (MR/214).
The ataressë was as close to a family name as most Elves got, and in some families these name included a common element. For example, Fëanor included the name of his own father Finwë in the names he gave to his male children: Turcafinwë “Strong Finwë”, Morifinwë “Dark (Haired) Finwë”, Nityafinwë “Little Finwë” (PM/352-353). In other families, though, the father names could take any form. The father name was in some respects the “public name” of an Elf, used in formal contexts and given first when more than one name was listed. It is not necessarily the name used in ordinary life, however.
The second name most Elves received was their mother name, their amilessë:
The mother-name was given later, often some years later, by the mother; but sometimes it was given soon after birth. For the mothers of the Eldar were gifted with deep insight into their children's characters and abilities, and many had also the gift of prophetic foresight (PM/339).
The mother-name was often descriptive of the child’s personality, and in some cases prophetic of the child’s future fate. A mother name could displace the father name in normal use, though never displacing it in formal use:
The amilessi tercenyë or mother-names of insight, had a high position, and in general use sometimes replaced, both within the family and without, the father-name and chosen name, though the father-name (and the chosen [name] among those of the Eldar that had the custom of the essecilmë) remained ever the true or primary name, and a necessary part of any “full title” (MR/217).
A third name many Elves had was their self-name or chosen name cilmessë, as mentioned above. Among the Noldor the chosen name was selected by an Elf when they came of age, in a special ceremony called the name choosing or essecilmë.
But every child among the Noldor (in which point, maybe, they differed from the other Eldar) had also the right to name himself or herself ... [in a] ceremony called the Essecilmë or “Name-choosing”. This took place at no fixed date ... but could not take place before the child was deemed ready and capable of lámatyávë, as the Noldor called it: that is, of individual pleasure in the sounds and forms of words ... The Essecilmë, therefore, the object of which was the expression of this personal characteristic, usually took place at or about the end of the tenth year (MR/214-215).
This formal ritual of the Essecilmë seems to be a tradition limited to the Noldor. Within that tribe, the cilmessë was considered a “true name” like that of the father name and mother name, but less public than these other names:
The chosen names were regarded by the Noldor as part of their personal property, like (say) their rings, cups, or knives, or other possessions which they could lend, or share with kindred and friends, but which could not be taken without leave. The use of the chosen name, except by members of the same house (parents, sisters, and brothers), was a token of closest intimacy and love, when permitted. It was, therefore, presumptuous or insulting to use it without permission (MR/215).
Thus among the Noldor, father names or mother names were most often used in public contexts, while the chosen name was a private name used only among family and close friends or lovers. A fourth type of name sometimes appears, acquired later in life, called a “given name” anessë or an “after name” epessë:
There was another source of the variety of names borne by any one of the Eldar ... This was found in the Anessi: the given (or added) names ... [these] “given names” were not true names, and indeed might not be recognized by the person to whom they were applied, unless they were actually adopted or self-given. Names, or nicknames, of this kind might be given by anyone, not necessarily by members of the same house or kin, in memory of some deed, or event, or in token of some marked feature of body or mind. They were seldom included in the “full title”, but when they were, because of their wide use and fame, they were set at the end (MR/216).
On occasion, such a “given name” became the one by which an Elf best known. The Quenya name Altáriel of Galadriel was given to her by Celeborn [in at least one story: UT/298], and in turn became the basis for her Sindarin name Galadriel. Such names were also used among Elves who were not Noldor:
In addition any of the Eldar might acquire an epesse (“after-name”), not necessarily given by their own kin, a nickname — mostly given as a title of admiration or honour. Later some among the exiles gave themselves names, as disguises or in reference to their own deeds and personal history: such names were called kilmessi “self-names”, literally names of personal choice (PM/339).
A self-name taken outside the ritual of essecilmë does not have the same formality, because though it is called a cilmessë it is technically an anessë (self-given) and thus not a “true name”. Nevertheless, such a name could become the one by which an Elf was generally known.
The most ancient Elvish names, such as Ingwë, Finwë or Elwë, often do not have clear meanings. But in later days, most Elvish names are built from recognizable word elements. An Elvish name cannot be a simple word like “Rose”, however:
The mere names of things, such as “hill, river, tree”, and especially of unique things, as “sun”, were not used, at any rate without differentiation: ... for that would imply some kind of total equation or identity. An Elf (or Man) would not be called Anar “Sun” even to depict great glory or radiant vigour ... But an Elf or Man could be called Anárion, Anardil (PE21/86).
As such, Quenya names usually consist of two elements in a compound, generally being three or four (or rarely five) syllables in length. Quenya names generally follow several patterns:
Masculinized names tend to end in -o, and feminized names in -ë (PE18/99; PE21/81-82). The name Ancalimë is the word ancalima “brightest” feminized by changing the a to ë; Ancalimon is a similar masculine variant. It is not an absolute rule that male/female names end in -o/-ë, just a tendency. Other masculine suffixes include -on, -mo, -ndo and other feminine suffixes -el, -(i)en, -issë, -llë, -ldë.
There are also the masculine and feminine patronymics -ion/-iel “son of/daughter of. These were often applied to things other than the person’s actual parent: Anárion “Son of the Sun”, Undómiel “Daughter of Twilight”; this indicates affinity rather than actual parentage. Of course, application to an Elf’s actual parent is possible as well: Finwion “Son of Finwë”, the father-name of Fëanor, Elerondiel “Daughter of Elrond”, a name of Arwen.
