Note: This article is deprecated and will receive no further updates. Please head to EXA_PICO Wiki for the latest version of the lexicon and grammar guides.

Hymmnos is a language created by Akira Tsuchiya for Ar tonelico, a video game series (also known as Hymnos in the English localization of said games). According to the game world's history, it was formerly a common, spoken language, but has fallen out of use by the time of the main story's events (circa A.D. 3700). In modern times, Hymmnos is used only for songs that interface with ancient machinery. While it is a language unique to the world of Ar Ciel, part of its grammar rules and vocabulary seem to be influenced by English, Japanese, Sanskrit and German.

A defining feature of Hymmnos is that it puts a very strong emphasis on conveying the emotions of the speaker.

It should also be noted that none of the songs featured in the prequel series Surge Concerto uses any of the Hymmnos dialects, excepting the Ar Ciela language that served as a predecessor to them all, and instead they use two other conlangs that were developed for it: Emotional Song Pact and REON-4213.


Hymmnos is derived from Carmena Foreluna, the language used by the Moon Chanters, an ancient line of shamans who invoked magic through spell casting. The Moon Chanters gradually constructed the currently known vocabulary through countless trial and error approaches in an attempt to find the words that allowed the easiest tranmission of feelings and that would result in the strongest amount of power when chanted or sung. Of the many words they created, the oldest ones belong to the Kurt Ciel Note, which is considered an obsolete dialect of Hymmnos nowadays that almost no one speaks or uses.

Some time after the First Tower of Ar tonelico was built, the people in charge of that project took the Hymmnos language as it was made by the Moon Chanters of the time, and optimized it for use with controlling the Tower and its communications. More details about the history of individual dialects can be found in the dialects section of this article.


Currently there are two writing systems in Hymmnos:

The Hymmnos Script is comprised of glyphs with designs based partly on those used by the Sanskrit language. The appearances of its numeric glyphs are based on square wave patterns.

This particular script was standardized by the El Elemia Tower Administration Bureau in 3018 AD, all of which are representations of a shape called the Emission Concentric Circle. The Emission Concentric Circle is formed by a central point surrounded by several concentric circles, which is also crossed over by several radial lines that are both vertical and horizontal in directionality. This central point is called the Sound Source.


The Emission Concentric Circle. Taken from the Ar tonelico 1 Perfect Guide.

This makes the Hymmnos letters recordings of the directionality and spreading of the words (Sound Energy) emitted by the Sound Source. Therefore, most of the Hymmnos letters are composed of straight lines that go over the Sound Source, concentric circles that have the Sound Source as their center, and horizontal lines and vertical lines around the Sound Source.
The arching energy (shape) represents "peace (α-Waves)", the radiating energy represents "stimuli (γ-Waves)" and the vertical and parallel energy represents "tension (β-Waves)", all of which indicate the feelings contained in the letters themselves.

The entirety of the glyphs that make up the Hymmnos script can be seen below:

Hym font ver updata

Hymmnos glyphs. Updated to the 1.1 version (borrowed from

However, there is also a secondary script known as the Ar Ciela Script, which is used to write the Ar Ciela language, as well as for writing the Carmena Foreluna language. However, the latter can also be written in normal Hymmnos script.

Font ar ciela

Ancient Font. Borrowed from

Grammar (Standard Hymmnos)[]

Note: Standard Hymmnos is the conjunction of all the dialects made during the First Era: Standard Central Note, Kurt Ciel Note, Ancient Metafalss Note, Alpha Note and Cluster Note. This means that words from all these dialects can be used freely in Standard Hymmnos sentences.

Basic Structure[]

Hymmnos was designed with the purpose of stating the speaker's (or singer's) emotions, so most sentences begin with a set of three words called "emotion sounds".

Let's look at an example:

Was yea ra chs hymmnos mea
(I will be glad to turn myself into a song)

The first three words, "Was yea ra", are emotion sounds. They are are subdivided into three levels:

Was is Level 1, used for expressing the intensity of the emotion.

yea is Level 2, which names type of emotion being felt.

ra is Level 3, and expresses whether you want the emotion to last, or describes the context in which the emotion is felt.

All known emotion sounds are listed in the following table:

Intensity Type Desire to Last or Context
  • Rrha - Trance state
  • Was - Very much
  • Wee - Fairly
  • Fou - A little
  • Ma - One's presence of mind
  • Nn - Lethargy, unconscious state
  • i - Impatience
  • yea - Happiness
  • waa - Happiness
  • paks - Excitement, nervouseness
  • num - Nothing
  • ki - Concentration
  • wol - Fervor, rush
  • apea - Immersed in happiness
  • au - Sadness
  • granme - Wishing to protect someone, courage
  • touwaka - Wishful, hiding wishes
  • quel - Earnestness, despair
  • yant - Scared, terrified
  • lau - Fear, dread
  • guwo - Resentment, anger, hatred
  • zeez - Revenge
  • jyel - Loneliness
  • zweie - Sincerity, hiding decisions
  • ga - I want this to end soon
  • ra - I want this to last
  • erra - I want this to last eternally
  • wa - It doesn't matter, I will accept the current situation
  • gayaI never want to go back to how I was before
  • gagis - I'm all right like this

The rest of the basic grammar is very simple. It follows this structure:

(verb) (object) (compound or object)

Returning to our earlier example, chs (to turn into) is the verb, hymmnos (song) is the object and mea (my or myself) is the compound-or-object.

Was yea ra chs hymmnos mea
(Very) (happy) (I want this feeling to last) (turn into) (song) (me)
(I will be glad to turn myself into a song)

Each sentence expresses one emotion, so you can have only one set of emotion sounds per sentence.

There are no verb tenses in Hymmnos; whether a verb refers to a past, present, or future action is implied by the context of the sentence.

By this point, you might be wondering where the subject is. Well, since Hymmnos sentences are first person by default, the subject is usually omitted.

Note: Apparently, the Lv. 2 Emotion Sounds can be used as well as adjectives, nouns and verbs. On the other hand, the adjectives that describe emotions (such "cyue" or "nyasri") can be used as Lv. 2 Emotion Sounds, as well.

Structure with a Subject[]

If you wish to speak about what someone else is doing, then you have to use the following structure:

(verb) rre (subject) (verb) (object) (compound or object)

"rre" is a subject definer. The noun it precedes becomes the subject of the sentence.

The first verb in the sentence describes what you are doing, and the second verb describes what the subject is doing. The sentence is still told from a first person perspective; emotion sounds apply to the speaker even if the speaker is not the subject. No sentence can contain more than one subject, so the subject definer can only be used once per sentence.

For example:

Wee ki ra hyma rre aquagon pagle wart
(Fairly) (concentrated) (I want this feeling to last) (listen) (subject definer) (aquagon) (talk) (words)
(I will concentrate on listening to the words that aquagon says)

Both the subject definer and the first verb are omitted if the subject is either a pronoun or being used as an object.

Also, the standard Subject-Verb-Object (SVO) grammar structure can be used in sentences devoid of Emotion Sounds. In these cases, the characteristic grammar rule of "the default subject is always I" of Hymmnos is ignored. And there is another form in which the structures are Verb-Object-Subject (VOS, with the subject being generaly separated from the rest of the sentence by a comma) and Verb-Subject-Object (VSO), containing Emotion Sounds and used for telling stories, but aren't very commonly used.


(Standard SVO structure in a emotionless sentence. Taken from EXEC_HARMONIOUS/. and EXEC_HARMONIOUS_FUSION/.)

Faura yerwe murfan anw sol ciel
(Little bird) (chirp) (feelings) (to) (world)
(The little bird chirps her feelings to the world).

(VSO structure sentence. Taken from EXEC_HIBERNATION/.)

Was yea ra vit bautifal faura anw dornpica en 1 dyyal nuih bexm
(Very) (happy) (I want this feeling to last) (see) (beautiful) (little bird) (to) (seed) (and) (night of the first day) (come)
(The beautiful little bird saw the seeds, and the first night came).

(VOS structure sentence with Emotion Sounds. Taken from EXEC_HIBERNATION/.)

Rrha cyuie gaya na ieeya crushue anw dornpica, Rhaplanca

(Trance-like) (sadness) (not wanting to go back to my previous state) (doesn't wish to craft) (to) (seeds), (Rhaplanca)

(Rhaplanca doesn't wish to craft the seeds anymore)


Pronouns can be used in two ways: as objects and as subjects. They differ slightly depending on how they are used.

Pronoun Used as Object Used as Subject
  • you
  • you (plural)
  • he / him
  • they / them (masculine)
  • she / her
  • they / them (feminine)
  • us / we
  • yor
  • yora
  • hes
  • hers
  • has
  • hars
  • mean
  • yorr
  • yorra
  • herr
  • herra
  • harr
  • harra
  • merra

Use of Particles (Prepositions/Postpositions)[]

If you need to use particles (or as we also know them, prepositions), you can insert them between the verb and the object. It's only needed if the action of the verb affects something other than yourself, though.

For example, "tes", a particle meaning "to", can be used here:

Was yea erra melenas tes ar ciel
(I will be eternally glad to give my love to this world)

But not here:

Was yea ra sonwe hymmnos mea
(I will be glad to sing my song)


Other than typical adjective lexicons, Level 2 Emotion sounds can also be used as adjectives.

Adjectives always come before the noun they affect, as we can see from these examples:

ridalnae sol ciel
(irreplaceable world)
tyui frawr
(small flower)

Passive Voice[]

If you want to change a verb into a passive voice, insert a "re" before the verb that you want to modify.

For example, prepending a "re" to the verb of this sentence:

Was ki ra gyuss lir
(I'm embracing the light)

Changes it into a passive voice statement:

Was ki ra re gyuss lir
(I'm being embraced by the light)

Negative Form[]

If you wish to negate a verb, just add "na" before the verb you want to modify.

For example:

Was yea ra chs hymmnos yor
(I will be glad to turn you into a song)

Can be negated by prepending "na" to the verb:

Was yea ra na chs hymmnos yor
(I will be glad for not turning you into a song)

This mechanism can also affect nouns, making them the opposite of what they would normally be.

For example:


Can be changed like this:

na yehah
(unhappiness, sadness)

Noun Phrases Indicating Ownership[]

If two nouns are placed next to each other, then it means that one noun "owns" the other. There isn't an actual rule indicating which noun owns which, however, so the direction of ownership is inferred from context.

For example:

Aurica forlinden
(Aurica's village)
revatail ciel
(Reyvateil's world)
sarla mea
(my song)
hyzik yor
(your body)

Ownership can also be indicated using the "oz" particle. When it's used, the noun that comes before it is "owned" by the noun that comes after it.

For example:

hymmnos oz faura
(song of the little birds)
sarla oz soare
(song of prayers)

Position Marking[]

If you want to indicate that something is inside, over, or in front of something else, you can insert a position marking particle before the noun that indicates the position. Use "ween" for inside, "won" for over, and "folten" for before and in front of. (There isn't yet a known particle for "below".)

For example:

won dor
(over the earth)
ween kapa
(inside the water)
folten forlinden
(in front of the village)

Special Character Pronunciations[]

These characters are used in either Hymmnos Binary or in special grammar functions that will be elaborated upon in later sections.

Character Pronunciation
  • =>
  • <=
  • >>
  • x
  • 0
  • 1
  • Xc=
  • ->
  • <-x
  • Tab
  • Tab'ra
  • Tras
  • Gu
  • O
  • I
  • Xeku
  • Pass
  • (S)pag

Note: The 0 and the 1 digits used here aren't the same as the ones in the numbers. These ones are used in Hymmnos Binary, and their use is signalled by this mark: #x#>>#### (replace the #s with 0s and 1s).

Hymmnos Binary is used in the execution of the programs within the Tower and in some hymns, like in this example: 0x010001110.


  • Nel - 0
  • Nnoi - 1
  • Zi - 2
  • Dri - 3
  • Fef - 4
  • Vira - 5
  • Ixa - 6
  • Hept - 7
  • Oct - 8
  • Nei - 9
  • Dec/de - 10
  • Hec/he - 100
  • Kik/ki - 1000
  • Mik/mi - 10000

The second form of the decimals is used when combining, in a manner similar to Japanese. For instance, zide is 20 and kik octohe ixade octo is 1868.

Numeral Alteration[]

(Unconfirmed Information)

Apparently, the New Testament of Pastalie dialect has different ways for pronouncing and using numbers, according to KWhazit:

"It has an alternate system for pronouncing numbers than the one normally used. The suggested pronunciation (which sounds close enough to be possibly correct) is ノイ・フェフラ・ノイ (nnoi fefra nnoi). Not only does it sound like what is actually sung, but it makes some logical sense as a shorthand for numbers with many zeroes as well. ノイ (noi) is the usual reading for 1, and フェフ (fef) (without the ラ (ra)) is the usual reading for 4. Putting that together, you have 1, a variation on 4, and another 1. Note that the thousands place, where the first 1 occurs, is the fourth position from the right. Supporting this theory is another note indicating that 5100 (used as the ID# for the Metafalica Hymn Crytal) sounds like ヴィラ・フェフラ・ノイ・トレル (Vira Fefra Nnoi Torel). 5 is ヴィラ, there's that フェフラ (fefra) again, and 3 is ドリ (dri). If we assume that トリル (Torel) is a mishearing of ドリラ (Drira), then we have a clear pattern: Add ラ (ra) to a number n to mean the nth digit from the right. Assuming this holds up, 1,000,000 could be read ノイ・ヘプトラ (Nnoi Heptra, since ヘプト (Hept) = 7), and so on. It would be comparable to writing 14z1 for 1001, 17z for 1,000,000, and so on."

Advanced Grammar Rules[]

Emotion Sound Keeper Definer Syntax[]

This structure allows a single set of emotion sounds to be shared by multiple sentences. Even though it's considered an advanced grammar rule, it's actually very simple. It is used in the following way:

(emotion sounds) 0x vvi. (keeper definer start)
1x AAs ixi. (keeper definer end)

What this means is that everything between the two keys, "0x vvi." and "1x AAs ixi.", will inherit the specified set of emotion sounds. For reference, "0x vvi" is pronounced "ogu vivi" and "1x AAs ixi" is pronounced "igu aas ixi".

Here is an example:

Was yea erra 0x vvi. ("Was yea erra" will be applied from this point forward)
chs hymmnos mea
hymme rre walasye hyma mea
sonwe anw sol ciel
rre sol ciel hyma hynne mea
1x AAs ixi. ("Was yea erra" will no longer applied beyond this point)
Wee apea ra rre yor melenas

The meaning of that passage is as follows:

Was yea erra chs hymmnos mea
(I will be eternally happy to turn into a song)
Was yea erra hymme rre walasye hyma mea
(I will be eternally happy that the people hear me)
Was yea erra sonwe anw sol ciel
(I will be eternally happy to sing to the world)
Was yea erra rre sol ciel hyma hynne mea
(And I will be eternally happy that the world hears me)
Wee apea ra rre yor melenas
(I will be immersed in happiness to love you)

The important thing to note is that even though "Was yea erra" was only spoken once, it was implied four times.

Double Register of Emotion Sounds[]

Even though writing Emotion Sounds in a sentence when using the Emotion Sound Keeper Definer Syntax is redundant, it can be done. It's only necessary if you want to indicate a change in emotion over one sentence between two parts that share the inherited emotion. Song servers give priority to explicitly specified Emotion Sounds when processing songs that make use of the Keeper Definer Syntax.

For example:

Wee apea ra 0x vvi.
(I will be immersed in happiness)
her ar ciel irs
(I'm immersed in happiness because I exist in this world)
hyma rre faura sonwe
(I'm immersed in happiness for listening to the singing of the birds)
Was yea erra sonwe hymmnos mea
(I will be eternally happy to sing my song)
sonwe sos rre ar ciel
(I'm immersed in happiness singing for the sake of this world)
1x AAs ixi.

As written here, the emotion associated with the third sentence is "Was yea erra", while all other lines carry "Wee apea ra".

Binasphere Chorus[]

This is one of the most complicated grammar rules existent in Hymmnos, and it's used to sing two lines of lyrics as if they were one. (Binasphere means "two worlds" in Hymmnos, this meaning that within the Binasphere Chorus song, there are two worlds contained). Doing this gives more power to the song since a song using this technique can have two simultaneous effects.

Historically, this was developed by the Moon Chanters a little after Hymmnos was standardized halfway during the First Era, in search of methods to make the language even more powerful. Since humans are unable to sing more than one song at once, their Binasphere Chorus method required two simultaneous singers to be executed. Also, it had another fatal flaw: since both singers were humans, no matter how much they tried to sympathize with each other's emotions, they were never be able to reach the synchronization level needed to create a Song Magic. Theoretically, if they had succedeed at their attempts, the songs of the Moon Chanters would have gotten roughly five to ten times more powerful than what they normally were. However, they still would have never be able to compete with Reyvateils that were born afterwards, since their power was only a hundredth of the one the Reyvateils could command with their songs.

Now that we have reviewed its history, let's give a look at the process of turning the following two lines into a single Binasphere Chorus line:

Was yea ra chs hymmnos yor
(I will be glad to turn you into a song)
en chsee fwal fwal yor
(and then, I shall spread out your wings)

First, take those two lines and break them up however you wish. Individual words may be broken up into multiple pieces, too. Rewrite everything in uppercase characters and on the same line, then mark each word that was broken into more than one piece with an "x" at the end of each fragment, except for the last fragment of each word.

For this example, we've broken up and combined the lines as follows:


Notice how word fragments from both the first and second lines are mixed with each other? Assign each fragment a number: 0 for fragments that originally came from the first line, and 1 for fragments that originally came from the second line:

    0  1   0   1  0  1  0   0   1    0   1   0   1    0   1

Use those numbers to make a Binasphere Formula for this song. Binasphere Formulas look like this:

EXEC hymme 2X1/0>>formula

The word formula should replaced with the string of numbers that we got after assigning 0s and 1s to each fragment. Following our example above, the formula for this case would be:


And thus, the Binasphere Formula for this example is:

EXEC hymme 2X1/0>>010101001010101.

The formula is always placed at the end of the song, and a song can't have more than one Binasphere formula. If you want to use the Binasphere format again in the same song, then your lyrics will have to be adapted to work with the first formula. The => (pronounced "Tab") is used to mark the point at which the Binasphere lines begin:

Let's use the following set of lines from EXEC_NULLASCENSION/. in an example of how to decode a set of Binasphere lines:

EXEC hymme 2x1/0>>01101010

We have to use the formula to figure out which fragments belong to which lines. Since this Binasphere formula is shorter than the total number of fragments in lyrics, we have to keep looping the formula until we reach the end of the song. (The "/" isn't actually used, but we use it in this example to show where the formula is looping):

    0    1    1   0   1   0   1  0 / 0   1   1

 0  1   0    1   0 /  0    1    1    0   1    0

 1   0 / 0  1   1  0   1   0    1    0 / 0   1

1   0   1  0   1   0 / 0   1   1

 0  1   0    1    0 /  0   1   1   0    1    0  1 
 0 /  0    1  1   0   1     0   1   0

Note: The end of the final loop has to coincide with the last fragment of the song.

After that, we need to put the fragments back together on their respective lines, as indicated by the formula.

This is how the lines look once they've been pieced back together:

First line:

Rrha apea gagis gran paul nosaash yanje en ini ar ciel
(In this delightful trance, I feel that the way will be opened, for the goddess to purify this world forever)

Second line:

Rrha guwo gagis tie innna gatyuny ini ar ciel la zahha
(In this trance of hatred, I shall tie this curse to the inside of my mind, and then, I shall purify and make advance this world)

First line:

Rrha apea ra hartes yora chyet walasye forgandal wassa ciel
(In the midst of this trance of delight, I will love your special persons, because of the festival of the world)

Second line:

Rrha guwo ga gatyuny ar ciel en ini sor gatyunla art sa fayra
(In this transient anger, I shall purify this accursed world through the flames of the hell)

Note: In theory, it's possible to use the Binasphere Chorus function in Pastalie sentences too, but so far, no more details on this have been revealed.

Emotionless Sentences[]

Normally, if a sentence doesn't begin with emotion sounds, then it's implied that it has the same emotion sounds as the sentence that came before it. But if none of the sentences in a song have emotion sounds (and if the sentences aren't using the Emotion Sound Keeper Definer Syntax), then the Tower can't process it as a spell. It's accepted as a simple song instead, used only to convey feelings without a specific effect to execute. Even though a song without emotion sounds can still invoke some magical effect, the effect wouldn't be very strong.

Here is an example taken from the lyrics of EXEC_HARMONIOUS/.:

Faura yerwe murfan anw sol ciel
(The little birds chirp their feelings to the world)
Faura sonwe murfan anw sol ciel ee
(The little birds sing their feelings to the people)
Ridalnae sol ciel yanyaue manaf
(This irreplaceable world, and these precious lives)
Presia yasra lusye enclone anw omnis
(Wish that the light of love would flood from the sun and wrap everything)
Faura selena anw Metafalica
(The little birds play the song of hope)

This song was used to convey feelings to someone in the story of Ar tonelico, but not for executing functions through the Tower. It still contains Hymmnos lines expressing the thoughts of the singer, though. The end result of this song is the fusion of feelings, but since it depends mostly on the feelings of the singer, the Tower doesn't do much in response to it.

Grammar (New Testament of Pastalie)[]

(All information in this section was contributed by Lazy, from the forums of A Reyvateil's Melody.)

Word Usage[]

Words from Standard Hymmnos can be used in Pastalie sentences, thanks to the emulation capabilities provided by Infel Phira, but if a word with similar meaning exists in the Pastalie vocabulary, then it's more commonly used due to the fact that Infel Phira requires additional processing power and time to use words from other dialects, thus reducing the overall execution efficiency and power of the song.

For example: "rawah" (from Pastalie Hymmnos) and "frawr" (from Standard Hymmnos) both mean "flower", but "rawah" is used more often in Pastalie sentences.

On the other hand, Pastalie words cannot be used in Standard Hymmnos sentences, except in dual-server songs with the EXEC_over.METHOD or EXEC_with.METHOD prefixes. This is because neither Ar tonelico or Harvestasha have any of the Pastalie words registered in their respective Hymmnoservers or possess the algorithms required to process them and their unique grammar mechanisms.

Basic Structure[]

Pastalie is designed on the concept of expressing several meanings and emotions with as few words as possible. The simplest Pastalie sentences can be formed from just one word, which is an "emotion verb". This type of verb is the basis for the New Testament of Pastalie's design; no sentence can be complete without it. There will be a whole section dedicated to emotion verbs later on, as it's very complicated.

For now, let's look at the simplest example ever:

(I will gladly sing for your happiness)

As with standard Hymmnos, the default subject of a sentence is "I". All Pastalie sentences end with /. Although it's actually a syntactical element for putting the given sentence into operation in a similar fashion to hymn names like EXEC_XYZ/., let's say that it is the normal "dot" at the end of each sentence for simplicity's sake. If it's way too emotional, you can replace "/." with "!". If it's a question, use "?" in place of "/."

All Pastalie sentences must express the speaker's emotion towards the matter mentioned in the sentence. The exception to that rule is the Xc=ABC -> XYZ/. syntax, which will be explained in the advanced grammar section.

The main advantage that Pastalie Hymmnos has over the Standard dialect is that you can explain multiple levels of emotions at once: your own emotion, emotion towards "you" (the person being spoken to), and emotion towards the world, all with a single verb. On the other hand, it cannot express the degree of the emotion and whether you want the emotion to last or not, as Standard Hymmnos can with its first and third emotion sounds.


When you want to add an object, just add it after the verb.

For example:

hEmmErYE hymmnos/.
(I will gladly sing a song)

This is similar to English, and the reverse of Japanese.

Particles (Prepositions/Postpositions)[]

In cases where particles (better known as prepositions or postpositions in English) must be used, insert them after the verb and before the object. Particles are used when the result of the verb's action affects something other than yourself.

You should use a particle when saying, "give blessing to the earth", for example, but not when saying, "sing (a) song". It's a bit complex and different from both English and Japanese, but you can get used to it by seeing it a lot. This rule is common to both Hymmnos dialects.

Here is an example of using the "ut" particle, meaning "to" or "towards":

lElLYEn ut doodu/.
(I gladly give a blessing to the earth)

And an example of when a particle is omitted:

hEmEmLYEr sarla/.
(I gladly sing a song)

Emotion Verbs[]

At a first look, words like hYEmmEr or fOwOrYUn look like crazy strings of letters from someone typing with the Shift key held and released rapidly. In actuality, all emotion verbs are made from 2 components.

The first component is the "base verb", which look like these: h.m.m.r., f.w.r.n., a.u.k., c.z. It consists of two or more lower case letters with a dot "." called "bank period" in between each character. A base verb is a complete word and has meaning, but it does not contain any emotion (and Hymmnos is all about expressing emotion).

The second component is the "emotion vowel", which are those upper case letters you see in emotion verbs. Their function is to express the emotions associated with the action of the verb.

Whenever a base verb has at least one emotion vowel, it becomes an emotion verb. Just replace a bank period with an emotion vowel. All other unused bank periods in the base verb are removed.

From left to right, bank periods are called "bank 1", "bank 2"... The smaller the bank number, the higher degree of emotion. In other words, from left to right, the degree of emotion decreases.

For example:

h.m.m.r. (base verb: to sing) +
E (emotion vowel: happy) +
A (emotion vowel: strength/concentration) =
hEmmrA (I gladly sing with my strength)
(But my happy feeling is higher than my concentration feeling)

Now add /. to the end and you have a complete sentence, "hEmmrA/.", which means "I will gladly sing with my strength".

When an emotion vowel is repeated, that means that the feeling is multiplied (hEmEmrA means that I'm much happier than hmEmrA).

"x." is a special verb, known as the "subject definer". It's used when the subject of the sentence is not the speaker/singer, and will be explained in a later section.

Finally, there are cases when base base verbs are used in sentences, such as when verbs are used as part of the object or subject, which will be elaborated upon further below.

Emotion Vowels[]

Emotion vowels express the subject's emotional state when performing the action described by the base verb. (The subject may be either the speaker/singer, or another subject defined in the sentence.) They are divided into 3 groups called "Levels".

Level 1: Emotion vowels that express the subject's own feeling.

Level 2: Emotion vowels that express the speaker/singer's feeling towards the target being spoken to (usually "you, the audience listening to this story"). This "you" is from the speaker/singer's perspective, which may not be the same as the subject's perspective.

Level 3: Emotion vowels that express the speaker/singer's feeling towards the people of the world (from now on known as "this world").

Here is a list of all emotion vowels and their meanings:

Subject Audience World
  • A - Strength, concentration, doing one's best.
  • I - Pain, agony, fear, feeling of wanting to run away.
  • U - Sadness, worry, (sometimes means caring too).
  • E - Happiness, pleasure.
  • O - Anger, malice.
  • N - Absent-mindedness, relaxation, (sometimes means negative).
  • YA - Your feeling.
  • YI - Your suffering, your pain, your death.
  • YU - Your sadness, your anxiety.
  • YE - Your happiness.
  • YO - Your anger.
  • YN - Your calmness.
  • LYA - This world's feeling.
  • LYI - This world's pain, this world's destruction.
  • LYU - This world's sadness, this world's tension.
  • LYE - This world's happiness.
  • LYO - This world's anger, this world's madness, this world's fighting.
  • LYN - This world's calmness, this world's stillness.

As you can see, with the exception of the difference between between A and YA/LYA, they are mostly the same. Level 1 vowels have different meanings when used with "x." (the subject definer), however, and those will be explained in a later section.

In addition to expressing the emotions of the subject, emotion vowels also indicate the target that's receiving the result of the action, provided that the target is either "I", "you" or "this world". This is quite hard to explain, so let's look at a few examples.

Example 1:

Standard Hymmnos:
Was yea ra chs hymmnos mea.
New Testament of Pastalie:
cEzE hymmnos/.

They both mean, "I will happily turn (myself) into a song". The "mea" in the Standard Hymmnos sentence refers to means "me", the target of the verb "chs" (turn into). But we don't see that in the Pastalie version because the two "E" vowels in the verb "cEzE" already define the subject as the target of the action. (Remember, though, that the subject may not necessarily be the same as the speaker/singer.)

Example 2:

Standard Hymmnos:
Was yea ra chs hymmnos yor.
New Testament of Pastalie:
cEzYE hymmnos/.

Both sentences mean, "I will happily turn you into a song". The "YE" means the target of the action "turning into" is "you (the person listening to this sentence)". Furthermore, it means that the action "turn you into a song" is happening for the sake of "your" happiness. So, if we translate each sentence word for word, we get:

Was yea ra chs hymmnos yor.
(very) (happy) (I want this to last) (turn into) (song) (you)
(With great happiness and wanting this happiness to last, I turn you into a song.)
cEzYE hymmnos/.
(turn into + my happiness + your happiness) (song)
(With happiness and for the sake of your happiness, I turn you into a song.)

Each type of Hymmnos features its own kind of advantage in expression. The same case is applied to level 3 emotion vowels; they, too, can indicate that "everyone" or "the world" is the target of the base verb's action.

Subject Definer[]

Time for the "x." to come into play. Remember that I said all Pastalie (and all Hymmnos, regardless of which dialect) sentences have the speaker/singer as the default subject? Well, now any sentence where the speaker/singer is not a subject will begin with the special verb "x.", also known as subject definer. In fact, most Pastalie lines you see in hymns and songs are this type.

Even though it's special, "x." is still a verb with only one bank period and you can insert an emotion vowel there. What it does is to express the speaker's emotion towards the matter mentioned in the sentence.

Example: xE yorr cEzE hymmnos/. (You will gladly turn into a song, and I feel happy about that)Subject

Subjects in Pastalie are quite hard to understand, but once you get it you will be able to use it quite easily (unlike emotion verbs which require a lot of time to figure out).

To make things simple, I'll just list out EVERY possible situations of subject that could ever happen through examples. For a person, we'll use Lazy, and for a non-person thing we'll use "lyuma" (star). Finally, for subject definer we'll use "xA" (I feel neutral about...)

Situation 1: The speaker/singer ("I") as the subject [verb]/.

Very simple, no subject definer needed, since "I" is the default subject in all Hymmnos sentences already.

Situation 2: "You", "He", "She" as the subject xA _____ [verb]/.

Check up the pronouns table above and fill the blank with the correct pronoun for your subject.

Situation 3: A particular person or thing as the subject xA rre ______ [verb]/.

Just add the noun you need. In case of noun phrase that indicates ownership... Well I'll explain it later, in next part.


- xA rre lazy hYEmEmArA/. (Lazy sings happily)

- xA rre lyuma hYEmEmArA/. (The star sings happily)

Situation 4: "That" or "Those" as the subject

xA sorr [verb]/.

Basically the idea of the last sentence is used as the subject in this sentence... Like this (direct example from METHOD_REPLEKIA/. lyrics, but I removed the emotion parts to make it more simple)

xA harr hLYUmLYUmOrO eje/. (She sings (the song of) her heart)

xA sorr kLYUvLYUr du qejyu/. (Her singing covers all people)

There is a faster way to express this though, through an advanced grammar: verb used as subject (known as "clause" in English). I'll explain it later, as always.

Situation 5: "This" or "These" as the subject

[noun or sentence] <-x [verb]/.

This will be explained in the <-x advanced grammar part.

Noun Phrases That Indicate Ownership[]

Simply put, those are nouns indicating normal things with an owner, kinda like they belong to someone/something else. There are two types of noun phrases: Things belong to person/people (Lazy's star) and things belong to a non-person thing (sound of waterfall, for example).

Type 1: Thing belonging to person/people

Add an emotional vowel infront of it, making sure that its level corresponds to the grade of person/people you are trying to indicate. The type of emotion in the vowel is also the emotion the owner express when owning the thing. For definite ownership, add a level 1 vowel followed by a lower dash "_" and then the owner's name.


- Alyuma (My star, which I feel neutral about)

- YIlyuma (Your star, which you feel jealous about)

- LYElyuma (People's star, which they feel happy about)

- Ulyuma_lazy (Lazy's star, which he feel sad about)

This is the only different from how it's used in Standard Hymmnos.

Type 2: Thing belonging to a non-person thing

First noun is the "owner", second noun is the thing being owned. Example:

- zalez ale ("sound of waterfall" or "waterfall's sound")

- zodal sechel ("capital of death" or "death's capital")

- vonn papana ("rain of darkness" or "darkness' rain")

Just imagine that there is an invisible "'s" in between two consecutive nouns.

And finally, both types of "ownership" can be mixed together, but of course this can be a bit complex:

ayulsa Asiance_qejyu (People's ideal land of eternity)

Advanced Grammar Rules[]


XXX/. -> YYY/.

This allows you to store an entire phrase as a word. However, it only works for the song in which the function is used. It can make the sentences harder to understand, but also allows to make them shorter


- colga/. -> cEzE colga sos qOgYIs/. (Ice -> I shall be happy to turn into ice to put an end to you).

Wherever "colga/." might be used from now on, it will mean the same as "cEzE colga sos qOgYIs/.".


Putting "rre" infront of a subject that normally doesn't require it (you, he, she...) will "emphasize" the subject.


- xA rre herr hYEmmrE/. (He sings, emphasis on "he", to say that it's that person, that "him" and not anyone else)

- xI rre yorr cEzE hymmnos/. (You turn into a song, emphasis on "you", it's "you" and not anyone else who turn into a song)

You can also emphasize the subject "I" by putting subject definer (expressing what you think about your own situation), followed by "rre mea".


xU rre mea hAmmrA hymmnos/. (I sing the song with all my strength, emphasis on "I", and also express that I feel sad about my own singing)

Passive voice[]

Simply add "eh" after the emotion verb to change the sentence totally from active to passive voice.

Example: xU rre lazy fYUwrUn lyuma/. (Lazy embraces the star) ---> xU rre lazy fYUwrUneh lyuma/. (Lazy is being embraced by the star)

Unlike in English or Japanese, you don't need a particle to indicate the action's performer (no need to add "by" like English or change "wo" to "ni" like Japanese).

Desirative Form[]

When you want to say "want to do something" in Pastalie (usually more like "want to be able to do something"), add "aye" after the verb.


xA rre lazy hYEmmrEaye hymmnos/. (Lazy wants (to be able to) sing a song)

Negative Form[]

Adding "zz" infront of an emotion verb or a noun means the negative of it.


hYOmmrU/. (I sing) ---> zz hYOmmrU/. (I don't sing)

arhou (hope) ---> zz arhou (hopelessness, despair)


To "quote", put your sentence in :/ and /: (looks like some sort of un-smiley face). The /: replaces the /. at the end of sentence too.


:/sYAlE yor/: ("I believe in you")


The word "en" can be used the same way as our normal "and". Just use it when you think it fits, or use it in the same way as the comma ","

Bare Verb[]

As it was mentioned before, the main element of a Pastalie sentence is emotion verb, but sometimes when the action is being performed without emotion, like by a machine, or in an unconscious state, you can use a bare verb as well, though this is uncommon. Note that if "I" is not the subject, then you still need an emotion vowel for the subject definer.


xA rre lazy h.m.m.r./. (Lazy sings without any emotion)

Note: The next kind of advanced grammar; and this one are different.

Verb treated as subject[]

Also known as "clause" in English. Let's say, directly from the lyrics of METHOD_REPLEKIA/. we have these lines:

xA harr hLYUmLYUmOrO eje/. (She sings (the song of) her heart)

xA sorr kLYUvLYUr du qejyu/. (Her song covers the people)

By now you should be able to understand the above 2 sentences.

Now let's say we combine them into 1 sentence... There are 2 situations, "Her singing covers the people" and "Her singing of her heart covers the people".

First situation is very easy, the "clause" doesn't contain the object "eje", and you can combine them to this:

xA harr hLYUmLYUmOrO kLYUvLYUr du qejyu/.

Note that the "xA" expresses the emotion of the main sentence, so if the second original sentence was xU sorr kLYUvLYUr du qejyu/. then the result would be:

xU harr hLYUmLYUmOrO kLYUvLYUr du qejyu/.

Second situation is a pain... I'll just give you the example result first: xA harr h.m.m.r. eje kLYUvLYUr du qejyu/.

If you add in the object "eje", you have to remove the emotion from the "hLYUmLYUmOrO" verb, to prevent mistaking information between the 2 clauses. Why remove emotion vowels from that verb, and not "kLYUvLYUr" verb? Because the latter is the main verb of the sentence, which is more important, and must have emotion.

Another example:

xN rre hLYImLYUmOrO a.u.k. zess quesa/. (Her singing is like thunder)

"a.u.k." is the "to be" verb, and "zess" means "similar to". This is your standard type of comparison sentence, in which "a.u.k." usually has its emotion removed due to it being the "to be" that usually doesn't have emotion, unless it's the main verb in sentence (Example, aEuk lazy/. means "I am (happy to be) Lazy).

Verb Used as Object[]

What could be worse when you want to say "I embrace her singing of her heart"? Is it fEwrEn harr h.m.m.r. eje/.? No. It is:

fEwrEn x. harr h.m.m.r. eje/.

Yes, the subject definer is used in bare form here as well. This only happen when the "clause" is used as object, and only when the subject of the clause is not "I" so that subject definer has to be used in that "clause".

Another example:

xA rre lazy fEwrEn x. harr h.m.m.r. eje/. (Lazy embraces her singing of her heart)

Even when no object is needed for the clause, you STILL need to remove the emotion vowels for the subject definer AND verb. Example:

xA rre lazy fEwrEn x. harr h.m.m.r./. (Lazy embraces her singing) (This does not mean that Lazy is a "her", we're talking about a particular "her", so please don't be mistaken)

What if it's "I embrace my singing of my heart"?

fEwrEn h.m.m.r. eje/.

But if you want to emphasize the "I" in the object clause:

fEwrEn x. rre mea h.m.m.r. eje/.

-za Suffix[]

This suffix turns the verb that preceeds it into a noun, allowing it to be used as a subject or noun. However, unlike the methods discussed before, it also allows the use of Emotion Vowels in it, allowing also more variety in emotional expression. The suffix can be attached to the end of the verb, or separated by a space.


- xE harr hLYEmEmErza wEwEjLYEnEcE ut sphaela/. (I'm happy that her joyful singing resounds to the world).

And this other one shows the "-za" separated form the verb by a space (taken from METHOD_IMPLANTA/.)

- wYAfA za rYAfrm 1001 Atitia_qejyu en vYAsk yor/. (My wish is to show you the 1001 good hearts of the people, so you can experience them). (Note: This is general information took from the Settei Book, and there are some instances of the use of -za that haven't been deciphered yet).

One Subject With Multiple Verbs[]

I'll just give you the example, since this part it's insanely hard:

xA sorr mLYOrArA du sphaela/. (Her singing reflects this world)

xO rre mLYOtOyOyO giz wOsLYI du giz/. (Her singing creates terror, gives birth to terror)

You see, on the second sentence there are two verbs in one sentence, this is not "clause" but just a normal sentence with more than one verb, thus I add the comma "," in between the two verbs. You also notice (do you?) that the second sentence seems not to have a subject, after "rre" it jumps right to the first verb. Well, I can only say like this: In case of multiple verbs in a sentence, the last subject is supposed to be implied to this sentence, provided that you add subject definer and "rre" to it.

Xc=ABC -> XYZ/.[]

The -> is NOT the same as the one used in Binasphere Chorus. Your standard "If...then..." with a small notice: Both "if" and "then" clause must have the same subject, and that must be the subject from the last sentence.


aEuk lazy/. (I am Lazy)

Xc=hLYEmYEmArA -> cEzLYE hymmnos/. (If I sing then I will become a song)

When the subject is not "I" then there won't be any subject definer on the Xc= and -> phrases, since that's carried over from the above sentence too.


There are two uses of "<-x". For the first use, let's take a look at an example from the song MIO:

zz arhou, balduo, ujes, Oqejyu, xA rre <-x aYAuAkN kajya LYAglansee qejyu/.

(Despair, darkness, malicious minds, hated people, THESE are also necessary for our hearts)

Basically, "xA rre <-x" replaces one or a couple of things you just listed out earlier. "xA rre <-x" works like "this" or "these" in English.

However, "<-x" can also be used when you want to replace a phrase, not just nouns. When you replace a phrase, put the phrase in front, then add "<-x". This is in fact the third way to combine 2 sentences that share a phrase.

Maybe an example should be better. When you want to say "I feel sad about the fact that Lazy's happily singing (a song) becomes a star" you have three methods...

- First method: Using "sorr". This forces you to use 2 sentences but also allows you to express a different emotion for each sentence.

xE rre Lazy hYEmmEr hymmnos/.

xU sorr cEzE lyuma/.

- Second method: Treat the phrase "Lazy's singing a song" as a verb.

xU rre Lazy h.m.m.r. hymmnos cEzE lyuma/.

- Third method: Using "<-x".

xE rre Lazy hYEmmEr hymmnos <-x cEzE lyuma

I hope the example is easy enough to understand. Take note that the subject definer in the second method is "xU" while the third method is "xE". They only express either emotion from either original sentence.

The above example is actually complex since the subject of the first sentence is not "I". Another easier example: "My happily singing (a song) becomes a star".

hYEmmAr hymmnos <-x cEzE lyuma/.

Unfortunately, due to the limit of the "<-x" grammar, you cannot express your feeling about the fact in the sentence (so "I feel sad about my..." and "I feel happy about my..." are the same when you use this type of grammar).

Unknown Grammar Mechanisms and Corrupted Pastalie Hymmnos[]

These are grammar mechanisms that are still unknown, but still posted in here, so if anyone it's interested can help us reach a solution.ygulatiaznouongauvdejuyygulatiaznouongauvdejuy

ygulatiaz nouon gauv dejuy ygulatiaz nouon gauv dejuy

A line from ee wassa sos yehar.

Awiaa_vamuebyuiujesbOsYlmdnhyungxL YlvO!xL YlvO!!xL YlvOgzavamue!

Awiaa_vamue byui ujes bOsYlm dn hyung xLYIvO! xLYIvO!! xLYIvOg za vamue!

Tons of unknown words and verbs.


wasse zz urgn wasse kieghi wasse hiew wasse zodal sphaela!!

(Praise for no laments! Praise for kieghi! Praise for sadness! Praise for this dying world!!) ("kieghi" is still an unknown word)

That's how aquagon handled this sentence. Yes, I'm absolutely sure the translation is correct, but seeing that all other lines are Pastalie, this line must be Pastalie as well, not Standard Hymmnos, especially with the upper case letters which definitely indicates emotional vowels.

Grammar (Carmena Foreluna)[]

This language is a derivation of the Ar Ciela language used by the Wills of Planet Ar Ciel and is the direct ancestor of Hymmnos. Therefore, it served as the basis for the development of the Kurt Ciel Note. However, unlike any of the current Hymmnos dialects, it doesn't possess an actual structured grammar, and instead, it's largely dependant on the interpretations of its speakers.

Basically, each letter contains an emotion and a power, which comes in effect as soon as they are pronounced, and by chaining them, the words are formed. This means that this dialect doesn't have a set vocabulary either.

The following table lists the effects that each letter is known to have.

Letter Power/Meaning Letter Power/Meaning
A Power N Nothing, empty, emptiness
B World, society O Evil, wicked, dark
C Grow/Raise, change P Life & Death
D Darkness, demon Q Ignorance
E Love R Life
F Convey, Transmit, Permeate, Spread S Request, wish
G Destruction, curse, punishment T I, myself
H Fire, heat, passion, love U Hatred
I Holy V Joy
J Unknown/Yet to be known, underworld W Mind, soul, heart
K Creation & Destruction X Protection, protect
L You, Other Person Y Light
M Compassion Z God, gods

The words in Carmena Foreluna can range from something as simple as "Xa" (protective power) to something as complex as "zacta" (Oh, God, through your power, bestow growth upon me of power). Additionally, all the sentences in this dialect are polite requests due to the fact that they are prayers, and the more a letter is repeated in a word, the more it intensifies the power of said letter in the overall sentence.

For example:

while "a" means "power", "aaa" means "incredible power".

Also, Carmena Foreluna has a few rules about how the combined letters work, which will be detailed below:

  • The more the same letters are put, the more its effects are amplified (the stronger they have to be pronounced)
  • Whenever there are vowels between the runes, the power flows from left to right, in this way: Left>Vowel>Right.
  • When there aren't any vowels between consonants, it means in pronunciation that there will be a silent short stop, and that the phrase won't have any power. It's merely a modifier, a phrase for fleshing out feelings further. Such phrases can be written however one wishes.
  • Capital letters denote reverence. The higher the grade of reverence, the further the original meaning of the original letters can change.

As an example, we have here a sentence used as a prayer by the Moon Chanters since ancient times:

yeerh zacta tyy maarr itt sss
(Oh, God, please bestow power upon me!)

And here, how the words are deconstructed to reveal their complete meanings:

yeerh [action]: Oh Light + through your great love + and to my passionate life
                    y               e e                        r    h
zacta [action]: Oh God + through your power + bestow growth upon me + of power
                  z              a                    c          t       a
tyy [object]: Illuminate me with a shining light.
                         t         y        y
maarr [action]: Oh Compassion + through your great power + to my vivid life
                     m                   a        a             r    r
itt [action]: bestow holy power + to me!
                    i               t t
sss [object]: That is what I wish the most.
                       s        s      s

And by joining the meanings, while eliminating the redundant ones, we get this:

Oh light, through your great love,
Oh God, through your great power,
illuminate me with a shining light,
and bestow power to my passionate life.
Oh, compassion, through your great love,
bestow a holy power upon my vivid life!
These are my greatest wishes.

Which we can simply condense as "Oh God, please bestow power upon me!"

Carmena Foreluna can also be used to deconstruct current Hymmnos words, to learn what their origins were.

For example:

hymme [action]: Through the passionate light + please bestow salvation upon me > with your love.
                            h           y                m        m                     e

Which can also be condensed as "Please, save me through the power of your love", which was the original intent of "singing" for the Moon Chanters.

Ar Ciela[]

Note: Due to this language still being in development, it will be left incomplete until further information is received from the official Hymmno Server site

Due to its special traits (detailed below, in the Dialects section), Ar Ciela has two methods in which it can be written, both of which use the Ar Ciela writing system:

Writing Methods[]

Ar Ciela Public[]

"Public" here represents the fact that texts written in this method represents only the sounds within human's audible spectrum, and thus only uses 26 letters. To normal people, there is no way to understand those words as there is no additional information needed for deciphering - this method of writing only writes down what human can hear, not the true meanings of the language. This was the method used for writing the lyrics in the Hymmnos Concert CDs' booklets.

Ar Ciela Compartment[]

Basically, Ar Ciela Compartment is fairly similar to Ar Ciela Public, but it features additional diacritics, named Compartment Codes. Those Compartment Codes explicitly represent the conditions of the sounds which belong to the frequency range inaudible to humans.

Humans cannot write down the Compartment Codes as they are uncapable of hearing them. Furthermore, normal human can also hardly understand the meanings behind those Compartment Codes and cannot make one-to-one correspondences between the written and spoken language. However, for the purpose of scientific analysis, Ar Ciela Compartment is the standard and necessary method of writing the language.

As explained above, Compartment Codes are diacritics that are further added to normal Ar Ciela texts. They are used to represent Ar Ciela's 5 frequency tendencies and 4 amplitude tendencies. It's through those Compartment Codes that the basics of Ar Ciela words are revealed - think of them as the radicals in kanji characters.

Sound frequency from 20000Hz to 600000Hz is broken into five parts, called sessions. For example, if you hear a 100kHz sound that sounds like "a," it could actually be "ka" or "sa" or something else, but your ears cannot distinguish those high frequency sounds.


Compartment Codes that represent frequency sessions are called FMCLs.

The five FMCLs are:

Diacritic FMCL not written down Fm1cu Fm2z Fm3 Fm4n
Compartment Name session-0 session-1 session-2 session-3 session-4
Corresponding Wave Band Frequency 20kHz~50kHz 50kHz~100kHz 100kHz~150kHz 150kHz~300kHz 300kHz~600kHz

The other type of Compartment Codes, AMCL, represents the amplitude tendencies - the shape of the sound wave above 20000Hz, per se. Those amplitudes are vital to recognizing sound pressure made by pronunciation of words, which is necessary for realizing those words. For example, between "ka" and "sa," the former has a shorter and more focused sound, leading to a sharper amplitude than the latter, which is softer and has a more flat sound pressure.

The four AMCL are:

Diacritic Am1y Am2u AMCL not written down Am3y
Compartment Name quad dual single half
Corresponding Amplitude Trend \___ / ̄\_ ____ /\

Normally, FMCL are written above the letter, while AMCL are written under them. However, those Compartment Codes are never used with vowels. That is because it is still possible to distinguish vowels even at high frequencies, and the confusion is only caused by the large variety of consonants that become unrecognizable.

Differences Between both Writing Systems[]

Let's show the differences between both Ar Ciela Public and Ar Ciela Compartment by showing how the words Ec Tisia are written in both ways:


In here, you can see how it looks written in Ar Ciela Public. As you can see, it is written normally like many other languages that use alphabets - including Hymmnos. However, it is still yet impossible to recognize the meaning of this word.


While in the other hand, here is the same phrase written in Ar Ciela Compartment. It becomes Ec[s-4/half] T[s-3/quad]is[s-4/dual]ia (you can get this by typing E%)c $&Ti%(sia on your keyboard, using the official Ar Ciela Compartment font).

Now, using the corresponding table below, you can finally decipher the meaning of the word by breaking it down into letters: "Sympathy + Growth" and "People + Sincerity + Forgiveness + Hope + Power of Emotion" respectively.

Meanings of the Sounds[]

Ar Ciela is the origin of Carmena Foreluna, thus, the pronunciation of each letter represents a different emotional state (which is converted into H-waves). While sharing similar "randomness" feature with Carmena Foreluna (which means you can freely recognize and translate the letters however you want), once the letters are fused into words, they form a solid meaning. Take Ec Tisia for example. As you can see above, its meaning is really random: "Sympathy + Growth" and "People + Sincerity + Forgiveness + Hope + Power of Emotion." However, once converted into H-wave, the feelings merge together and, as a result, Ec Tisia as a whole means "Forgive everything." That said, the meanings of words are not limited to just a mix of each separate letter's meaning. It was through a long process of mixing, adding and alternating countless emotions that Ar Ciela came to the language of today which we call Hymmnos.

Below is a chart that summarizes all known emotion in each separate letter known up till now. Vowels are not divided according to their frequency, because they require the addition of consonants in order to acquire meaning. Speaking in Wave Science's terms, vowels are not classified - they remain the same regardless of Compartment. To human, the meaning of each vowel might sound varied and only loosely connected, but as for the Planet, she feels that those are one and only unified emotion. Those unified emotions are surrounded by [brackets] in order to distinguish them from humans' derived meanings. That said, it is not possible to for human to alter the words as they please. Words in Ar Ciela are crafted by the Planet herself, and is a synthesized form of emotion, so if you for example replace "A" with "U" or "N" then the meaning of the word will be altered, and it will no longer be Ar Ciela anymore. It is assured that the emotions of vowels and consonants cannot be expressed by words.

Wave Frequency Band Levels (Each Level Has Emotions Corresponding to their Representative Pronunciation)
A [Scalar] Neutrality, Equality, Fairness, Energy, Power of Emotion, Balance, Love
I [Prayer] Fear, Anxiety, Respect, Hope, Spirit, Kindness
U [Experience] Aspiration, Envy, Friendship, Cooperation, Friendly Love
E [Neighbor] Compassion, Restraint, Understanding, Sympathy, Sexual Love
O [Karma] Hatred, Anger, Sadness, Admonition, Narcissism
N [Mandala] Emptiness, Comfort, Enlightment, Benevolence
B Oneself Neighborhood Region World Love
C Transformation Revolution Growth Love
D Evil Corruption Darkness Light Love
F Conduction Influence Penetration Love
G Hell Destruction Change Reincarnation Cultivation Love
H Heat Flame Burning Passion Love
J Ignorance Unknown Fantasy Singularity Love
K Creation and Destruction (exact frequency bands unknown) Love
L Stranger Neighborhood Friend Companion Trustworthy Love
M Serenity Pity Salvation Love
P Life and Death (exact frequency bands unknown) Love
Q Foolishness Ignorance Innocence Love
R Fomenting Throbbing Life Love
S Lust Dependence Wish Prayer Gratitude Forgiveness Love
T Ego We People World Love
V Joy Praise Love
W Interior Spirit Love
X Aggression Defense Protection Love
Y Darkness Sacred Love
Z God Master Interior Ego Love

Only the known emotions are written in the above table, according to research materials from the Archia Think Tank. Furthermore, their divisions into I to IX are purely for the purpose of researching, they have absolutely no meaning to the actual Ar Ciela language. So, say, if you divide a section - III for example - into 10 smaller subsections, the "feeling" will be divided even further and become much more detailed, bringing forth many more new emotions.


The origins of Hymmnos and the Hymmnos script are ancient, and existed before the father of Sound Science discovered the [Sound Magic] in the First Era. It was not yet called [Hymmnos] at that time, but the incantations that all sorts of shamans used, called [Moon Chanters] in those days, became the origins of Hymmnos. Hymmnos was later formalized as a language optimized for control of and communications with the Towers, but its grammar and vocabulary sounds are based in the incantations of these shamans. The generic Hymmnos that spread throughout the world from there subtly changed format depending on the region and sect, and the vocabulary also steadily altered as history progressed. Therefore, archaisms, dialects, and so forth also exist in Hymmnos, just as they do in Japanese and other languages.

The dialects are called "Notes", which comes from the musical terminology for easier understanding and recognition.

These are the Hymmnos dialects that are still in use during the present, the Third Era:

  • Central Standard Note (Standard Dialect)
  • Kurt Ciel Note (Ancient Language From Before the First Era)
  • Cluster Note (Cluster Area Language)
  • Alpha Note (Origin Spell)
  • Ancient Metafalss Note (Sacred Language from the First Era)
  • New Testament of Pastalie (Pastalia Dialect)
  • Carmena Foreluna (Ancient Language from Before the First Era))
  • Corrupted Pastalie Hymmnos
  • Ar Ciela (Planet Language)

Central Standard Note (Standard Dialect)[]

[Central] refers to the First Tower [Ar Tonelico], called the core of Song Magic in the First Era. In other words, the incantations optimized to control the Tower are called Central Standard Note. However, this is not to say that only this Central Pure Note can control it, and in practice it is possible to control it using any Note. That is to say, Central Standard Note is nothing more than the format that the tower administration in Sol Ciel's capital city [El Elemia] decided upon as a global standard (like the ISO standard on Earth).

As a result, everything standardized from each [Note] becomes [Central Standard Note], so everything acknowledged in lookups and so forth is all written as Central Standard Note. Central Standard Note and other forms thus often exist mingled together.

Kurt Ciel Note (Ancient Language From Before the First Era)[]

This Kurt Ciel Note is itself a refined form of the spells of the shamans, the [Moon Chanters/Tsukikanade], and is the origin of Hymmnos. It has largely been forgotten in modern times, and it can be said that this line of vocabulary scarcely remains.

Cluster Note (Cluster Area Language)[]

Passed down and established during the First Era in the Sol Cluster region, a region far away from the Sol Ciel region (on the opposite side of the planet), is the [Cluster Note]. From the standpoint of Central Standard Note, it is a [dialect] of sorts, but the practical effect is scarcely different. At present, there are said to be none in the vicinity of Ar tonelico who can use this [Cluster Note], but somewhere in the world, a colony centered on Cluster Note might exist even in the present day.

However, the Sol Cluster people despised the people from Sol Ciel, and apparently, they created this dialect as a response to the creation of the powerful Central Standard Note. Probably it was created with the purpose of surpassing the faculties of the Central Standard Note, but since there aren't any speakers and it isn't used in Sol Ciel, or in Metafalss, the exact details are unknown.

Alpha Note (Origin Spell)[]

The [Reyvateil Origins], of which there are said to be only three in the world, can put together optimized Hymmnos terms in their own minds. Terms formed like this are called [Alpha Note]. This Alpha Note is somewhat troublesome in that it is unable to produce maximal power if not used with the tower that Origin manages. That is because, in contrast to the way the other [Notes] put together terms that are within the current specifications of the tower (strictly speaking, Reyvateils just pick out terms that are already present), this Alpha Note alone expands the receiving tower's functionality and creates a Hymmnos term to deal with it. In other words, in her own tower exists the functionality to handle that term, but it is an invalid term on other towers lacking this, and is thus naturally not general-purpose.

(For example: the Chronicle Key Hymn is an Alpha Note, Eolia Genus word, meaning that it was crafted in the first Ar tonelico Tower, managed by the Origin Eolia, and is unusable in the other Towers because these Towers don’t have any of the sealing mechanisms or programs that the hymn needs to be executed.)

Ancient Metafalss Note (Sacred Language from the First Era)[]

A unique Note invented in the early First Era by the holy land of [Metafalss] that dominated the world back in the time of the 2000s. The power that one word holds far exceeds the average value for Central Standard Note words. The reason they produced such high-powered Hymmnos terms is said to be in their land. In this land, spiritual ideology thrived more than in other regions, and the study and development of songs through thanks to the Moon Chanters was extensive. For that reason, even as other regions were reaching the limits of song power, they overcame these little by little, training themselves to craft yet stronger songs, then refining and improving them.

For better or worse, the influence of their spiritual ideology meant that even as other regions began studying the songs scientifically, and discovered the importance of the entonation and emotions in the songs, the spiritual thinking of the Moon Chanters lived on even then, and the civilization rapidly started rolling back after entering the 3000s because of that. Though thanks to the diplomatic relations with Sol Ciel, the Metafalss Note came to be registered on Song Magic servers.

New Testament of Pastalie (Pastalia Dialect)[]

A new dialect created in the land of Metafalss around the year 3400. This was created for [Infel Phira], a different Song Magic server from [Ar tonelico], and only the I.P.D.s(Infel-Phira Dependant), the Reyvateils linked to that server, can use this dialect. Thanks to its special characteristics, its compatibility with the other dialects is almost zero, from grammar to vocabulary. The execution commands are thus also changed into the unique declaration [METHOD_XXXX/.], instead of the [EXEC_XXXX/.] one. However, in practice, other forms are also generally available through emulation. They are translated into New Testament of Pastalie within Infel Phira, but on the user level, it is possible to use [Central Standard Note], for example, without modification.

The grammar differs entirely from the other Hymmnos formed in the First Era, so a separate grammar guide has been prepared. It should be consulted as well.

Carmena Foreluna (Ancient Language From Before the First Era)[]

It’s an ancient dialect, and it’s the Moon Chanter language itself. For those reasons, it isn't considered as a Hymmnos dialect by some. Its grammar was based on the unification of [Pronunciation (letters)] and meanings, where each letter had a meaning and effect. Since the extent of its effects are very small and required entire choruses of people singing in harmony to actually bring about any effects, the majority of the Reyvateils don't support (understand or speak) this dialect.

Corrupted Pastalie Hymmnos[]

There isn't any information available yet. Unofficial name.

Ar Ciela (Planet Language)[]

Ar Ciela, commonly known as "Language of the Planet," is not a language used by the humans. This language is used when the Planet wants to express her will and intention, and also when the Planet wants to craft something (comparable to Earth's activities ranging from weather to volcanic eruption to the cycle of life). In such situations, Ar Ciela is used to materialize those wills of the Planet. You can also say that Ar Ciela is the origin of everything that belongs to the Planet Ar Ciel, and was also the initial point for the development of the Carmena Foreluna language, which in turn also served as a predecessor for Hymmnos.

Ar Ciela has a really special trait, that is, you cannot completely comprehend its meaning within just the audible spectrum. That is because, the pronunciation of the language itself becomes H-wave and takes effect in that very form. Human's normal audible range is within 20~20000Hz, but Ar Ciela's pronunciation can easily reach up to 20~600000Hz (30 times higher than the audible spectrum) of H-wave. With such an extensive range like this, necessary sampling rate for recording Ar Ciela would be 1200kHz (in comparison, CD uses 44.1kHz sampling rate).

So what will happen when normal humans listen to Ar Ciela? There is one method through which you can experience this yourself. First, record your speaking or your singing into your computer, then use an audio editor program to cut off all the sound waves that are over 100Hz in frequency. The result will be a big mess of muddy and stuffy voice which is difficult to understand. That is the same effect as when a normal person listens to Ar Ciela.

For example, record your saying the word okaasan ("mother") into your computer. Then, if you cut off the sound waves that are over 100Hz, the word will sound like "ohoohon." Now, you can still tell a few things from this albeit highly distorted word: it has five morae, accent falls onto the 2nd and 3rd morae, and you can also feel where the vowels are. As such, even though in reality that word might even sound similar to something random like "notaahau" (which has no meaning at all), your brain will automatically come to realize that this word is okaasan. However, with Ar Ciela - a completely strange language, you cannot "guess" like in Japanese. For example, take the word Tisia. It's the result of the "cutoff" process mentioned above, and as you can see, there are countless ways to infer the original pronunciation of that word. But certainly, if you could somehow magically become capable of listening to sounds all the way up to 600kHz, you would be able to realize the original and precise pronunciation of that word, and would understand its meaning.

Another example. The Japanese are often known for not being able to tell the difference between /r/ and /l/, so without proper training, Japanese people will have difficulty distinguishing the two English words grass and glass - they both sound "garasu" to them.

As such, when a normal person listens to Ar Ciela, in order for him to be able to recognize the words' true meaning, Ar Ciela has been divided into categories. Furthermore, those categories are also split into dimensional visual subdivisions - that is, subcategories that attempt to visualize Ar Ciela pronunciation in dimensional forms.


The currently known vocabulary for all dialects of the language can be found over at Hymmnos:Lexicon.

Bonus: Hymmnos Spells[]

And as a small extra, here are some of the Hymmnos Spells that the humans can use to interface with the Tower:

Ma num ra exec guol CA07795 sos yor.

"I'll close (the door) CA07795 for you"

This Hymmnos Spell is used to activate an elevator in the Conductor Activator.

Ma num ra tek tes weaken.

"I'll move to the right"

This Hymmnos Spell is used to move the bridge in Sol Marta to the right.

Ar tonelico Download Spell[]

This is the spell used for downloading Hymn Crystals in the Reyvateils that are connected to either Ar tonelico, Sol Marta or Harvestasha. It comes in two versions:

Version 1 (used in Ar tonelico 1)[]

Fou paks ra exec hymmnos (hymn name)

Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec drone hymmnos (hymn name)

enter (Reyvateil's hymn code)


Fairly nervous, I shall execute the song (hymn name)

I will be happy to turn you into a song,

and into the wings you carry.

Execute the download of the song (hymn name)

into (Reyvateil)


Fou paks ra exec hymmnos PAJA

Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec drone hymmnos PAJA



Fairly nervous, I shall execute the song Purger

I will be happy to turn you into a song,

and into the wings you carry.

Execute the download of the song Purger

into Misha Artsellec Lune.

Version 2 (used in Ar tonelico 2 and 3)[]

Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec drone hymmnos (hymn name)

enter (Reyvateil's hymn code)


I will be happy to turn you into a song,

and into the wings you carry.

Execute the download of the song (hymn name)

into (Reyvateil)


Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec drone hymmnos METAFALICA



I will be happy to turn you into a song,

and into the wings you carry.

Execute the download of the song Metafalica

into Frelia.

Ar tonelico Uninstall Spell[]

This is the spell used for extracting some items that are installed in a Reyvateil that can't be taken out by other means (such as the D-Cellophane):

Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec na drone (binaries)

oter (Reyvateil's hymn code)


I will be happy to turn you into a song,

and into the wings you carry.

Execute the extraction of this target

out from (Reyvateil)


Was yea ra chs hymmnos yor,

en chsee fwal fwal yor.

Exec na drone 0x11000101101



I will be happy to turn you into a song,

and into the wings you carry.

Execute the extraction of this target

out from Frelia.

Infel Phira Download Spell[]

This is the spell used to download Hymn Crystals that are used to control Infel Phira into the IPD Reyvateils. However, this will be explained more in depth, given that this spell is more complex:

cEzYA hymmnos

(Hymn Crystal ID#) --x tArm azit tn='(Hymn Name)'

es tn=>sol.infel-phira.mea


I will be happy to turn you into a song.

Extract and expand the song (hymn name) out from the Hymn Crystal (ID#)

and upload it into me, the one that resides inside of Infel Phira.


cEzYA hymmnos

#5100 --x tArm azit tn='METAFALICA'

es tn=>sol.infel-phira.mea


I will be happy to turn you into a song.

Extract and expand the song Metafalica out from the Hymn Crystal #5100

and upload it into me, the one that resides inside of Infel Phira.


cEzYA hymmnos = New Testament of Pastalie direct equivalent to the "Was yea ra chs hymmnos yor" line from the Ar tonelico download spell.

#5100 = This is the ID tag under which the Hymn Crystal is recorded in the Infel Phira database. It's required for security purposes.

--x = This indicates the way of transference, indicating that the transference of feelings is from a Hymn Crystal to a Reyvateil. There are two other indicators: c (--c), which indicates that the transference is from an Hymn Crystal to other Hymn Crystal, and r (--r), which indicates that the transference is from a Reyvateil to a Hymn Crystal.

tArm = Emotion Verb for "realize" and "expand". Bare form: "t.r.m.".

azit = The format of the emotions that are going to be recorded in the Hymn Crystal or downloaded in the Reyvateil. No further details are known about this.

tn='METAFALICA' = Declares the feelings of the song being extracted from the crystal or Reyvateil as the definition of the "tn" variable.

es tn=>sol.infel-phira.mea = Orders the transference of whatever is stored in the "tn" variable to the recipient of the feelings, identified by the set phrase "sol.infel-phira.mea".

Infel Phira Uninstall Spell[]

Similar to the Ar tonelico Uninstall Spell, with the difference that it can only be used by IPDs.

cEzYA hymmnos.

#9990 --r mAtyy trix tn=b'0x11000101101'

es tn<=sol.infel-phira.mea


I will be happy to turn you into a song.

Using the Hymmnos Spell #9990, store a target in binary form into the "tn" variable.

and then, realize its materialization out from me, who resides inside of Infel Phira.



Wiki page made and mantained by Shirasagi. Information contributed by aquagon and Lazy.

Don't use without crediting the contributors and the maker of this wiki, or at least putting a link to this wiki.