Other popular name suffixes include -(n)dil “friend of” (Valandil “Vala-friend”), -(n)dur “servant of” (Isildur “Moon-servant”) and -tur “master of” (Ciryatur “Ship-master”). These are only attested in masculine names, and so it is likely there are feminine variants ending in ë, but only -(n)dilmë appears among published Quenya names: Vardilmë presumably meaning “Varda-friend (f.)” [this is a guess; other meanings are possible].
For compounds, if the first element ends in a vowel and the last begins with a consonant then the words can usually just be combined to create a name: Angamaitë “Iron-handed” = anga “iron” + maitë “handy; handed”. But one or both elements may be simplified in the compound to produce a more pleasing name, particularly if this reduces the syllable count: cala “light” + macil “sword” might become Calmacil. Some combinations of vowels or consonants may alter their sounds when they come into contact: elen “star” + macil “sword” becoming Elemmacil rather than **Elen-macil.
A full list of all possible combinations and how they might change is beyond the scope of an introductory course. Elvish name crafting is an art rather than a science, and surprisingly it is one of the most advanced tasks in using Elvish (Sindarin is even worse). Another common request is to translate realworld names into Elvish names, a process made more complex by the fact that real-world naming conventions often do not align with Elvish naming conventions. Fiona Jallings wrote an interesting series on the topic of “Your Name in Elvish”, which is a good exploration of the complications of Elvish name crafting: https://realelvish.net/tag/your-name-in-elvish/.
Here are some vocabulary words having to do with the elements, some of which we’ve seen before:
Tolkien actually gave the form of †hriz- for “to snow”, an archaic form from the root √SRIS (PE17/168). In the Noldorin dialect of Quenya (the basis of modern Quenya in Middle-earth), this z dissimilated away from the preceding r, becoming s (PE19/74). Thus likely its modern Quenya form is ^hris-.
Translate the following to English:
Translate the following to Quenya:
Answers are in Answer Key 4.5 at the end of this chapter.
Tolkien composed a version of the Markirya poem in the 1960s which was an updated version of the one he wrote around 1930. Both versions were published posthumously in Monsters and the Critics (MC) in 1983; it is the 1960s version of the poem we will discuss here (MC/221-222). First a bit of vocabulary, some new and some from previous chapters:
The words cirya, hresta, ve have appeared in previous chapters, but you may not remember them. The remaining vocabulary for the poem appears in this chapter.
The first five lines of the poem are:
Man cenuva fána cirya
métima hrestallo círa,
i fairi nécë
ringa súmaryassë
ve maiwi yaimië?
Broken down literally this is:
Man cen-uva fána cirya
Who see-will white ship
métima hresta-llo círa,
last shore-from (is cutting),
i fair-i néc-ë
The phantom-(pl.) pale-(pl.)
ringa súma-rya-ssë
cold bosom-her-in
ve maiw-i yaim-ië?
like gull-(pl.) wailing-(pl.)
Several of the above words are plural nouns or adjectives: fairë “phantom” → fairi “phantoms”, néca “pale” → pl. nécë, maiwë “gull” → maiwi “gulls”, yaimëa “wailing” → pl. yaimië. The verbs in the first two phrases are future cenuva “will see” (cen- “to see”) and present círa “is cutting” (cir- “to cut”). The second line is a continuation of the first and therefore its subject is the ship seen in the first line. My interpretation of the second line’s [cirya] círa métima hrestallo is that the ship is “cutting/cleaving from the last shore” with the metaphorical meaning that it is leaving, based on Tolkien’s English translation (see below).
The last three lines can be interpreted as a “to be” expression with omitted ná, meaning something like “the pale phantoms [are] in her cold bosom like gulls wailing”. From context, the “pale phantoms” are probably the ship’s sails, wailing in the wind. Earlier drafts of the poem explicitly referred to “sails” in the first few lines (PE16/74-75), making this interpretation of “phantoms” = “sails” more plausible. Likewise the “her” in the phrase probably refers to the sea, in whose “cold bosom” the ship now resides. Earlier drafts of the poem explicitly referred to the sea in the first stanza (MC/220; PE16/62).
Tolkien did not translate the 1960s version of the poem, but the translation of the 1930 version (MC/241-215) seems to still apply:
Who shall see a white ship
leave the last shore,
the pale phantoms
in her cold bosom
like gulls wailing?
This English translation aligns pretty closely to the literal interpretation above if you accept “is cutting/cleaving from” as a poetic expression for “is leaving”. Compare English “the last shore” to Quenya métima hresta, which lacks i. This seems to be another case where English uses the definite article but Quenya does not: when a noun has a determining adjective like “first, last”. English itself omits “the” after numbers: “one ship, two ships”, but it seems Quenya omits it with some other qualifiers as well.
Finally, the first line of the original poem actually has Men cenuva fána cirya with e instead of a in man “who”, but this is clearly some kind of error, since the word for “who” is man in all later stanzas of the poem.
Tengwar Names Part 1:
Body Parts:
Colours:
Food and Drink:
Example Verbs:
The Elements:
Miscellaneous:
The tengwar summary is deferred to Chapter 5, §5.1 until after we cover the other tengwar. Likewise, past tenses are not summarized here since the rules are simply too complex; for details see section 4.4 Basic Past Tenses.
4.7.2.1 Past Tenses of Basic Verbs Summary: This summary outlines common past forms of basic verbs, but not every verb follows these patterns.
4.7.2.3 Word Order Summary: Quenya word order is similar to English, but not identical:
4.7.2.4 Object Suffix Summary: There are two object suffixes which are added to verbs: -s “him/her” and -t “them”.
Tengwar answers:
Tengwar answers:
Tengwar answers:
Tengwar answers: