Relformaide Dictionary:Grammar

From Hacked by Del.icio.us!
Jump to: navigation, search

This page is currently a work-in-progress draft, and as such is not complete nor refined yet. Material below is subject to change or improvement over time.


Relformaide, the subject of this Referata wiki, is a constructed auxiliary language inspired by the Romance branch of the Indo-European family, with agglutinative influences throughout. The name "Relformaide" is the language's own word for "reformed",[1] which represents its initial efforts to reform the Romance languages and amend several defects in their grammar, particularly those related to gender.

Relformaide is written in the standard Latin script employed by English and various other European, American, Austronesian, and indigenous Pan-American and Australian languages. Its alphabet contains all standard letters except C and Q—which are only found in imported surnames—and adds a CH digraph in place of the C.

Thanks to its flexibility, Relformaide can emulate not only its major source languages (English plus the Romance and Germanic families), but also various others from different regions. It is a split-ergative language whose standard word order is Subject–verb–object (SVO). There are no standalone grammatical cases save for the nominative and genitive (possessive) in nouns and pronouns; the accusative in pronouns; and the ergative/unmarked absolutive in certain conditions.

Relformaide consists of several hundred base morphemes, all of which are either free (capable of standing alone as either roots or affixoids) or bound (only found in derivations and inflections).[2] Roots in Relformaide end in consonants (except for s), and are designed and chosen to be free of as much orthographical and semantic conflict as possible. Many are borrowed from various Romance languages, as well as their ancestor Latin; some more are sourced from other Indo-European branches and language families elsewhere.

Among those roots are several dozen adpositions, nearly all of which can also serve as either standalone prepositions, or postpositional mesoclitics attached to the end of a stem. Emphasis is focused on the final major root in any given combination, especially in the case of postpositions. Depending on the circumstances, a Relformaide sentence can consist of several small-to-medium words, or a very long one-word phrase.

Rules

  1. Relformaide uses a Latin-based alphabet of 25 letters for its native roots and words, leaving out C and Q (except in imported surnames and terms) and adding a digraph, CH.
  2. The language is spoken as it is written, with monophthongs, diphthongs, and diacritics to assist in the pronunciation, spelling, and marking of words.
  3. It consists of hundreds of free and bound morphemes, which can either serve as roots or affixes, and are used to form words of various lengths and constructs.
    1. Free morphemes are also called Base roots (or Ziegeltimes), the vast majority of which serve as Core Base roots (Júrekziegeltimes). Core Base roots represent concepts found in most natural languages, human cultures, and various fields.
    2. Many bound morphemes also serve as Termisons (Fimättimes)—suffixes which are placed at the end of most words.[3]
  4. The language's morphology is extremely flexible, and its agglutinative nature also allows users to build long words à la Hungarian, Finnish, Turkish, the Bantu languages, and others of their ilk. In rare cases, extremely long one-word sentences can rival those occasionally found in the indigenous languages of North America, such as Greenlandic.
  5. Words are head-final—in which the emphasis is placed on the last root in any given combination—while standard sentences are head-initial.
  6. The language is non-configurational , allowing it to easily emulate the standards of English, as well as various Romance and Germanic languages. Constituent order is therefore flexible (with a few caveats); Subject–verb–object (SVO) is the declarative default. Statements in the passive voice assume Object–verb–subject (OVS), where the subject (plus any associated modifiers) utilises the ergative intrafix -ieb- if a noun, or remains unchanged if a pronoun; polar questions and optative remarks are formed with Verb–subject–object (VSO).
  7. All words are categorised into nine classes: articles, nouns, pronouns, verbs, adjectives, adverbs, adpositions, conjunctions, and interjections. The first six enable termisons, as does the last one in some cases.
    1. Nouns and pronouns assume gender, depending on whether they are animate (living) or inanimate. Animate subjects and objects end with -o (for masculine/male forms), -a (for feminine/female forms), and -e (for cases where the form's gender is unknown/undetermined, as well as for groups and demonstrative text). Inanimate subjects and objects almost always assume the neuter form;[4] -e is also applied. If articles precede them, then they also assume the noun's gender.
    2. Adjectives and adjectival phrases end with -i; adverbs and adverbial phrases with -u.
    3. Adpositions and conjunctions assume their original root forms.
    4. Interjections can also assume their root forms, or end with -(a)t.
    5. As in English, French, Spanish, and Portuguese, -s is the plural termison. This applies to all articles, nouns, and pronouns, along with instances where adjectives agree with the subsequent plural forms. The remaining parts of speech are never pluralised.
    6. All verbs have -ar as their infinitive ending—the base of a conjugation system that involves -at (for indicative forms), -ant (for the progressive/continuative aspect), -aid (for past participles), and so forth.
    7. Tense is marked with nupé- (recent past), pé- (simple past), plé- (discontinuous past), fé- (future), and péfé- (future in the past) before roots; the present carries no indication.
  8. Numbers, interrogatives, correlatives, determiners, and affects constitute special classes that span across the parts of speech; determiners behave the same way as nouns do.
  9. The nal- and nem- prefixes express negation, and always precedes tense markers; nal- is also the leftmost possible morpheme of any given word.
  10. Several dozen standalone roots—primarily adpositions—also serve as case mesoclitics before the stems they modify.
    1. To indicate possession by a referent, either the genitive -oz- (for inalienable possession) or the possessive -orz- (for alienable possession) is placed between the root and the termison. If a complement is directly related to a possessee, then one of two proprietive markers—-zol- or -ten-—is employed in certain cases. In complex scenarios, the invariable der or den precedes the possessee information.
  11. A special variant intrafix, -uez-/-ouz-, is placed between the stem and either a case suffix, verb termison, or -i in complex plural forms.
  12. With the exception of the word class (-o/-a/-e for nouns; -i for adjectives; -u for adverbs) and tense (nupé-/pé-/plé-/fé-/péfé-) markers, all native morphemes must end in a consonant. Roots cannot end with s, which is reserved as the plural termison.
  13. Base morphemes cannot contain double letters of any sort, but double consonants are permitted in compound forms.
  14. Imported terms (including surnames) are mostly exempt from the previous two rules.
  15. Anytime a vowel termison precedes another vowel, then an ń (cf. the movable nu in Ancient Greek) is placed as a mandatory buffer:
    1. at the start of the next word (if native to Relformaide), or
    2. directly after the termison (before native proper names and imported terms).
  16. As in French, quotations in text are enclosed by wilémètes («»); all other punctuation is used as in English et al.

Graphemics

Orthography

Characters

Standard Latin convention, as well as English, possesses an alphabet of 26 letters:

a b c d e f g h i j k l m n o p q r s t u v w x y z

In Relformaide, there are 25:

a b ch d e f g h i j k l m n o p r s t u v w x y z

The ch digraph substitutes the c, which is otherwise reserved for imported surnames and terms, as is q. Familiar examples of the two stray letters' use can be found in Cousteau and QWERTY.

Relformaide has no c in its native words, as the letter is better represented by its common phonetic equivalents of k, s, and z. Within imported names, the accented variants of ć (for /k/) and ç (for /s/ and /z/) are optional, resulting in the likes of Ćousteau and Luçiano among other forms.

Relformaide uses the same numbers as the Latin script, viz 0 1 2 3 4 5 6 7 8 9.

Diacritics

In Relformaide, a variety of diacritics are used to represent various sounds and assist in letter-marking. As in French, these variants are not part of the language's standard alphabet.

Type Letters Function
a e i o u b c d f g j k l n p r s v y z
Acute á é í ó ú ć ȷ́ ń ŕ ś ý ź
  • Marks the natural sounds of the vowels at the end of syllables.
  • c denotes the /k/ sound in imported surnames and terms.
  • j is used to mutate the root-final g before another j (/dʒ/) in compounds.
  • n is attached to a vowel-initial word after a preceding termison, or the end of a native word before an imported term.
  • r is used to mutate the root-final l before morphemes containing another l as their second or later letter (i.e. in doseŕaili [educational; dosel + -ail + -i]).
  • s is used to mutate the root-final z before affixes starting with ch, h, k, m, n, p, t, or another s.
  • y denotes the /ɪ/ sound at the end of select few roots, amfiebý- (amphibian) among them.
  • z denotes the /ʒ/ (zh) sound in Aiźa (Asia) and mokeźe (landmine).
Grave è Only used when e precedes a consonant and another e at the end of some words (e.g. aumbrède [young human], toapète [small stone = pebble]). In words ending with -eche (e.g. leché [milk]), the last e is marked with an acute instead.
Breve ă ĕ ĭ ŏ ŭ Distinguishes certain root formations from other existing valid combinations. An example is the place name Jămăika, which otherwise means "a female who should be jamming" (jam- + -aik [conditional/subjunctive termison] + -a).
Circumflex â ê î Indicates that the vowel is pronounced exactly like its English literal (/eɪ/, /ɪ ~ iː/, /aɪ/).
Macron ā ē ī ō ū Signifies the start of syllables in some words.
Tilde ĩ ũ Denotes the /ɜː ~ ɝ/ sound found in English words such as herd and hurt.
Umlaut ä ë ï ö ü Placed on the first unaccented vowel of certain roots, whose last letter then becomes the next morpheme's first consonant in certain compounds.
Underdot
  • The underdotted vowels help distinguish root-final -ar/-at/-et/-id &c. from their regular conjugation/affix counterparts; and also substitute the standard -e in internationally recognised neuter nouns. Found in words such as chauklạite (chocolate), Magyạre (Hungary), praujẹte (project), kasinọ (casino), and guavạ (guava).
  • In compounds involving said international nouns, an underdotted l or r follows the root plus original vowel before the next morpheme. Thus, kasinoḷinti (in the casino); guavaṛseulu (only the guava).
  • A preceding o plus underdotted r help buffer certain consonant clusters in compounds, as in námúzoṛganar (achieve honours).
  • Underdotted s serves as epenthesis in certain compounds:
    • When -Vr precedes er-/ir-/or-/ur-, as in vousererive (letter clearance).
    • Between double instances of an: ganant (winning), dranant (being correct).
  • Underdotted y is similarly placed between two consecutive instances of on, as in fonondi (silent).
Cedilla ç Marks the sound of /s/ (and sometimes /z/) in imported surnames and terms.
Dot ġ Only used to mutate and soften the root-final j before another g (/g/) in compounds.
Hook ɓ ƈ ɗ ƒ ɠ ʝ ӄ þ ѵ
  • In compounds, the ƒ is used to mutate the root-final v before h, s, t, z, or another f.
  • Respectively transforming v before b/d/g/j/k/p are ɓ, ɗ, ɠ, ʝ, ӄ, and þ. (The first four are borrowed from the International Phonetic Alphabet, the fifth from Cyrillic, and the last from Scandinavian/Old English.[5])
  • ƈ optionally substitutes the ch diagraph in forms such as ƈúzar (cook), ƈinƈila (chinchilla), leƈé (milk), sandwiƈe (sandwich), and Ƈîna (China).
  • ѵ, which transforms f in front of another v, also originates from Cyrillic.

Punctuation

Relformaide carries the same inventory of punctuation marks found in various Indo-European languages. Its quotation marks (wilémètes, «») are borrowed from French; other symbols ([{<- — _ , ; : . ... ? ! & @ * # % $ € £ ¥ ¢ † ‡ § ~ + × ÷ = º / \ |>}]) remain as-is. (For usage examples, see § Parts of speech: Referential pronouns and § Syntax: Quotations and punctuation.)

Phonology

Relformaide boasts a highly phonemic orthography, meaning that its graphemes correspond to the sounds they represent most of the time.

Letters

The language is pronounced in much the same way as English, as are most of its letterals. All letter names are adapted from existing roots with different meanings.

Letter Sound Name IPA symbol(s) English example(s) Relformaide example(s) Meaning(s) of example(s)
Standard
A ah ade /æ/ art, cat santaline mistletoe
/a(ː)/ ábili able to; can
/eɪ/ day flâme fire
/ɑ/ bar mongar consume
B bey /b/ bay bar, obène, roubi be, house, red
CH chu choule /t͡ʃ/ chat chimèle, ploché jelly, area
D dee dèye /d/ delight dostar, sliedar get, slide
E ey ède /ɛ/ egg esine, vúdèle this, back/posterior
/eɪ/ évitar avoid
/iː/ develop dêvelopar develop
/ɜː ~ ɝ/ herd rgim so/because/therefore
/Ø/ choose maunde world
F eff /f/ fair fásili, sif, baufale easy, if, buffalo
G gee guve /g/ go ganar, rógeli win, green
H hoal hine /h/ humble hanadu, méhoute there, kiwi
/Ø/ uh-oh flohe flea
I ee ide /ɪ/ interval, bid int in
/i(ː)/ síprane surprise
/aɪ/ ice hîdrole water
/ɜː ~ ɝ/ circle sĩrkauze circus
J jay joale /d͡ʒ/ jack jodé, plajé shoe, beach
K kay kaute /k/ kennel korte, eskole line, school
L ell /l/ long lezar, dumale read, cheetah
M em /m/ milk mólen, dúbime though/despite, bear
N en nide /n/ nest nákole, ńaistar, lubone story, make, throne
O oh oate /o/ owned ond, stúdiole, lechoṛsúkrane without/-less, studio, lactose
/oː/ óvile sheep
P pey /p/ pray puerté, sèpe door, seven
R ar roaze /ɹ/ rest ríante, jerowe, besneŕelde parent, manatee, tradition
S ess sẽrte /s/ simmer seristes, sauvar, vasiti cherries, save, intentional
T tee /t/ tote, boot toute all/every
U ub usté /u/ accuse stuk! stop!
/uː/ Tudor úlem under/below
/ɜː ~ ɝ/ hurt dũr during/while
V vee /v/ advance vieslar, suavé whistle, kiss
W wee touvvé /w/ wink weste, túwalu west, from the looks of it/seems that...
/Ø/ show chowe roof
X eks exane /ks/ exit guximi, sorèxe brave, shrew
Y yel yorbe /j/ yellow yoale, yúglane, fluyé, choyaide yawn, walnut, river, pet
/ɪ/ hairy ye tea
/i/ hockey haukýe hockey
/Ø/ choyprini about pampering
Z zed zège /z/ zone zụstime, mozé agreement, my
/ʒ/ Aiźa, mokeźe Asia, landmine
Imported
*C seeta seutène /k/ magic
/s/, /z/ cite
*Q cue kúbạ /kw/ quite

Clusters

Consonants

Another four digraphs are represented. (Even though sch uses three characters, it comprises two letters in Relformaide—s and ch.)

Combo IPA symbol English example Relformaide example Meaning of example
Digraphs
SH /ʃ/ shilling roshólar decide
SCH /ʃ/ Schultz beschérar conjugate
NG /ŋ/ sing lingaili linguistic
NY /ɲ/ Like canyon kanyaune canyon

Ten triple-consonant clusters, all sibilant, are also permitted in words:

Combo IPA symbol English example Relformaide example Meaning(s) of example(s)
Consonant triples
SCHL /ʃl/ schlepping
SCHM /ʃm/ schmaltz
SCHN /ʃn/ schnauzer
SCHR /ʃɹ/ Schroeder
SCHW /ʃw/ Schwartz
SHR /ʃɹ/ Shrove shroute collapse
SKR /skɹ/ scream skríbar write
SPL /spl/ splash resplobe refrain
SPR /spɹ/ spree strouśprini about a street
STR /stɹ/ strudel strúbar, strouze build, street
Vowels

In addition, the language possesses four monophthongs (vowel pairs represented by single sounds); eight diphthongs (vowel pairs sounded together); one semi-diphthong (which acts as either a monophthong or diphthong); one triphthong (comprising three phonemes); and nine hiatuses (vowel pairs sounded separately).

Combo IPA symbol(s) English example(s) Relformaide example(s) Meaning(s) of example(s)
Monophthongs
AU /ɔ(ː)/ caulk, gauze maunde world
/ʌ/ Like mull saume sum
IE /iː/ Bernie, fiend mietire metre
OU /ʊ/ ghoul bouklar, woufe defend, woof
Diphthongs
AI/É /eɪ/ aim Paipo, livré Pope, book
EU /juː/ feud pleuve, jeure rain, rude/offensive
Î /aɪ/ nice sîgloane cyclone/hurricane/typhoon
OA /oʊ/ foal poartar carry
OI /ɔɪ/ boil pointe point
UA /wa/ squad tuami female
UE /wɛ/ pueblo kuelendar torment/torture
UI /wɪ/ squid kuibiène thief/robber
Semi-diphthong
EA /ɪə/, /iː/ appeal, jeans voaŕineale, gleazar, musíkeande airplane, order/arrange, musical instrument
Triphthong
UE /wɛə/ Like swear suertide luck
Hiatuses
AO /a.o/ Like aorta Bilbaone Bilbao
EI /eɪ.ɪ/ deity mâyoneiz mayonnaise
ÉO /eɪ.o/ eon déoze awe
ÍA /iː.æ/ industrial tevolíaku on the table
ÍE /iː.ɛ/ Diego veyne, póliezíena watcher, policewoman
IO /i(ː).o/ lion raidiole radio
ÍU /iː.u/ diurnal díuve kite
OE /o(ː).ɛ/ poem pze poem

Inventories

Relformaide carries almost the same individual phonemes as those of English—26 for consonants and 16 for vowels. (Corresponding graphemes are bolded.)

Consonants Labial Dental Alveolar Postalveolar Palatal Velar Glottal
Nasal m [m] n [n] ny [ɲ] ng [ŋ]
Stop Voiceless p [p] t [t] k [k] - [ʔ]
Voiced b [b] d [d] g [g]
Affricate Voiceless ch [t͡ʃ]
Voiced j [d͡ʒ]
Fricative Voiceless f [f] th [θ] s [s] sh [ʃ] h [h]
Voiced v [v] z [z] ź [ʒ]
Approximant Plain l [l] r [ɹ] y [j] w [w]
Labial ui [ɥ]
Vowels Front Central Back
Close i/í/ie/ea [i(ː)] ú [uː]
u [u]
Near-close i [ɪ] ou [ʊ]
Mid e [ɛ] é [e] V [ə] /ĩ/ũ [ɜː] o/ó [o(ː)] au [ʌ/ɔ/ɒ]
Open a [æ] a/á [a(ː)] ar [ɑ]

IPA correspondence

IPA a æ ɑ ɒ b d e ɛ ɜː ə f g h i ɪ j ju k ks l m n ŋ ɲ o ʌ ɔ(ː) ɔɪ p ɹ s ʃ t θ u ʊ v w ɥ z ʒ
Grapheme(s) a a a au á î b d j é e/è ẽ/ĩ/ũ ai/â/é a/e/i/o/u f g/ġ h i i ea/ê/ie/í/ý y eu k x l/ḷ m n/ń ng ny o oa ó au au oi p r/ŕ/ṛ s/ś sch/sh t ch th u ú ou v u/w ui z ź

Termisons

Relformaide has six uniliteral termisons, of which five (a, i, o, s, and u) are always pronounced in full at the end of words. Depending on the word, e is either pronounced or left silent, as shown in these tables for the roots ed-, ótr-, lapin-, strouz-, lech-, and sandwich-. S always follows a vowel termison (except for u) in plural forms.[6]

Example 1: ed- (young/youth)
Form IPA English meaning
ède ɛd youngster, kid, lad, tyke
èdes ɛds youngsters, kids, lads, tykes
edo ɛ.do boy
eda ɛ.dɑ girl
edi ɛ.di young
edu ɛ.du youngly (Rare)
Example 2: ótr- (other/different)
Form IPA English meaning
ótré oː.tɹeɪ another one
ótrés oː.tɹeɪs others
ótro oː.tɹo another one (masculine)
ótra oː.tɹɑ another one (feminine)
ótri oː.tɹi different
ótru oː.tɹu differently
Example 3: lapin- (rabbit)
Form IPA English meaning
lapine læ.pɪn rabbit
lapines læ.pɪns rabbits
lapino læ.pɪ.no rabbit buck
lapina læ.pɪ.nɑ rabbit doe
lapini læ.pɪ.ni rabbit-associated
lapinu læ.pɪ.nu like a rabbit
Example 4: strouz- (street)
Form IPA English meaning
strouze stɹʊz street
strouzes stɹʊ.zɪs streets
strouzi stɹʊ.zi street-associated
strouzu stɹʊ.zu like a street
Example 5: lech- (milk)
Form IPA English meaning
leché lɛ.tʃeɪ milk
lechés lɛ.tʃeɪs servings of milk
lechi lɛ.tʃi milk-associated
lechu lɛ.tʃu like milk
Example 6: sandwich- (sandwich)
Form IPA English meaning
sandwiche sænd.wɪtʃ sandwich
sandwiches sænd.wɪ.tʃɪs sandwiches
sandwichi sænd.wɪ.tʃi sandwich-associated
sandwichu sænd.wɪ.tʃu like a sandwich

Stress

In most Relformaide words, stress falls on the final or lone syllable, and also with long vowels (represented by grave accents): nâye (dog), touvarde (two million), pla (beach), koulibré (hummingbird), telar (deceive/trick).

In some words like nemnonde (something unsilent), the focus is on the syllable with the accent/long vowel. Many causative verbs exhibit stress before the -inz suffix used to form them, as in frólinzar (enlighten).

Morphophonology

Ablaut

A handful of Relformaide roots exhibit ablaut, a morphophonological process otherwise known as apophony. Ablaut involves the change of vowel sounds to form different words.

Indo-European

The more predominant type, Indo-European ablaut, uses i [ɪ] ~ a [æ] ~ au [ɔ] to convey deixis.

Base root (e) Proximal (i) Medial (a) Distant (au) Etymon(s)
English meaning
*aubem aubim aubam aubaum Iteri lʌboli (little)
size small midsized big
aulxem aulxim aulxam aulxaum French ault + Latin proximus (nearest)
height short medium tall
esen esin esan esaun Estonian ese/Finnish esine
thing this that yon
*hen hin han haun Basque hemen (here)/han (there)
place/location here there yon(der)
proxem proxim proxam proxaum Latin proximus
distance/length near far beyond
woxem woxim woxam woxaum Lower Sorbian wob (during) + Latin proximus
duration short medium long
Gender-based

This pattern, involving e [ɛ] ~ o [o] ~ oa [oʊ] ~ a [a], is only found in the gender adjective series.

Base root (i) Neuter (e) Masculine (o) Unisex (oa) Feminine (a) Etymons
English meaning
sugim sugem sugom sugoam sugam Estonian/Veps/Võro sugu; cf. Finnish suku
grammatical gender
*tuim tuom tuam À priori root
natural gender

Allomorphy

Relformaide's most predominant allomorph is the neuter noun marker, -e, and its plural counterpart -es.

  • -e itself is unvoiced (/Ø/) in many nouns, such as paundale (zebra; /pɔn.dæl/), proximade (vicinity; /prok.sɪ.mæd/), and esène (thing; /ɛ.sɛn/).
  • It takes on the pronunciation of /eɪ/ as in French, when marked as in words like nèke (cat; /tʃo.neɪ/), ábilídé (ability; /aː.bɪl.iː.deɪ/), and leché (milk; /lɛ.tʃeɪ/).
  • The plural -s (/s/) is always pronounced: paundales (zebras; /pɔn.dæls/), bevaurdes (beverages; /bɛ.veɪds/); esènes (things; /ɛ.sɛns/); nèkes (cats; /tʃo.neɪs/).
  • In certain words whose roots end with z, -e remains unvoiced in the singular, but is sounded as a soft i (/ɪ/) in the plural: strouze (street; /stɹʊz/) → strouzes (street; /stɹʊ.zɪs/).
  • As already mentioned, -és is mandatory for plurals of roots ending in -ech: leché (/lɛ.tʃeɪ/) → lechés (/lɛ.tʃeɪs/). Those ending in -ich adhere to the normal rules for -e(s): sandwiche (/sænd.wɪtʃ/) → sandwiches (/sænd.wɪ.tʃɪs/).
  • Except in the case of -èCe forms, verbal nouns always end with (/eɪ/): manké (eating; /mæn.keɪ/), sauté (jumping; /sɔ.teɪ/), lèze (reading: /lɛz/), noagèle (swimming; /noʊ.gɛl/).

Movable Ń

Relformaide employs a special character, ń, which buffers hiatuses between words in similar fashion to Ancient Greek's "movable nu", and carries no semantic meaning.

  • When native words are involved, the ń is placed at the start of the following word: boavondi ńagrotabode (farm without cows), un'eda ńeskoladi (a girl at school), Prósimu ńaistait les pastèles (Please make the cakes).
  • If imported or capitalised terms follow native words, then ń is attached to the latter's termison: Senoroń Obama (Mr. Obama), Jaunoń Updike (John Updike), Guimbrovrageń: Aulttemi Pleuƒfoareste (FernGully: The Last Rainforest).
  • Vice versa, the regular procedure applies: Clementine ńättrúlat mozo jurèke ńoid naulótra nalhaizat. (Clementine warms my heart like no other lady can.)

Phonotactics

Relformaide's syllable structure is (s)(C)2(V|M|D)2(C)2(s|z); common syllable patterns include CVC, CDC, CV, CCV, sCCV, VC, VCC, and VCCs. Standalone roots and affixes follow a (s)(C)2(V|M|D)2(C)2(z) pattern; CVCVC, CVC, CVVC, CDC, and VC are prevalent. Standalone termisons are either an s (plural) or (V|D)(C(C)).

All words must have at least one vowel; abbreviations and uniliteral roots are exempt. Words cannot start with /ks/ (⟨x⟩) or /ŋ/ (⟨ng⟩).

Onset

The onset of a Relformaide word can only reach up to three consonants ( (s)(C)C ), with /b t͡ʃ d f g h d͡ʒ s v ks z/ ⟨b ch d f g h j s v x z⟩ prohibited as the second one in any given cluster; words and roots beginning with vowels omit onsets.

Permitted onset clusters in Relformaide
First consonant Second consonant
/k/ ⟨k⟩ /l/ ⟨l⟩ /m/ ⟨m⟩ /n/ ⟨n⟩ /p/ ⟨p⟩ /ɹ/ ⟨r⟩ /t/ ⟨t⟩ /w/ ⟨w⟩ /j/ ⟨y⟩
/b/ ⟨b⟩ /bl/ ⟨bl⟩ /bɹ/ ⟨br⟩ /bw/ ⟨bu + V⟩ /bj/ ⟨be+u⟩
/t͡ʃ/ ⟨ch⟩ /t͡ʃw/ ⟨chu + V⟩ /t͡ʃj/ ⟨che+u⟩
/d/ ⟨d⟩ /dɹ/ ⟨dr⟩ /dw/ ⟨du + V⟩ /dj/ ⟨de+u⟩
/f/ ⟨f⟩ /fl/ ⟨fl⟩ /fɹ/ ⟨fr⟩ /fw/ ⟨fu + V⟩ /fj/ ⟨fe+u⟩
/g/ ⟨g⟩ /gl/ ⟨gl⟩ /gɹ/ ⟨gu⟩ /gw/ ⟨gu + V⟩ /gj/ ⟨ge+u⟩
/h/ ⟨h⟩ /hj/ ⟨he+u⟩
/d͡ʒ/ ⟨j⟩ /d͡ʒw/ ⟨ju + V⟩ /d͡ʒj/ ⟨je+u⟩
/k/ ⟨k⟩ /kl/ ⟨kl⟩ /kɹ/ ⟨kr⟩ /kw/ ⟨ku + V⟩ /kj/ ⟨ke+u⟩
/l/ ⟨l⟩ /lw/ ⟨lu + V⟩ /lj/ ⟨le+u⟩
/m/ ⟨m⟩ /mw/ ⟨mu + V⟩ /mj/ ⟨me+u⟩
/n/ ⟨n⟩ /nw/ ⟨nu + V⟩ /nj/ ⟨ne+u⟩
/p/ ⟨p⟩ /pl/ ⟨pl⟩ /pɹ/ ⟨pr⟩ /pw/ ⟨pu + V⟩ /pj/ ⟨pe+u⟩
/ɹ/ ⟨r⟩ /rw/ ⟨ru + V⟩ /ɹj/ ⟨re+u⟩
/s/ ⟨s⟩ /sk/ ⟨sk⟩ /sl/ ⟨sl⟩ /sm/ ⟨sm⟩ /sn/ ⟨sn⟩ /sp/ ⟨sp⟩ /st/ ⟨st⟩ /sw/ ⟨su + V⟩ /sj/ ⟨se+u⟩
/ʃ/ ⟨s(c)h⟩ /ʃl/ ⟨schl⟩ /ʃm/ ⟨schm⟩ /ʃn/ ⟨schn⟩ /ʃɹ/ ⟨s(c)hr⟩ /ʃw/ ⟨s(c)hu + V⟩ /ʃj/ ⟨s(c)he+u⟩
/t/ ⟨t⟩ /tɹ/ ⟨tr⟩ /tw/ ⟨tu + V⟩ /tj/ ⟨te+u⟩
/v/ ⟨v⟩ /vl/ ⟨vl⟩ /vɹ/ ⟨vr⟩ /vw/ ⟨vu + V⟩ /vj/ ⟨ve+u⟩
/z/ ⟨z⟩ /zl/ ⟨zl⟩ /zw/ ⟨zu + V⟩ ⟨zw⟩ /zj/ ⟨ze+u⟩

Nucleus

A nucleus consists of either a vowel (V), a monophthong (M), or a diphthong (D). Vowels end all Relformaide articles, nouns, adjectives, and adverbs. One diphthong, /eɪ/ (⟨é⟩), is found in all basic tense augments ( (C)CD ) and neuter nouns containing it at the end ( (s)(C)CD ).

Coda

The coda consists of at least one consonant; as with onsets, up to three are allowed ( C(C)(s|z) ). In clusters, /ɹ w j/ ⟨r w y⟩ are prohibited as the last possible letter. All bare roots end with codae, as do adpositions, conjunctions, and some interjections.

Permitted coda clusters in Relformaide
Penultimate consonant Last consonant
/b/ <b> /t͡ʃ/ ⟨ch⟩ /d/ ⟨d⟩ /f/ ⟨f⟩ /g/ ⟨g⟩ /d͡ʒ/ ⟨j⟩ /k/ ⟨k⟩ /l/ ⟨l⟩ /m/ ⟨m⟩ /n/ ⟨n⟩ /p/ ⟨p⟩ /s/ ⟨s⟩ /t/ ⟨t⟩ /v/ ⟨v⟩ /ks/ ⟨x⟩ /z/ ⟨z⟩
/b/ ⟨b⟩ /bs/ ⟨-bes⟩
/d/ ⟨d⟩ /ds/ ⟨-des⟩
/f/ ⟨f⟩ /fs/ ⟨-fes⟩ /ft/ ⟨ft⟩
/g/ ⟨g⟩ /gs/ ⟨-ges⟩
/k/ ⟨k⟩ /ks/ ⟨-kes⟩
/l/ ⟨l⟩ /lb/ ⟨lb⟩ /lt͡ʃ/ ⟨lch⟩ /ld/ ⟨ld⟩ /lf/ ⟨lf⟩ /lg/ ⟨lg⟩ /ld͡ʒ/ ⟨lj⟩ /lk/ ⟨lk⟩ /lm/ ⟨lm⟩ /ln/ ⟨ln⟩ /lp/ ⟨lp⟩ /ls/ ⟨-les⟩ /lt/ ⟨lt⟩ /lv/ ⟨lv⟩ /lks/ ⟨lx⟩ /lz/ ⟨lz⟩
/m/ ⟨m⟩ /mb/ ⟨mb⟩ /mt͡ʃ/ ⟨mch⟩ /mf/ ⟨mf⟩ /mg/ ⟨mg⟩ /md͡ʒ/ ⟨mj⟩ /mk/ ⟨mk⟩ /mp/ ⟨mp⟩ /ms/ ⟨-mes⟩ /mt/ ⟨mt⟩ /mv/ ⟨mv⟩ /mks/ ⟨mx⟩ /mz/ ⟨md⟩
/n/ ⟨n⟩ /nb/ ⟨nb⟩ /nt͡ʃ/ ⟨nch⟩ /nd/ ⟨nd⟩ /nf/ ⟨nf⟩ /ŋ/ ⟨ng⟩ /nd͡ʒ/ ⟨nj⟩ /nk/ ⟨nk⟩ /np/ ⟨np⟩ /ns/ ⟨-nes⟩ /nt/ ⟨nt⟩ /nv/ ⟨nd⟩ /nks/ ⟨nx⟩ /nz/ ⟨nz⟩
/p/ ⟨p⟩ /ps/ ⟨-pes⟩ /pz/ ⟨pz⟩
/ɹ/ ⟨r⟩ /ɹb/ ⟨rb⟩ /ɹt͡ʃ/ ⟨rch⟩ /ɹd/ ⟨rd⟩ /ɹf/ ⟨rf⟩ /ɹg/ ⟨rg⟩ /ɹd͡ʒ/ ⟨rj⟩ /ɹk/ ⟨rk⟩ /ɹl/ ⟨rl⟩ /ɹm/ ⟨rm⟩ /ɹn/ ⟨rn⟩ /ɹp/ ⟨rp⟩ /ɹs/ ⟨-res⟩ /ɹt/ ⟨rt⟩ /ɹv/ ⟨rv⟩ /ɹks/ ⟨rx⟩ /ɹz/ ⟨rz⟩
/s/ ⟨s⟩ /sk/ ⟨sk⟩ /sp/ ⟨sp⟩ /st/ ⟨st⟩
/t/ ⟨t⟩ /ts/ ⟨-tes⟩ /tz/ ⟨tz⟩
/v/ ⟨v⟩ /vs/ ⟨-ves⟩

Graphotactics

Initial letters

A native Relformaide word can begin with any of 24 letters, but never x (outside of imports).

a b ch d e f g h i j k l m n o p r s t u v w x y z

Final consonants

There are 19 consonants that roots can end with; the vowels and s are reserved as termisons.

a b ch d e f g h i j k l m n o p r s t u v w x y z

Bigrams

With 25 letters in its alphabet, the maximum number of possible two-letter combinations in Relformaide words and syllables is 625 (25²).

  1. Unlike most natural languages, and in similar fashion to Lojban's gismu, Relformaide does not allow double letters (e.g. bb, dd, ee, ff, nn, ss) in its base roots.
    1. Double vowels only occur in surname imports, as in Boorman and Deere.
    2. Double consonants are permitted in native compounds, such as addez (tell; ad-, to object + dez-, say).
  2. In compounds, the final z of a root becomes ś before ch, h, k, m, n, p, s, and t.
  3. Similarly, v becomes ƒ before ch, s, t, z, or another f; f turns into ѵ before another v.
  4. The letter x cannot begin any word apart from surname imports (most notably Xavier); nor can roots end with s (the plural marker).
  5. Among consonant-only bigrams ending with x, only lx/nx/rx/yx are allowed within words; the rest are disallowed outside imports.
  6. chh is not permitted in base roots, but permitted in compounds and imports. Also following this rule are homophones gh (= f/g), ph (= f), th (= d/f), rh (= r), wh (= w), wr (= r), sw (= sui) (only in imports), and tw (= tui).
  7. kh is conditionally permitted in base roots like lakh (an Indian term for 10,000); otherwise, it turns into k.
  8. These pairs are strictly prohibited:
    1. xk/xs/xz (within standalone roots; = xoṛk/xoṛs/xoṛz in compounds)
    2. sz/ae (except in imports)

After excluding double vowels aa/ee/ii/oo/uu and double consonants ss/xx, one is left with 618 (25² - 7) possible bigrams in native roots. From here, the remaining number depends on various factors as shown in the table below:

Elimination rule Combos left
Base Compound
At start Within At end At start Within At end
Rest of double consonants (bb, dd, ll, tt, &c.) 600 600 600 600 618 600
Pairs xk, xs, and xz 597 597 597 597 618 597
Rest of x-initial pairs (e.g. xd, xg, xn, xt) 576 576 576 576 618 576
sz 575 575 575 575 617 575
Pairs chh, gh, ph, rh, th, and wh 569 569 569 569 617 569
hch/vch/zch 566 566 566 566 617 566
Rest of C + ch pairs (except sch) 552 566 566 552 617 566
l/r + b/d/f/g/j/v 540 566 566 540 617 566
Rest of C + b/d/f/g/j/v pairs 444 566 470 444 617 470
lx/nx/rx/yx 440 566 470 440 617 470
Rest of C + x pairs 425 551 455 425 602 455
Rest of C + s/C + z pairs 390 551 420 390 602 420
sh 390 551 419 390 602 419
Rest of C + h pairs (except kh) 380 551 409 380 602 409
lk/rk 378 551 409 378 602 409
Rest of C + k pairs (except sk) 363 551 394 363 602 394
C + l pairs (except [b/f/g/k/p/s]l) 351 551 394 351 602 394
C + m/n/p/t pairs (except sm/sn/sp/st) 283 551 394 283 602 394
C + r pairs (except [b/d/f/g/k/p/t]r) 272 551 394 272 602 394
C + w pairs (except zw) 255 551 394 255 602 394
ae 254 550 393 254 601 393
ih, iy, and uy 251 550 393 251 601 393
iu and oe 249 550 391 249 601 391
ua, ue, and uo 246 550 388 246 601 388
uw 245 550 388 245 601 388
pe/le/fe 245 550 385 245 601 385
Rest of C + V pairs 245 550 293 245 601 293
Rest of V + V pairs 245 550 274 245 601 274

Therefore, one is left with 245 possible opening bigrams, or 39.2% of the initial maximum (as seen in the bigram chart).

After removing bigrams which end with vowels and s (per language rules), and factoring in the tense markers /plé//péfé, there are 134 potential biliteral morphemes (or 21.44% of the initial maximum, or 54.694% of all legal combos). Of these, 10 (or 7.463%, not counting plural markers) are documented in Relformaide.[7]

Native roots can also end with one of 274 possible combinations.

Intra-letter changes

In forming compound words, several letter combinations may appear awkward if not checked. As a result, Relformaide has several insertion/mutation rules to prevent them from surfacing, as the table below demonstrates:

  • -z and -v before certain initial letters
  • -v-C ̡ before b, d, g, j, k, and p
  • -f before another v
  • Insertion of -oṛ- between difficult consonant clusters
Last letter in compound root #1 First letter in compound root #2
a b ch d e f g h i j k l m n o p r s t u v w y z
b ba bb bch bd be bf boṛg bh bi boṛj bk bl bm bn bo bp br bs bt bu bv bw by bz
ch cha choṛb choṛch choṛd che chf choṛg chh chi choṛj chk chl chm chn cho choṛp chr choṛs choṛt chu chv chw chy choṛz
d da db dch dd de df dg dh di dj dk dl dm dn do dp dr ds dt du dv dw dy dz
e[mut 1] era éb éch éd ere éf ég éh eri éj ék él ém én ero ép ér és ét eru év éw éy éz
f fa foṛb fch foṛd fe ff foṛg fh fi foṛj foṛk fl fm fn fo foṛp fr fs ft fu ѵv fw fy fz
g ga goṛb goṛch goṛd ge goṛf gg gh gi ȷ́j gk gl gm gn go goṛp gr gs gt gu goṛv gw gy gz
h[mut 2] ha hb hoṛch hoṛd he hoṛf hoṛg hh hi hoṛj hk hl hm hn ho hp hr hs hoṛt hu hoṛv hw hy hz
j ja joṛb joṛch joṛd je joṛf ġg jh ji jj joṛk jl jm jn jo joṛp jr joṛs joṛt ju joṛv jw jy joṛz
k ka koṛb koṛch koṛd ke koṛf koṛg kh ki kj kk kl km kn ko kp kr ks kt ku kv kw ky kz
l la lb lch ld le lf lg lh li lj lk ll lm ln lo lp lr ls lt lu lv lw ly lz
m ma mb mch md me mf mg mh mi mj mk ml mm mn mo mp mr ms mt mu mv mw my mz
n na nb nch nd ne nf ng nh ni nj nk nl nm nn no np nr ns nt nu nv nw ny nz
p pa pb pch pd pe pf pg ph pi pj pk pl pm pn po pp pr ps pt pu pv pw py pz
r ra rb rch rd re rf rg rh ri rj rk rl rm rn ro rp rr rs rt ru rv rw ry rz
t ta tb tch td te tf tg th ti tj tk tl tm tn to tp tr ts tt tu tv tw ty tz
v va ɓb ƒch ɗd ve ƒf ɠg vh vi ʝj ӄk vl vm vn vo þp vr ƒs ƒt vu vv vw vy ƒz
w wa wb wch wd we wf wg wh wi wj wk wl wm wn wo wp wr ws wt wu wv ww wy wz
x xa xoṛb xch xoṛd xe xf xoṛg xh xi xoṛj xoṛk xl xm xn xo xp xr xoṛs xt xu xoṛv xw xy xoṛz
y ya yb ych yd ye yf yg yh yi yj yk yl ym yn yo yp yr ys yt yu yv yw yy yz
z za zoṛb śch zoṛd ze zoṛf zoṛg śh zi zoṛj śk zl śm śn zo śp zr śs śt zu zv zw zy zz
  1. Only in the tense augments pé-, plé-, fé-, and péfé-.
  2. Found mainly in foreign place names.

For roots which end with a consonant followed by l or r, a regular -i- interfix is added before another consonant. Root-final clusters following this rule include bl, fl, kl, pl, sl, br, fr, kr, pr, and tr.

Morphology

Relformaide consists of several hundred base morphemes, all of which are either free (capable of standing alone) or bound (only found in derivations and inflections).[2] Depending on their function and position in a word, they can either serve as roots (capable of standing alone semantically) or affixes (capable of inflecting a root).

Roots

Source languages

Roots comprise the vast majority of Relformaide's base morphemes; all of them end with a consonant (save for s) and vary in length. Many are borrowed from earlier and more modern varieties of English, as well as the Romance languages (French , Spanish , Portuguese , Italian , Romanian , &c.) and their ancestor, Latin . Other sources include:

Lengths

Note Note:
The 3+-letter morphemes shown in this section are chosen at random via Semantic MediaWiki. A different list will therefore be displayed on most page visits unless cached.

Five standalone uniliteral roots, all part of the Core (Júrèke) group, are the shortest in Relformaide:

  • l- (DEF — the art def)
  • m- (1 — 1st person; I/me/my; we/us/our prn)
  • t- (2.FAM — 2nd person, familiar; you[r] prn)
  • b- (be v)
  • v- (go)

There are several two-letter morphemes, including:

  • -ad- (LOC/DAT — at/to)
  • -av- (PERF — perfect verb)
  • -et- (DIM — diminutive suff)
  • -ieb- (ERG — ergative marker)
  • -id- (state/quality of being)
  • -if- (characterised by suff)
  • -oz- (GEN — genitive/possession marker)
  • un- (INDF — a[n]/some art indef)

Most basic roots contain 3-5 letters, with some bearing six or seven:

  • 3: bel — beautiful, pretty, pulchritudinous, handsome, fair, lovely, beauty, belle; ond — without, lacking, empty, void, deficient, devoid, free of, bereft, lack, -less, -free, abessive case marker; ver — back, backward, backwards, reverse, revert, turn back, non-, un-, de-, dis-, mis-, mal-, im-, in-, il-, ir-, a-, counter-, revertive marker; ben — well, in good health; chow — roof; kây — cay, key; men — hand; sab — know, be aware of, be acquainted with, be familiar with, have knowledge of; han — there; zev — it is heard, I hear it, auditory evidence marker
  • 4: krez — grow, growth; flev — vein; trug — turn, spin, rotate, change direction; paum — palm; guod — enjoy, take pleasure in, enjoyment, pleasure; déoz — awe; pauz — pause, interrupt, take a break, freeze, stop, halt, cease, interruption, intermission, interval, break; trul — comfortable, cosy, snug, comfort; guld — pay, spend, payment; dual — fence
  • 5: kochul — transvestite, crossdresser, cross-dresser, drag, drag king, drag queen, crossdress, cross-dress, transvest; paxul — colon; gímel — pin; nímel — in charge of, master, lord, maestro, boss, comptroller, controller, manager, supervisor, overseer, administrator, admin, control, subdue, monitor (over), supervise, manage, handle, oversee; búrov — donkey; hóleg — shadow, shade; skild — draw, sketch, doodle, drawing; cháluz — steel; shoar — each other, one another, mutual, reciprocal marker; meubl — furniture
  • 6: píjaun — pigeon; véloar — sail, curtain, drape, veil, cover; haurig — chariot; shólob — chute; shaman — shaman, witch doctor; doavan — award, reward, award ceremony; pajaul — shark; fushil — fuchsia; dorimb — grenade; akolem — academy
  • 7: êlekter — electric, electric device, electro-; fouskal — blister; moarxel — coffin, casket; múskrib — compose, write music, composition; Yẽrbíom — erbium; drauven — raisin; gastrov — gastropod; politik — politics, political; sitrauz — citrus; ziguist — flicker, glimmer

Beyond this limit, relatively few are longer:

  • 8: shárazan — utopia, dreamland; zibaldon — hodgepodge, miscellany, miscellaneous, misc., jumble; sínagaug — synagogue; protévon — primate; Paskuyen — Easter egg; skaureal — freezer; dolashum — circulatory, circulatory system; galẽriev — gallery; moarprin — obituary, obit; zaumbiel — zombie
  • 9: teriyakin — teriyaki; Silikogen — Silicon Valley; goumieder — hullabaloo; kaunfrenz — conference, convention; boumẽrang — boomerang, return to sender, rebound, boomerang child; Tairokíem — Earth Day; kromosoam — chromosome; silikorut — silicon chip, microchip, microprocessor; testament — testament; orangutan — orangutan

The longest standalone Core roots are:

Root Table Length Meaning
auraundismaunt Roots/auraundismaunt 14 arrondissement
estroadinair Roots/estroadinair 12 extraordinary
lábrotaurit Roots/lábrotaurit 11 laboratory
lab
helikauptẽr Roots/helikauptẽr 11 helicopter
chopper
kauntravẽrz Roots/kauntravẽrz 11 controversial
disputed
kaumpromiez Roots/kaumpromiez 11 compromise
pampelmouz Roots/pampelmouz 10 grapefruit
tamboadrol Roots/tamboadrol 10 whirlpool
dîmenshaun Roots/dîmenshaun 10 dimension
sîglopiedý Roots/sîglopiedý 10 encyclopedia
foaraskoap Roots/foaraskoap 10 stethoscope
hapokaisil Roots/hapokaisil 10 community spirit
melíagriez Roots/melíagriez 10 turkey
saulfoguor Roots/saulfoguor 10 sulphur spring
sulfur spring
astregánul Roots/astregánul 10 astrology

Affixes

Affixes in Relformaide are either inflectional/derivational markers, or normal roots modifying the meaning of a preceding stem. As in the Romance languages and English, they are either prefixes (at the start of words) or suffixes (at the end). Many in the latter group represent various word classes as Termisons (or Fimättimes), so called because of their purpose and position.

Nouns, pronouns, and articles

Relformaide reserves three vowels to mark gender at the end of all words except imperatives, standalone adpositions, conjunctions, and interjections:

  • -o (MASC — male)
  • -a (FEM — female)
  • -e (N — neuter/unspecified/pan-gender; only in nouns, pronouns, and articles)

Adjectives

All adjectives and adjectival phrases (ADJ) end in -i, and are derivational. Comparatives are formed by suffixing -ausm after the root, and superlatives with -ausom; the same applies for adverbs. (The corresponding inverse suffixoids are -eusm and -eusom.)

Adverbs

Adverbs and adverbial phrases (ADV) end in -u, and are also derivational. -u is equivalent to English -ly, French -ment, and Spanish -mente in most given cases.

Verbs

Conjugation of Relformaide verbs is straightforward, and involves the -ar termison for infinitives; -at for indicative forms; -ait for the imperative; -ant for the continuative and progressive moods (present participle); -aid for past participles; and so forth. An -av between the root and select verbal suffixes gives variants in the perfect aspect.

Of Relformaide's five tenses, only the present is unmarked. The rest are handled by five prefixes, henceforth known as augments: nupé-/nuper- (recent past), pé-/per- (simple past), plé-/pler- (discontinuous past), fé-/fer- (future), and péfé-/péfer- (future in the past). The extra r in each one serves as an interfix before vowel-initial roots.

Nal- signifies negation, and occupies a word's leftmost possible slot.

Interjections

Interjections either assume their unmarked root forms, or end with -(a)t.

Adpositions and conjunctions

Adpositions and conjunctions always assume their root forms, and take no termisons.

Plurals

As in English, French, Spanish, and Portuguese, plural forms are denoted with an -s at the end of nouns, pronouns, and articles, as well as any adjectives preceding nouns. This is the final possible slot in Relformaide's word formation system; as a rule, adverbs do not take plural endings.

  • -os (MASC-PL — male plural)
  • -as (FEM-PL — female plural)
  • -es (N-PL — neuter plural)
  • -is (ADJ-PL — plural adjective)

In complex compound forms, -*s (as applies to nouns) becomes -uez/-ouz before the focus root, as in the possessive verb möutozar (to be ours) and the case-inflected tuameduezómistu (for the girls' sake).

Possessive adjectives inflect for gender with prefixoids tuom- and tuam-: tüortozi (of yours; male singular), tüallumöutozi (of theirs; female plural agreeing with plural noun).

When dealing with plural pronouns at the end of complex verb conjugations, -es, -os, or -as is employed: Álïsmmatas (We give up; female speakers), Benaigavirújoṛtatos (You've done really well; to male listeners), Noutävvoalaijlumatés (They're about to fly now; referring to unspecified subjects or a crowd/group/team).

Affix charts

Non-termison
Stem Meaning Type
Prefixes
nal-/naŕ- predicate/complement negation marker Free
nupé-/nuper- recent past tense; just, recently Bound
pé-/per- simple past tense; was, were, did Bound
plé-/pler- discontinuous past tense (used to); former(ly), ex- Bound
fé-/fer- future tense; "-to-be" Bound
péfé-/péfer- future-in-the-past tense (would) Bound
nem- not among, non-, un- Free
rel- re-, repeated, again, anew Free
ver- back(wards) Free
vuiz- before (in sequence) Free
suiv- after (in sequence) Free
saub- sub- Bound
supẽr- super- Free
hîpẽr- hyper- Free
Affixoids
-ad- to (of indirect objects); at (of places) Free
-aup- to(wards) (a location or thing) Free
-aurd- for (a place or thing, in designation) Free
-int- in (a place or thing) Free
-tranz- through (a place or object) Free
-tug- out of, outside (a place or thing) Free
-tüvv- out of (a place or thing, in departure) Free
-weg- off the surface of (an object of origin) Free
-aseb- with, accompanied by (someone/something) Free
-auvek- with, accompanied by (as part of a group) Free
-emek- with (an instrument/tool/device) Free
-proxim- near, next to, close by (a location or thing); nearly/almost # Free
-prev- before; pre- Free
-áprev- after; post- Free
-aiv- each/every (prefix); per object (suffix) Free
Suffixes
-ard 1,000,000x (one million [of], in numbers); augmentative suffix (for other words) Bound
-raz times #, #-fold; multiple of x (with numbers) Free
-tam about/approximately/roughly # (only with numbers) Free
-tem #st/nd/rd/th (only in ordinal numbers) Bound
-tim x/# (in fractions); a small piece/portion/segment (of inanimate subjects) Free
-tuim #-plet (when referring to twins) Free
-esil (originating) from a place, person, or thing Free
-íak on (the surface of) Free
-ósot up to the point of (a place), until Free
-äffim ending up at (a place) Free
-ómist for/(dedicated) to (in media); for the benefit/sake of Free
-opon against, opposing, anti- Free
-anfil prone/susceptible/liable to (something) Free
-oajem brief(ly), in a flash; momentane aspect Free
-ámel frequent(ly), often, occasional(ly) (of an action or state) Free
-oguil sometimes, part-time Free
-esel maybe, probably, perhaps Free
-úpot almost, nearly (also in numbers) Free
-ábil able to; can Free
-euz full of... Bound
-ond without, -less Free
-seul -only, exclusively, exclusive to Free
-oid like, resembling; -oid/-ine (of animals) Free
-ausik similar to, acting/behaving like Free
-kuam more than/much as/less than (an object) (after plain adjectives, and comparatives formed with -ausm/-eusm) Free
-úvel step-(relative); ersatz/mock (before other roots) Free
-úbel bad, poor, shoddy, inferior Free
-úvam excellent Free
-truz set or group of... Free
-mult collection of... Free
-budin team of... (only with animate subjects) Free
-tánul the study/field of... Free
-ail of or pertaining to a place, person, or thing Bound
-íen someone who is (from a place), comes from/lives in/resides in (a place), does (an action), plays (a game/sport), specialises in Bound
-íer something that does... (with action verbs) Bound
-íom -ium (in chemistry) Free
-ead medium-sized (chiefly in clothing/footwear terms) Bound
-et small (in size), miniature, brief (in time) Bound
-eb baby/newborn (of animals), seedling/sapling/sprout (of plants) Free
-ed young in age Free
-eld old in age, senior, elder Free
-ausm comparative (cf. English -er) Free
-ausom superlative (cf. English -est) Free
-eusm less... Free
-eusom least... Free
-ẽrgim because (of), thanks to Free
-inz causative verb marker (to cause/make...) Free
-iz the act of turning/transforming into/converting to... Bound
-id the quality or state of being... Free
-if the nature of (being)/characterised by... (adjectives and adverbs) Bound
-aig the use/action of... Free
-uit/-oip intensifier Bound
-ieb ergative marker (chiefly used for passive statements in OVS, as well as in appositive phrases) Bound
-eun topic marker (cf. English "speaking of/as for SBJ") Bound
-erij appositive marker (only in certain sentence orders; see below) Bound
-iruj emphasis marker Bound
-oz possessive marker (mostly in nouns and pronouns; cf. English 's) Bound
-zol proprietive marker (in complements associated with a referent; see below) Free
-tuig (#) of Free
-zeg quotative/indirect speech marker Free
-zem visual evidence marker Free
-zeng sensory evidence marker Free
-zev auditory evidence marker Free
-ódiv "I wonder if..."; cf. Finnish han/hän Free
-uez/-ouz complex plural Bound
-av perfect verb[8] Free
-aij prospective marker[9] Bound
-aisom accomplished/achieved; completive aspect Free
-aubil starting to; inchoative aspect Free
-aubel no more, no longer; cessative aspect Free
-ain the result of an action or state Bound
Termison
Suffix Class Function
-e Noun Neuter
-o Masculine
-a Feminine
-i Adjective/Adjectival phrase
-u Adverb/Adverbial phrase
-s Plural
-ar Verb Infinitive
-at Indicative
-aik Conditional/Subjunctive[10]
-ait Imperative
-ant Progressive
-asant Continuative
-aid Past participle
-Ø/(a)t Interjections
Adpositions
Conjunctions

Morphological processes

Morphological processes in Relformaide
[After Epo (2014), p. 13https://inter.payap.ac.th/wp-content/uploads/linguistics_students/Yrrah_Epo_Thesis.pdf]
Process Examples
Prefixation nalintar (to be not inside) (nal- [not] + int + -ar)
nemvoardi (unofficial) (nem- [not among] + voard [true/real/authentic] + -i)
rëmmitaid (resent) (rel- [re-] + mit + -aid)
saubtime (subsection) (saub- [sub-] + tim + -e)
verjautaine (answer) (ver- [reverse] + jaut [ask] + -ain [RSLT] + -e)
péríantes (ancestors) (pé- [PST] + ríant [parent] + -e + -s)
féhódieze (upcoming event) (fe- [FUT] + hódiez + -e)
péfévanzat (...that SBJ would win) (pé- [PST] + fé- [FUT] + vanz + -at)
Suffixation neuvière (reference work) (neuv [show the way (to)/advise] + -íer [something that...] + -e)
seuddeziène (liar) (seud [lie] + dez [say] + -íen [someone who...] + -e)
nekède (kitten) (nek [cat] + -ed [youngster] + -e)
kimódra (princess) (kim [monarch] + -ódr [child] + -a)
ékuinoze (horse's) (ékuin + -oz [GEN] + -e)
Nazaulproximtúgu (outside Nassau's vicinity) (Nazaul + -proxim [nearby] + -tug [outside] + -u)
véyar (to see) (vey- + -ar)
imájinat (SBJ imagines) (imájin- + -at)
veselant (SBJ is probably going) (v- + -esel + -ant)
nougeŕámelaik (SBJ ought to swim often) (nougel- + -ámel + -aik)
fraulaija (bride) (fraul [spouse] + -aij [PRSP] + -a)
Iteration touɖdèye-touɖdèye (the year 2020) ((touv [2] + dey [10] + -e)2)
Reduplication vuivuivuipey! (hip-hip-hooray!) (vui3pey [yippee!])
glauglaung (sound made by a gong) (glau2ng)
Compounding nâyobène (doghouse) (nây [dog] + oben [house] + -e)
dínerlúdar (to gamble) (díner [money] + lud [play] + -ar)
ästrettánule (astronomy) (astrel [star] + tánul [study] + -e)
sulittoaké (audition) (sulit [test] + toak [perform] + -e)
Fusion vorite (eye) (vey [see] + rit [body organ] + -e)
touzièle (universe) (tout [whole/entire] + zíel [sky] + -e)
astregánule (astrology) (astrel [star] + tegin [sign] = astregin [star sign] + tánul [study] + -e)
doamube (bed) (doarm [sleep] + meubl [furniture] + -e)
moarzar (to kill) (moart [die] + inz [CAUS] + -ar)
ríablo (uncle) (ríant [parent] + sibl [sibling] + -o)
Zero modification All standalone adpositions/conjunctions, and most interjections; der and den are Relformaide's only invariable roots

Inflectional categories

Inflectional categories in Relformaide
[Modelled after Conklin (1949), p. 10 (quoted in Epo (2014), p. 14https://inter.payap.ac.th/wp-content/uploads/linguistics_students/Yrrah_Epo_Thesis.pdf)]
Category Types Productive in
# Label
Case 80+ See Case handbook and § Word formation: oben paradigm Nouns
Pronouns
Adjectives
Adverbs
Verbs
Number 3 Singular (-Ø)
Plural (-s)
Nouns
Pronouns
Adjectives
Verbs
Case plural (-uez/-ouz) Nouns
Pronouns
Adjectives
Adverbs
Verbs
Gender 3 Masculine (-o)
Feminine (-a)
Neuter (-e)
Nouns
Pronouns
Adjectives
Verbs
Tense 6 Present (-Ø)
Recent Past (nupé-/nuper-)
Simple Past (pé-/per-)
Discontinuous Past (plé-/pler-)
Future (fé-/fer-)
Future in the Past (péfé-/péfer-)
Nouns
Pronouns
Adjectives
Adverbs
Verbs
Aspect 4 Progressive (-ant)
Continuative (-asant)
Perfect (-av)
Prospective (-aij)
Nouns
Pronouns
Adjectives
Verbs
Mood 3 Indicative (-at)
Conditional/Subjunctive (-aik)
Imperative (-ait)
Nouns
Adjectives
Verbs
Comparative 4 Comparative (-ausm)
Superlative (-ausom)
Negative Comparative (-eusm)
Negative Superlative (-eusom)
Adjectives
Adverbs
Evidential 4 Quotative (-zeg)
Visual (-zem)
Sensory (-zeng)
Auditory (-zev)
Nouns
Pronouns
Adjectives
Adverbs
Verbs

Slots

In Relformaide, a single standalone root is the least a word can be formed with; this applies to adpositions, conjunctions, and many interjections. Most words contain one or more roots followed by a termison; negated forms, and verbs in any tense but the present, carry prefixes.

Relformaide morphology involves a 22-slot system, as laid out in the table below. Although technically possible, no common words are formed with all slots. The typical formation pattern is (1/3+)4+(5+)(7+)8+(9+)(11+)13+(15+)19(+22) for most words; 8+13+19(+22) for pronouns; and 8 for adpositions, conjunctions, and interjections.

Word formation is subject to the processes summarised in the stem change chart and elaborated in the table of intra-letter changes.

Level 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17a 17b 18 19 20 21 22
Category NEG CPLX.GND NEG.EXCL TENSE DESID SUB RE/VER ROOT TM1
(PT.PTCP)
AFFCT ART CPLX.PL CASE RCV1 TM2
(PR.PTCP)
RCV2 PERF PF.PTCP VRPN TM3 RCV3 TM4 PL
Morphemes nal- ~ naŕ- tuom-
tuam-
nem- Ø
nupé- ~ nuper-
pé- ~ per-
plé- ~ pler-
(pé)fé- ~ (pé)fer-
soul-
voul-
boul-
saub- rel-
ver-
Root -aid Various -(i)l
-un
-uez
-ouz
Case Repeat levels 2–12 -ant
-asant
Repeat levels 2–14 -av -ant
-asant
Ø -u Ø
Ø -e Ø -s
-o
-a
-i
Prn. / Dem. -ar Ø -o
-a
-at
-ant -e
-o
-a
-asant
-aik Repeat level 16
-ait Ø

Split ergativity

A feature of Relformaide is its split ergativity, whereby pronouns and determiners utilise a nominative–accusative system, but passive-voice sentences using only nouns follow an ergative–absolutive pattern.[11] Below is a comparison involving the inanimate noun livré (book); the animate feminine noun lapina (rabbit doe); and the third-person masculine pronoun lumo.

Case Word
livré lapina lumo
Ergative livriebe lapinieba lumo,
lumiebo
Nominative livré lapina lumo
Absolutive
Accusative lumio

Gender alignment

In nearly all cases, the grammatical gender of a Relformaide word matches its natural gender. Terms for water transportation, inanimate in the real world but nonetheless feminised per English tradition, are the notable exception.

Type Gender Termison(s)
Gramm. Nat.
Animate masculine creatures in kingdom Animalia M M -o
Animate feminine creatures in kingdom Animalia F F -a
Animate creatures in Animalia, whose gender is unknown/undetermined/irrelevant N C -e
Groups of animate masculine members from Animalia M M -o(s)
Groups of animate feminine members from Animalia F F -a(s)
Groups consisting of both masculine and feminine creatures from Animalia N C -e(s)
Animate beings in kingdom Plantae N C -e(s)
Inanimate material objects (except water transportation) N N -e(s)
-ọ(s)
-ạ(s)[12]
Water transportation F N -a(s)
Given names of masculine beings M M -o
Given names of feminine beings F F -a
Names of languages, fields, specialties, movements, events, and seasons N N -e
Names of places (continents; bodies of water — oceans, seas, bays, straits, gulfs; countries; subdivisions — states, provinces, districts, parishes, prefectures, counties, communes; cities and towns; villages and hamlets) N N -o
-a
-e[13]

Compounds

As in various agglutinative languages and the already established Esperanto (also a constructed language itself), new words in Relformaide can be formed out of existing base roots and affixes (as will be seen in the next segment of the grammar, § Word formation). As shown earlier on, certain combinations with difficult consonant clusters utilise an epenthetic -oṛ- interfix depending on the first vowel of the next morpheme: lech + ten = lechoṛten (with milk); plaj + tug = plájoṛtug (outside the beach); lof + kuam = lofoṛkuam (...-er than the hill).

Word formation

Relformaide is an agglutinative language capable of producing words of various lengths, and with various inflections and derivations. As such, one can make new words out of given roots and end markers.

Formulation

Rules

Base roots and affixes in Relformaide are formulated and chosen so that:

  1. they can be as machine-readable as possible, as with Lojban (another constructed language); and
  2. their spellings do not conflict with those of other roots or legible combinations.[14]

Thanks to the above factors, it is possible to form new roots that combine existing ones with nonexistent or invalid strings:

  • déoz- (awe), sourced from Greek, contains the -oz- possessive marker at the end, but the first two letters () are an invalid combo ending with a vowel. Similarly, ríant (Welsh for "parent") combines the invalid with the continuative verb mood marker, -ant.

Extra care should be taken when forming roots that begin with pé/per, plé/pler, and fé/fer (the verb tense markers), l/m/t/b/v, or z/d/f/r:

  • fer- (or féroz-) cannot represent the Latin for "wild" or "savage"; instead, Armenian-sourced vairim is used. Similarly, lokand (a Gujarati borrowing) corresponds to the Latin ferrum, or iron.
  • dúven-, a portmanteau of English/Latin/Vietnamese/Dutch elements, stands in for the problematic dait(ạ) (data/information). (Otherwise, for instance, ótr-i-dait [different data] = ótr-id-ait [different-state-IMP].)

Unless marked with breved vowels, etymons that resemble combinations of existing roots cannot be safely used in Relformaide:

  • The English word "pirate" cannot be imported as is, since the combination of pir + -at = pirat (SBJ is used for...) already exists. The compromise term is gorag-, derived from Tajik.
  • Likewise, "plantain" = "the result of being a plant" (plant + -ain), so kuand- (from Cameroon Kol) is used.

Limitations

  1. If the source word serves as an adposition and/or conjunction, then the Relformaide root should not end with a consonant followed by l or r (e.g. -fl, -dr, -pr, -tr), as such combos may prove unwieldy to pronounce. Interjection roots can also end with Cr; as a rule, -at is the mandatory marker.
  2. Double consonants are not permitted in base roots, but are acceptable in compound forms (e.g. addez — tell [ad, "to object" + dez, "say"]).
  3. Unless the underdotted is employed, Core roots cannot begin with any of the verb markers (-ar/-ant/-ait/-aid et al.) or tense prefixes (pé-/plé-/fé-/péfé-).
  4. If a root begins with b/l/m/t/v, then no verb markers should follow that first letter unless an ă is used to avoid morphological conflict.

Formation

Rules

  1. Words are normally formed as in English and the Romance languages.
    1. Nouns end in -e if the subject is inanimate or neuter; -o if masculine; or -a if feminine.
      • nây- (dog) + oben- (house) + -e (neuter marker) = nâyobène (doghouse)
      • livr- (book) + usont- (institute) + -íen (someone from/working at) + -a (feminine marker) = livrusontíena (female librarian; literally "female worker at a book institute")
      • eskũr- (squirrel) + -ed (juvenile/young creature) + -o (masculine marker) + -s (plural marker) = eskũredos (boy squirrels)
    2. Verbs end with -ar in the infinitive, -at in the indicative, and so forth.
      • vev- (live) + -ar (infinitive) = vévar (to live; to be alive). To aid in pronunciation, the e is marked with an acute (‘) as in French.
  2. Emphasis is placed on the last main root (marked below in bold) in any valid combination.
    • ad- (to) + ven- (come) + -at (indicative) = advenat (SBJ arrives)
    • toukel- (hammer) + -emek (instrumental marker) + -u (adverb marker) = toukelēmeku (with [the help of] a hammer)
  3. Roots that start with a bound suffix and end with a bound prefix are not permitted, and vice versa.
  4. Pronouns use the affixes allocated in their declension system, along with -aik (in phrases such as Sif mé taik... [If I were you...]).
  5. At least one combination consisting only of pronouns is liberally allowed: m- (I/we) and t- (you) are found with an interfix in mit (send) and tim (piece).
  6. The definite-article root l- can only be followed by the gender plus plural markers (or the emphatic marker -iruj in between), and cannot be conjugated as a verb nor combine with another root.

Summary by stem type

Affix(es) Stem type
Noun Adjective Verb Preposition
-Ø- Root Preposition
Suffixes
Termisons and TAM particles
-e Entity Something that possesses the attribute; something that is the attribute's description That which is the action; an act of doing something That which is PREP
-o/-a Animate entity Someone who possesses the attribute Someone who is PREP
-i Associated with entity Attribute Associated with the action Which is PREP
-u In the way/manner of the entity; like the entity would be or do In the way/manner of the attribute; cf. English -ly In the way/manner of the action In a PREP way/manner
-e-s Entities Things possessing the attribute; things that fit the attribute's description Acts of doing something Things that are PREP
-o-s/-a-s Animate entities Sentients who possess the attribute Sentients who are PREP
-i-s Associated with entities Attribute (of plural noun) Associated with the actions Which are PREP
-ar To be the entity To possess the attribute; to be ADJ To do the action (Verb) To be PREP
-at SBJ is the entity SBJ possesses the attribute; SBJ is ADJ SBJ does the action SBJ is PREP
-ant SBJ is being the entity SBJ is possessing the attribute SBJ is doing the action SBJ is being PREP
-asant SBJ is still being the entity SBJ is still possessing the attribute SBJ is still doing the action SBJ is still being PREP
-ant-av-at SBJ has been being the entity SBJ has been possessing the attribute SBJ has been doing the action SBJ has been being PREP
-ant-e Something that is being the entity Something that is possessing the attribute; something that is being ADJ Something that is doing the action Something that is being PREP
-ant-i That is being the entity That is possessing the attribute; that is being ADJ That is doing the action That is being PREP
-av-ar To have been the entity To have possessed the attribute; to have been ADJ To have done the action To have been PREP
-av-at SBJ has been the entity SBJ has possessed the attribute; SBJ has been ADJ SBJ has done the action SBJ has been PREP
-av-e Something that has been the entity Something that has possessed the attribute; something that has been ADJ Something that has done the action Something that has been PREP
-av-i Associated with having been the entity That has possessed the attribute; that has been ADJ That has done the action That has been PREP
-av-ant-u Having been the entity Having possessed the attribute Having done the action Having been PREP
-aij-e An entity that is about or ready to be Something that is about to possess the attribute; something that is about to be ADJ An action about or ready to be done Something that is about to be PREP
-aij-i About or ready to be the entity About to possess the attribute; about to be ADJ About or ready to do About to be PREP
-aij-ar To be ready to be the entity To be ready to possess the attribute; to be ready to be ADJ To be ready to do the action To be ready to be PREP
-aij-at SBJ is about to be the entity SBJ is about to possess the attribute; SBJ is about to be ADJ SBJ is about to do the action SBJ is about to be PREP
-aij-ant SBJ is about to be being the entity SBJ is about to be possessing the attribute SBJ is about to be doing the action SBJ is about to be being PREP
-aik SBJ ought to be the entity SBJ ought to possess the attribute; SBJ ought to be ADJ SBJ ought to do the action SBJ ought to be PREP
-aik-e Something that ought to be the entity Something that ought to possess the attribute; something that ought to be ADJ An action that ought to be done Something that ought to be PREP
-aik-i That ought to be the entity That ought to possess the attribute; that ought to be ADJ That ought to or should be done That ought to be PREP
-aid-e Something that is done
-aid-i Done
-aid-ar To be done
-aid-at SBJ is done
-aid-ant SBJ is being done
-aid-asant SBJ is still being done
-aid-av-at SBJ has been done
-aid-aij-at SBJ is about to be done
-aid-aik SBJ ought to be done
-ait Be the entity! Be ADJ! Do it! Be PREP!
-aid-ait Be done!
-aisom-ar To fulfill an action
-aisom-e An action fulfilled
-aubel-ar To cease being an entity To no longer possess the attribute To cease doing an action To no longer be PREP
-aubel-e An entity that has ceased to exist An entity no longer possessing the attribute An action that has ceased An entity that is no longer PREP
People and sizes
-oz-(e/o/a) Belonging to an entity; an entity's... Belonging to something possessing the attribute; an ADJ entity's... An action's... Belonging to something that is PREP; a PREP entity's...
-oz-íen-(e/o/a) A member of an entity (group, team, or organisation)
-íer-e Something that produces the entity Something that turns things into the attribute Something that does the action Something that makes things PREP
-íen-(e/o/a) Someone who is, does, plays, comes from, or specialises/works in the entity Someone who possesses the attribute; someone who is ADJ Someone who does the action Someone who is PREP
-ódr-(e/o/a) A child (son/daughter) or progeny of an animate entity; an offspring or spinoff of an inanimate entity A child who is ADJ
-úvel-(e/o/a) A step-relative of an animate entity; an ersatz form of an inanimate entity An ersatz, ADJ entity An ersatz action
-oajan-(e/o/a) A god-relative of an animate entity A god-relative who is ADJ
-eb-(e/o/a) Animate newborn entity A baby/newborn who possesses the attribute; a baby/newborn who is ADJ A baby/newborn who is PREP
-ed-(e/o/a) Juvenile/young entity A young entity possessing the attribute; a youngster who is ADJ A young entity that is PREP
-eld-(e/o/a) Old/elder entity An old entity possessing the attribute; an old entity that is ADJ An old entity that is PREP
-et-(e/o/a) Diminutive form of entity Something that is a little/briefly ADJ Brief action Something that is a little PREP
-ead-(e/o/a) Medium-sized entity Something that is moderately ADJ Moderate action Something that is moderately PREP
-ard-(e/o/a) Augmentative entity Something that is greatly ADJ Extended action Something that is greatly PREP
-et-i Associated with a diminutive form of the entity A little/briefly ADJ Associated with the brief action A little PREP
-ead-i Associated with a midsized form of the entity Moderately ADJ Associated with the moderate action Moderately PREP
-ard-i Associated with an augmentative form of the entity Greatly ADJ Associated with the extended action Greatly PREP
-et-ar To be a miniature form of the entity To be a little/briefly ADJ To do an action briefly To be a little PREP
-ead-ar To be a midsized form of the entity To be moderately ADJ To do an action moderately To be moderately PREP
-ard-ar To be a large version of the entity To be greatly ADJ To do a prolonged action To be greatly PREP
-eset-e A small thing that is ADJ
-esead-e A midsized thing that is ADJ
-esard-e A small thing that is ADJ
Comparatives and superlatives
-ausm-e Something more ADJ/ADJ-er
-ausom-e Something most ADJ/ADJ-est
-eusm-e Something less ADJ
-eusom-e Something least ADJ
-ausm-i More ADJ/ADJ-er
-ausom-i Most ADJ/ADJ-est
-eusm-i Less ADJ
-eusom-i Least ADJ
-ausm-u More ADJ-ly
-ausom-u Most ADJ-ly
-eusm-u Less ADJ-ly
-eusom-u Least ADJ-ly
Emphasis
-iruj-e An entity as emphasis An ADJ entity as emphasis An action as emphasis A PREP entity as emphasis
-iruj-i Pertaining to the entity as emphasis ADJ as emphasis Associated with the action as emphasis Which is PREP as emphasis
-iruj-ar To be the entity as emphasis To be ADJ as emphasis To do the action as emphasis To be PREP as emphasis
-uit-e A stronger variant of the entity An entity that is strongly ADJ A serious action That which is really PREP
-uit-i Associated with a stronger variant of the entity Strongly, seriously, or really ADJ Associated with serious actions Really PREP
-uit-ar To be a stronger variant of the entity To be strongly ADJ To seriously do the action To be really PREP
Verbal/duration particles
-ábil-e Something that can be the entity Something able to possess the attribute; something that can be ADJ A potential action A potential state of being PREP
-ábil-i Able to be the entity Able to possess the attribute; able to be ADJ Able to do the entity Able to be PREP
-ábil-ar To be able to be the entity To be able to possess the attribute; to be able to be ADJ To be able to do the action To be able to be PREP
-ábil-id-e The ability of being the entity The ability of possessing the attribute; the ability of being ADJ The ability of doing the action The ability to be PREP
-aibil-e Something that can be done
-aibil-i Able to be done
-aibil-ar To have the potential to be done
-aibil-id-e The ability of an action being done
-ain-e The result of being the entity The result of being ADJ The result of the action The result of being PREP
-ámel-e An occasional or frequent entity Something often or frequently ADJ A frequent action Something often or frequently PREP
-ámel-i Often or frequently the entity Often or frequently ADJ Often or frequently the action Often or frequently PREP
-ámel-ar To be the entity often or frequently To possess the attribute often or frequently; to be ADJ often or frequently To do the action often or frequently To often or frequently be PREP
-oguil-e A part-time entity Something occasionally ADJ A part-time action Something occasionally PREP
-oguil-i Sometimes the entity Sometimes ADJ Sometimes the action Sometimes PREP
-oguil-ar To be the entity sometimes To be sometimes ADJ To do the action sometimes To be sometimes PREP
-oajem-e A momentary entity; something that appears for a split second Something momentarily ADJ A momentary action Something momentarily PREP
-oajem-i Momentarily the entity Momentarily ADJ Momentarily the action Momentarily PREP
-oajem-ar To be the entity for a split second To be ADJ for a moment To do the action in a flash To be PREP for a moment
-esel-e A probable entity Something probably or perhaps ADJ A probable action Something probably or perhaps PREP
-esel-i Maybe, perhaps, or probably the entity Maybe or perhaps ADJ Maybe or perhaps the action Maybe or perhaps PREP
-esel-ar To probably be the entity To probably possess the attribute; to probably be ADJ To probably do the action To probably be PREP
-ódiv-at What about the entity? What if SBJ were ADJ? What if SBJ should do the action? What if SBJ were PREP?
-aubog-e Something that favours or supports an entity Something that favours the attribute Something that favours or supports the action
-aubog-i In favour or support of an entity; pro-entity In favour of the attribute In favour or support of the action; pro-action
-aubog-ar To favour or support an entity; to be pro-entity To favour the attribute To favour or support the action; to be pro-action
-opon-e Something that opposes an entity Something that opposes the attribute Something that opposes the action
-opon-i Against an entity; anti-entity Against the attribute Against the action; anti-action
-opon-ar To oppose an entity; to be anti-entity To oppose the attribute To oppose the action; to be anti-action
Attributionals
-un-et-e Part or some of the entity Part or some of which possesses the attribute Part or some of the action
-euz-e Something that is full of the entity Something that is full of the attribute Something that is full of the action
-euz-i Full of the entity Full of the attribute Full of the action
-euz-ar To be full of the entity To be full of the attribute To be full of the action
-ail-i Pertaining/relating to the entity Pertaining/relating to the attribute Pertaining/relating to the action Pertaining/relating to PREP
-if-i Characterised by the entity Characterised by the attribute Characterised by the action
-id-e The state, quality, or condition of being the entity; cf. English -(i)ty/-ness/-hood/-dom The state, quality, or condition associated with the attribute The state, quality, or condition associated with the action The state, quality, or condition of being PREP
-oid-i Like/in the form of the entity Like/in the form of the attribute Like/in the form of the action
Causatives
-inz-ar To cause or bring forth the entity (on something) To make ADJ To cause the action To make PREP
-inz-aid-i Transformed with instances of the entity Made ADJ Made done (by an agent) Made PREP
-inz-aid-ar To be transformed with instances of the entity To be made ADJ To be made done (by an agent) To be made PREP
-iz-ar To turn into the entity To turn ADJ To turn into the action To turn into a PREP entity
-iz-aid-i Turned into the entity Turned ADJ Turned into the action Turned into a PREP entity
-iz-aid-ar To be turned into the entity To be turned ADJ To be turned into the action To be turned into a PREP entity
-ẽrgim-u Caused by/because of the entity Caused by/because of the attribute Caused by/because of the action Caused by/because of being PREP
Specialties
-íom-e An element named after the entity An element possessing the attribute An element named after the action An element named after PREP
-tánul-e The study of the entity; the field associated with the entity The study of the attribute; the field associated with the attribute The study of the action; the field associated with the action The field associated with PREP
-eand-e A tool devoted to the entity A tool that is ADJ A tool for doing the action A tool that is PREP
-iz-eand-e A tool that tranforms things into the entity A tool for making things ADJ A tool for making things PREP
-ineal-e A machine devoted to the entity A machine that is ADJ A machine for doing the action A machine that is PREP
-iz-ineal-e A machine that transforms things into the entity A machine for making things ADJ A machine for making things PREP
Prefixes
Functionals
nal- Not the entity Not the attribute SBJ does not do the action SBJ is not PREP
nid- Not the entity in question Not possessing the attribute SBJ does not do the action in question SBJ is not PREP in question
rel- Something that is the entity again Possessing the attribute once again SBJ redoes/repeats the action SBJ is PREP again
ver- The entity in reverse The inverse of the attribute SBJ does the action in reverse, or reverses/reverts it SBJ is PREP, but in reverse
voul- An entity that is wanted or desired An attribute that is wanted or desired An action that is wanted or desired SBJ wants to be PREP
boul- An entity that is needed or important An attribute that is needed or important An action that is needed or important SBJ needs to be PREP
seb- A self-serving entity Self-ADJ An action done by oneself Self-PREP
shoar- Entities to each other ADJ towards one another An action done by each other/one another PREP in relation to each other
tep- Great-/grand-
tout- A full or complete entity Completely or totally ADJ; omni-ADJ Full action Completely PREP
Affects
jout- A sudden or unexpected entity Suddenly or unexpectedly ADJ A sudden or unexpected action Suddenly or unexpectedly PREP
vasit- Intentional entity Intentionally ADJ Intentional action Intentionally PREP
tasiv- Accidental entity ADJ by accident Accidental/unintended action PREP by accident
International roots
hîdrol- Hydro-entity Hydro-ADJ Hydro-action
hîpẽr- Hyper-entity Hyper-ADJ Hyper-action
mîkrol- Micro-entity Micro-ADJ Micro-action
saub- A sub-entity Sub-ADJ Sub-action Sub-PREP
supẽr- Super-entity Super-ADJ Super-action
télev- Something transmitted from afar; tele-entity Tele-ADJ Something done from afar; tele-action
Tense augments
pé-
(per-)
A past or ex-entity Once or formerly the attribute SBJ did the action SBJ was PREP
plé-
(pler-)
An entity that once was, but is no longer Once or formerly, but no longer, the attribute SBJ used to do the action SBJ used to be PREP
fé-
(fer-)
A future entity; an entity-to-be Planned or anticipated to be the attribute SBJ will do the action SBJ will be PREP
pé-fé-
(pé-fer-)
An entity that would be Which would be the attribute ...that SBJ would do the action ...that SBJ would be PREP

Examples

Simple

ótr

Relformaide words change meanings based on the termison and/or prefix, as shown in this paradigm for ótr- ("other/different"). As with most adjective-based roots, ótr- employs all five uniliteral termisons—plus the verbal ones—in its simplest form. (It becomes ńótr- when placed after a vowel.)

With termisons
Grammatical form Termison Word Meaning
Noun Neuter -e ótré another one
Masculine -o ótro another one (masculine)
Feminine -a ótra another one (feminine)
Adjective -i ótri different
Adverb -u ótru differently
Verb Infinitive -ar ótrar to be different; to differ
Indicative -at ótrat SBJ is different; SBJ differs
Perfect -av-at ótravat SBJ has been different
Prospective -aij-at ótraijat SBJ is about to be different
Conditional / Subjunctive -aik ótraik SBJ ought to differ
Imperative -ait ótrait Be different! Stand out!
Progressive -ant ótrant SBJ is differing
Continuative -asant ótrasant SBJ is still different
Past Participle -aid ótraid[15] been different
With prefixes
Grammatical form Prefix Word Meaning
Negation Verbal nal- nalótrat SBJ is not different; SBJ is the same
Nominal nem- nemótrat SBJ is not among those who are different; SBJ is homogeneous
Iterative rel- relótrat SBJ is different again
Volitive soul- soulótrat SBJ wishes to be different
Desiderative voul- voulótrat SBJ wants to be different
Debitive (Necessitative) boul- boulótrat SBJ needs to be different
Tense Present Ø- ótrat SBJ is different; SBJ differs
Recent Past nupé- ~ nuper- nuperótrat SBJ was just different
Simple Past pé- ~ per- perótrat SBJ was different
Discontinuous Past plé- ~ pler- plerótrat SBJ used to be different
Future fé- ~ fer- ferótrat SBJ will be different
Future in the Past pé-fé ~ pé-fer- péferótrat SBJ would be different
ed

This next paradigm demonstrates the similar versatility of the animate root, ed- (young). (It becomes ńed- when placed after a vowel.)

Word Affix(es) Meaning
Word/Phrase Affix(es)
Basic words
ède[16] -e youngster, youth, youngling, child, kid, tyke neuter noun
edo -o boy masculine noun
eda -a girl feminine noun
edi -i young adjective
edu -u youngly (Rare) adverb
edar -ar to be young infinitive verb
Compound words
édausmé -ausm-e the younger comparative + neuter
édausmi -ausm-i younger comparative + adjective
édausome -ausom-e the youngest superlative + neuter
édausomi -ausom-i youngest superlative + adjective
edómisti -ómist-i children's dedicated to + adjective
édeuzi -euz-i youthful full of + adjective
édeuzu -euz-u youthfully full of + adjective
édide[16] -id-e youth state/condition + neuter
ednandu -nand-u as a child, while young as... + adverb
reledi rel-R-i young again again + young + adjective
relédinzar rel-R-inz-ar to rejuvenate again + young + cause + infinitive verb
relédinzière rel-R-inz-íer-e the Fountain of Youth again + young + cause + agent + neuter
ódrède ódr-R-e young offspring/progeny child + young + neuter
tuerède tuer-R-e cub, whelp, young creature creature + young + neuter
úvelède úvel-R-e young step-relative step- + young + neuter
oajanède oajan-R-e young god-relative god- + young + neuter
Ámeríkíenèdes Ámerik-íen-R-e-s "Kids in America" America + demonym + young + neuter + plural
vey/veiz/vorit

Words in Relformaide can be formed from existing morphemes based on attribute, thus reducing the need for different roots when necessary. These charts, dealing with the inanimate vey- ("see") and related veiz- (show)/vorit- (eye), demonstrate the language's derivational versatility.

Word Affix(es) Meaning
Word Affix(es)
Verbs (with infinitive -ar)
véyar see, watch, view
véyetar -et glance a little
véyardar -ard stare a lot
véyaidar -aid be seen; appear past participle
véyondar -ond be blind without
naŕvéyar nal- not see/notice (something) verb negator
prevvéyar prev- foresee, preview before
évïvvéyar évit- ignore, disregard avoid
advéyar ad- look at, admire at something/someone
äuvvéyar aup- face (something/someone), look toward(s) toward(s)
sebvéyar seb- look at oneself self
shoarvéyar shoar- see each other, stare face-to-face each other
intäuvvéyar int-aup- peer (at) in + towards (into)
tränvvéyar tranz- scan through
wegvéyar weg- shun off/away from
Nouns (with neuter -e)
vèye sight (act of seeing)
véyaide -aid sight (something seen) past participle
veyábilidé -ábil-id vision able to... + state of
véyaine -aine insight the result of...
véyiène -íen watcher, looker, seer someone who...
véyondide -ond-id blindness without + state of
veyúbèle úbel- bad vision, poor eyesight bad
prevvèye prev- foresight, preview before
févèye fé- outlook future tense marker
véyaurode -aurod lookout, vantage point, observatory place
veyäurffásine -aurd-fásin videotape, videocasette, VHS for purpose + tape
véyeande -eand glasses tool
véyeandiène -eand-íen optician tool + occupation
sebvéyaide seb-R-aid reflection self + see + past participle
sebvéyeande seb-R-eand mirror self + see + tool
Adjective/Adverb (with -i/-u)
véyi -i visual
véyondi -ond-i blind without
véyu -u visually
véyondu -ond-u blindly without
Word Affix(es) Meaning
Word Affix(es)
veizar -ar show, display, demonstrate infinitive verb
veize -e show, demonstration neuter noun
Word Affix(es) Meaning
Word Affix(es)
Nouns (with neuter -e)
vorite eye
voritiène -íen oculist, eye doctor occupation...
voritosivaine -osiv-ain glaucoma, eye disease sick + resulting from
tuarenvorite tuaren- cataract fog
lounvorite loun- moon blindness, night blindness moon
drautvorite draut- dry eye dry
vörisshoule -shoul sclera white
lẽrn

This chart for the inanimate verbal root lẽrn- (learn) highlights the affixational differences between Esperanto, a well-known constructed language, and Relformaide itself.

Affix(es) Word Meaning
EO RFM EO RFM
lern' lẽrn- Root
-i -ar lerni lẽrnar to learn
-ad -ant lernadi lẽrnantar " study
-eg -ard lernegi lẽrnardar " cram
-ig -inz lernigi lẽrninzar " cause to learn
-iĝ -iz lerniĝi lẽrnizar " learn intuitively
-et lerneti lẽrnetar " dabble in learning
dis- -úbel dislerni lẽrnúbelar " learn in a desultory manner
ek- -prem eklerni lẽrnpremar " begin to learn
el- shoum- ellerni shoumlẽrnar " learn thoroughly
mal- ver- mallerni verlẽrnar " unlearn
re- rel- relerni rellẽrnar " learn again
-ant -ant-(e/o) lernanto lẽrnante, lẽrnanto a pupil, a learner (neut./masc.)
-ant-in -ant-a lernantino lẽrnanta a pupil, a learner (fem.)
-ej-an -íen-o lernejano lẽrníeno a schoolboy
-ej-an-in -íen-a lernejanino lẽrníena a schoolgirl
ge-R-ant-o-j -íen-e-s gelernantoj lẽrniènes pupils (masc. and fem.)
-ant-ar -ant-truz lernantaro lẽrnanttruze a class
-ej -abod lernejo lẽrnaurode a school (EO); a learning institution (RFM)
-ej-et -abod-et lernejeto lẽrnaurodète an elementary school (EO); a small learning institution (RFM)
-ej-ar -abod-truz lernejaro lẽrnäbottruze a university (EO); a cluster of learning institutions (RFM)
-ul -av-(e/o) lernulo lẽrnave, lẽrnavo a learned person, a learned man, a savant
-ul-in -av-a lernulino lẽrnava a learned woman, a "blue stocking"
-aĵ lernaĵo knowledge (EO)[17]
-il -eand lernilo lẽrneande intelligence (EO); flashcard (RFM)
-ind -val lerninda lẽrnvali worth learning
-o -e lerno lẽrne act or action of learning
-ebl -aibil lernebla lẽrnaibili learnable
-ec -aid-id lerneco lẽrnaidide learnedness
-em -anfel lernema lẽrnanfeli studious
-er aurd-esim lernero lẽrnaurdesime a subject of a curriculum
-ar aurd-truz lernaro lẽrnäurttruze a curriculum
-a -i lerna lẽrni learned (EO); associated with acts of learning (RFM)
-e -u lerne lẽrnu learnedly (EO); in the manner of an act of learning (RFM)
-u -ait lernu! lẽrnait! learn!
oben

This declension chart for the inanimate root oben- ("house"; ńoben- post-vowel) is another example of Relformaide's agglutinative nature. Relformaide has an array of affixes and affixoids corresponding to more than 80 grammatical cases found in various natural and constructed languages (whether single or compound), as well as a handful conceived for Relformaide itself.

Whereas cases in various languages are placed after their host root, those in Relformaide act as postpositional mesoclitics between the root and termison. The intrafixes -ieb-, -iz-, -orz-, -oz-, and -únet-, plus the -u suffix (and the -i- in accusative pronouns), are the language's only bound case markers; -eun- and -e(s) mark topics and singularity/plurality respectively; and -ouz- (otherwise represented as -uez-) represents a plural subject before case particles.

Case Abbr. Suffix Derivation Meaning
Singular Plural
Grammatical
Nominative
Accusative
Absolutive
Subject NOM
ACC
ABS
-e obène obènes (the) house(s)
Topic -eune obenēune obenēunes as for/speaking of the house(s)
Ergative ERG -iebe obeniebe obeniebes done by the house(s) (passive agent)
Appositive[18] APP -erije obenerije obenerijes ...which is the house
Dative DAT -ade obenade obenades (to) the house(s)
Partitive PART -únète obenúnète obenúnètes part of the house(s)
Vocative Familiar VOC -té obenté obentés O house(s)
Formal -usté obenusté obenustés O house(s)
Adverbial/Functive ADV/FUN -u obenu in the manner/way of a house; like a house
Relational
Genitive
(Inalienable)
GEN -oze obenoze obenozes (the) house's/houses'
Possessive
(Alienable)
POSS -orze obenorze obenorzes associated with the house(s)
Proprietive Inalienable PROP -zol obenzol obenouzzol ...with the house(s) (as part of someone's belongings);
owning the house(s)
Alienable -ten obenten obenouśten ...with the house(s) temporarily, having houses
Ornative ORN -abuam obenabuam obenouzabuam endowed/equipped with a house/houses,
provided a house/houses (by a donor)
Associative
Instrumental/
Vialis/Instructive
INSTR
VIA
-emek obenemek obenouzēmek with (the help of) the house(s), using the house(s) (for a task or purpose); by way of/via/through the house(s); house-wise
Utilitative UTL -útil obenútil obenouzútil using the house (but not for any task or purpose)
Utilitive/
Applicative/Purposive (Supine)
UTIL -aurd obenaurd obenouzaurd for use in the house(s); (intended) for home use
Benefactive BEN -ómist obenómist obenouzómist for (the benefit of) the house(s), for the house's/houses' sake,
(dedicated) to the house(s)
Causal/Motivative[19] CSL -ẽrgim
-póvaud
obenẽrgim
obenpóvaud
obenouzẽrgim
obenouśpóvaud
because of/thanks to the house(s)
Comitative/
Sociative/Associative
COM -aseb obenaseb obenöutāseb (along) with the house(s)
Conjunctive CNJ -auvek obenauvek obenouzāuvek (along) with the house(s)
(as part of the entity/group)
Predicative PRD -tegem obentegem obenouśtegem depending/relying on a house/houses
Spatial
Originative OGN -esil obenesil obenöutesil (originating, coming) from the house(s) (of)
Egressive EGR -äpprem obenäpprem obenouzäpprem starting from the house(s) (in movement)
Initiative INIT -premad obenpremad obenouśpremad starting from the house(s) (as the source)
Terminative TERM -ósot
-äffim
obenósot
obenäffim
obenöutósot
obenouzäffim
up to the point of the house(s),
ending up at the house(s)
Locative LOC -ad obenad obenouzad at the house(s) (of)[20]
Allative/Versative/
Approximative
(Directional)
ALL -aup obenaup obenouzaup towards the house(s) (in position)
Lative LAT -äuvv obenäuvv obenouzäuvv to(wards) the house(s) (in movement)
Orientative ORI -élogaup obenélogaup obenouzélogaup facing the house(s)
Oppositive/Situative OPS
SIT
-esreg obenesreg house to house; houses facing/compared with each other
Cisative CIS -túdelad obentudelad obenouśtúdelad at the side of the house(s)[21]
Apudessive APUD -proxim obenproxim obenouśproxim near/next to the house(s), by the house(s)
Apudlative APLT -proximad obenproximad obenouśproximad at the vicinity of the house(s)
Apudallative APLL -proximaup obenproximaup obenouśproximaup towards the vicinity of the house(s)
Pertingent PERT -toag obentoag obenouśtoag touching/against the surface of the house(s)
Adessive ADE -íak obeníak obenouzíak on the surface of the house(s)
Altessive ALT -ault obenault obenouzault above the house(s)
Superessive SUPE -auload obenauload obenouzauload atop the surface of the house(s); on the roof(s)
Superlative[22] SUPL -äulvv obenäulvv obenouzäulvv over the top of the house(s); over the roof(s)
Perlative PERL -tranz obentranz obenouśtranz through the house(s) (in movement)
Prolative PROL -korttranz obenkorttranz obenouśkorttranz along the house(s)
Inessive I INE -int obenint obenouzint in the house(s)
II -intad obenintad obenouzintad within the house(s), inside the house(s)
Illative ILL -intaup obenintaup obenouzintaup into (in towards) the house(s)
Interessive/Intrative Dual INTER -prolad obenprolad between the (two) houses
Plural -multintad obenmultintad among the houses
Intertransitive Dual INTRT -proltranz obenproltranz in between the two houses
Plural -multtranz obenmulttranz going amidst the houses
Contessive CONT -multint obenmultint among the houses, amidst the houses
Exessive EXE -tug obentug obenouśtug outside the house(s) (in position)
Ablative ABL -túgaup obentúgaup obenouśtúgaup towards the limits of the house(s); heading outside the house(s) (in movement)
Elative ELA -tüvv obentüvv obenouśtüvv off the house(s), out of the house(s) (in movement)
Delative DEL -weg obenweg obenouzweg off (the surface of) the house(s)
Subessive SUBE -úlem obenúlem obenouzúlem below the surface of the house(s),
underneath the house(s)
Sublative SUBL -úlemad obenúlemad obenouzúlemad at the bottom of the house(s); in the basement(s)
Chronological
Immediate IMD -oantem obenoantem obenouzoantem first the house(s)
Precursive PCV -prev obenprev obenouśprev before the house(s)
Postcursive POS -áprev obenáprev obenouzáprev after the house(s)
Quantitative
Abessive ABE -ond obenond obenouzond without the house(s); homeless
Privative PRIV -eriv obeneriv obenouzeriv removing/taking away the house(s)
Exceptive EXC -moin obenmoin obenouśmoin except the house(s)
Selective SEL -oantuig
-plũrimtuig
obenōantuig obenplũrimtuig[23] one of the houses, several of the houses
Inclusive INCL -nend
-ausin
obennend
obenausin
obenouśnend
obenöutausin
including the house(s), the house(s) too/as well
Distributive DIST -aiv obenaiv obenouzaiv per house; per group/set/cluster of houses
Exclusive EXCL -seul obenseul obenouśseul only the house(s)
Qualitative
Propositive PRO -aubog obenaubog obenouzaubog for the house(s) (in opinion); in support of/supporting the house(s); pro-house
Oppositional OPP/ANTI -opon obenopon obenouzopon against the house(s) (in opinion); anti-house
Aversive/Evitative AVE/EVIT -évit obenévit obenouzévit avoiding the house(s)
Contrastive CRS -pótal obenpótal obenouśpótal instead of the house(s), rather than the house(s)
Transpositive TSP -uslen obenuslen obenöutuslen on behalf of the house(s)
Concessive CNC -mólen obenmólen obenouśmólen despite/in spite of the house(s), the house(s) in spite, although a house/houses
Postulative PTL -sif obensif obenouśsif if the house(s), then...
Conversive CNV -nemsif obennemsif obenouśnemsif if not for the house(s), then...
Essive ESS -nand obennand obenouśnand as a house/houses
Translative TRSL -iz obeniz obenöutiz changing into a house/houses, becoming a house/houses
Compositive CPS -esit obenesit obenöutesit made (out) of the house/houses
Semblative/Formal SEMBL
FORM
-oid obenoid obenouzoid like/resembling the house(s) (in appearance), in the form of the house(s)
Similative SIMIL -ausik obenausik obenouzausik like the house(s) (in manner or habit)
Identical IDT -aibem obenaibem obenouzaibem the same as the house(s)
Comparative[24]/Equative CMPR
EQU
-kuam obenkuam obenouśkuam (...er) than the house(s); (as ...) as the house(s)
Referential REF -prin obenprin obenouśprin about the house(s); concerning the house(s)
Considerative CSD -omil obenomil obenouzomil according to the house(s)

Relformaide also utilises two cases exclusive to time-based expressions and words for events:

  • Temporal (TEMP, -wob): kinwob (at five o'clock), sẽrtnevwob (at 9:00 p.m.), Kainmasinwob (in July), Krismasinwob (during the Christmas season), toakardwob (during the concert).
  • Limitative (LMT, -maurad): kinmaurad (by five o'clock), sẽrtnevmaurad (by 9:00 p.m.), tośtueldíemmaurad (by your birthday), Paskelmaurad (by Easter), anovaubrimaurad (by the New Year).

In addition, a variation of the Essive—the Linguistic Essive (ESS.LG, -nong)—denotes material written in a stated language:

Mandạrinnong (in Mandarin), Japaunnong (in Japanese), Suahilinnong (in Swahili), Euskạrnong (in Basque), Ingilnong (in English).
nek

While various natural languages give different terms to various animals (depending on age and gender), Relformaide uses one common root and various affixes to accomplish the same effect; this also applies with various species-associated compounds. Below is an example involving the Japanese-derived animate root, nek- (cat).

Relformaide word English meaning
Masculine Feminine Neuter/Pan-gender Masculine Feminine Neuter/Pan-gender
Names
neko neka nèke tom queen cat
neketo neketa nekète small(-sized) cat
nekedo nekeda nekède kitten, young cat
nekebo nekeba nekèbe newborn/baby kitten
nekoido nekoida nekoide feline
nekoidedo nekoideda nekoidède feline cub, young feline
nekoidebo nekoideba nekoidèbe newborn/baby feline
vairimneko vairimneka vairimnèke wild cat, feral cat
aupönsneko aupönsneka aupönsnèke stray cat
Postpositionals
tüonnekadu tüannekadu nekadu to the cat, at the cat
tüonnekedadu tüannekedadi nekedadu to the kitten, at the kitten
tüonnekaupu tüannekaupu nekaupu towards the cat
tüonnekproximu tüannekproximu nekproximu near the cat, next to the cat
tüonnekaultu tüannekaultu nekaultu above the cat
tüonnekúlemu tüannekúlemu nekúlemu below the cat's feet/paws
Associated terms
nekmankaurde cat food
nektaulite litterbox
esper

This chart serves to address and amend the derivational defects in Esperanto; esper-'s definition remains as "hope".

Word Meaning
espère hope
esperi associated with hope
esperu in the manner/way of hope
esperiène one who hopes, a hopeful one; hoper
esperar to hope
esperat SBJ hope(s)
esperait! hope!
esperant[25] SBJ is hoping/continues to hope
esperanté[25] that which is hoping/continues to hope
esperantiène[25] someone who is hoping/continues to hope
esperaijat about to hope, going to hope
esperaijiène someone who is about to hope
esperaid hoped (past participle stem)
esperavant having hoped
esperaide that which is hoped
espereuzi full of hope; hopeful
espereuzu hopefully
esperonde a desperate one
esperondi hopeless, desperate, despondent
esperondu hopelessly, desperately
esperondide despair (n), hopelessness, despondency
esperondar despair (v)
esperanfeli hope-prone
esperaine the result of hoping
esperide state of hope
esperesilabode source of hope
aurdesperar to hope for
aurdesperat SBJ hope(s) for
aurdesperant SBJ is hoping for/continues to hope for
aurdesperanté that which is hoping for/continues to hope for
aurdesperantiène someone who is hoping for/continues to hope for
aurdesperaijat about to hope for, going to hope for
aurdesperaijiène someone who is about to hope for (something)
aurdesperaid hoped for
aurdesperavant having hoped for
aurdesperaide that which is hoped for
tropesperar to overhope
esperṣerivar to dash one's hopes
Esperlinge Esperanto (literally "language of hope")
Esperlingparliène someone who speaks Esperanto
Esperlingdoseliène teacher of Esperanto
Esperlinglẽrniène student of Esperanto
Esperlingiziène Esperanto translator
Esperlingsábante Esperanto expert/professional
Esperlingloaké Esperanto devotee/fan
Espernongi in Esperanto
Joalespère Good Hope
east

This next table compares forms in Kayardild , an extinct Aboriginal Australian language, with their Relformaide equivalents. (Based on Evans (2010), p. 164https://books.google.com/books?id=A7dNZLHvW-cC.)

Root: ri- (east)
Kayardild Relformaide English
riya easte east
rilungka eastaupu to the east, eastward
riyananganda eastaupu to the east of
rilumbanda eastesile easterner
riinda easttüvvant moving from the east
riliida eastäuvvasant heading ever eastward
riliji houteastaupu far to the east
rinyinda eastoṛkúzobadi at the eastern extremity of
ringurrnga proltrannyeasti east across a geographical discontinuity
riinkirida parlaupeasttüvvänkkúzobadu at the boundary you meet moving from the east toward the point of speech
rimali eastätté hey you in the east!
riinmali easttüvvantté hey you coming from the east!
rilumali eastäuvvantté hey you going eastward!
rilumirdamirda eastoṛdúgaungtenadu in the dugong grounds to the east
rilunganda eastesiŕsúflé easterly wind
rilurayaanda vüëinnochemöreastägguersile previous night’s camp in the east
rilijatha (seb)eastauproular turn (self) round to the east
rilijulutha eastäummouvvar move something to the east
eastäuttuedtendoarmar sleep with one’s head to the east
rimarutha eastäuvvéyar look to the east
riinmarutha eastesilvéyar look from the east
fraul

Below is a comparison between Relformaide and Novial, an alternative to Esperanto. Novial was devised by Danish linguist Otto Jespersen in the late 1920s; this sample is taken from his 1928 work, An International Language, with variations of mari/fraul-/marry (and in the English version, related terms) marked in bold.[26]

Novial Relformaide English
Li pastoro ha mari Paul e Anna. Paul ha mari se a Anna. Les blid mari yer; dunke les es nun marit. Paul es Annan marito, e la es li marita de Paul. Anna esed charmanti kom marienda. After li mario li du marites departad a Paris por li mari-voyaje. Nus espera ke li mariteso sal es felisi. Lo pastoro fraulinzavat Paulo nend Anya. Paulo fraulizavat auvek Anya. Lumés pébat fraulizaid vüëiddíemwobu; ńẽrgim lumés noutadu fraulizaidat. Paulo bat Anyoza fraulo, nend luma bat Paulozo fraula. Anya pébat chạrmanti nand fraulaija. Áprev le fraulfeste, le fraultousène peräuvvat Párieze wob selbtouvoze fraulvé. Més esperat le fraulide fébat froli. The clergyman has married Paul and Anna. Paul has married Anna. They were married yesterday; thus they are now married. Paul is Anna's husband, and she is Paul's wife. Anna was charming as bride. After the wedding the married couple left for Paris on their wedding-trip. We hope the marriage (state of being married) will be happy.

Complex

In Relformaide, a phrase like "Although you really tried to make this play unsuccessful" can be rendered as the more analytic Mólen té pépógirújat haulemízante der esine dúlane, or the more morphologically complex Tośpédúlanēsinhaulemiśpógirujmólenu. The first form is used in general speech and writing, while the latter is mainly confined to print demonstrations of the language's agglutinative tendencies.

Morpheme breakdown
Morphemes Form Meaning Notes
haulem Haulem- Unsuccessful, a failure Hawaiian
haulem iz Haulemiz- To turn into a failure Translative case suffix
dúlan haulem iz Dúlanhaulemiz- To render a play (for stage or screen) unsuccessful Tagalog dula(an)
dúlan esin haulem iz Dúlanēsinhaulemiz- To render this play unsuccessful Demonstrative pronoun; Estonian/Finnish
dúlan esin haulem iz pog Dúlanēsinhaulemiśpog- To try rendering this play unsuccessful Dutch pogen
dúlan esin haulem iz pog iruj Dúlanēsinhaulemiśpógiruj- To strongly try rendering this play unsuccessful Emphatic marker; Marshallese
dúlan esin haulem iz pog iruj dúlanēsinhaulemiśpógiruj- SBJ strongly tried rendering this play unsuccessful Past tense augment
t oz dúlan esin haulem iz pog iruj Tośpédúlanēsinhaulemiśpógiruj- Your past serious attempt to make this play unsuccessful 2nd person familiar + genitive case suffix
t oz dúlan esin haulem iz pog iruj mólen Tośpédúlanēsinhaulemiśpógirujmólen- Although you really tried to make this play unsuccessful Concessive case suffix; Greek
t oz dúlan esin haulem iz pog iruj mólen u Tośpédúlanēsinhaulemiśpógirujmólenu Although you really tried to make this play unsuccessful Adverb termison
Note Note:

Word classes

All words in Relformaide are categorised into nine classes: articles, pronouns, nouns, verbs, adjectives, adverbs, adpositions, conjunctions, and interjections.

Articles

Both of Relformaide's articles, l (the) and un (a/an/some), are always attached to their host roots à la Romanian.

Definite

English has one definite article, the, which translates into -lo (masculine), -la (feminine) and -le (neuter) in Relformaide:

ríantlo (the father), ríantla (the mother), ríant (the parent).

When a plural noun is used, -s is likewise added to the end:

ríantlos (the fathers), ríantlas (the mothers), ríantlés (the parents).

-Lo/-la/-le exhibits elision similar to French when it becomes l' before a vowel:

l'aumbro (the man), l'aumbra (the woman), l'aumbré (the person).

It switches back to -los/-las/-les in plural form, however:

aumbrilos (the men), aumbrilas (the women), aumbriles (the persons/people).

Countries with traditional plural forms remain that way when converted into Relformaide:

Verpuem-Územtimiles (United States), Ulemúzemiles (Netherlands),[27] Mauldivlés (Maldives), Filipiniles (Philippines), Bahamlas (Bahamas).

Indefinite

English's indefinite article, a(n), corresponds to -uno (masculine), -una (feminine), and -une (neuter) in Relformaide:

nekuno (a tomcat), nekuna (a queen cat), nekune (a cat);
mosheluno (a male meerkat), mosheluna (a female meerkat), moshelune (a meerkat).

Before vowels, it turns into un':

un'aumbra (a woman), un'audiène (a listener), un'edo (a boy), un'élemente (an element).

Pluralised, it becomes the word for some:

edunos (some boys), aumbrunas (some women), aumbredunes (some children), kimunes (some monarchs); gómaidunes (some secrets).

When a case mesoclitic is involved, the host root is assumed to be definite unless un- is placed between both:

obeninti (in the house); eskoltúgu (outside school); nekúlemu (below the cat's paws); móbilunabuamu (furnished with a car).

Relformaide also makes use of the partitive case in forms such as bauterúnète (some butter), fáloavúnète (some bread), and lechúnète (some milk).

Pronouns

Personal

Relformaide's pronoun system is modelled after those of Spanish and Quechua, and honours the T–V distinction found in various Romance languages.[28] All pronouns decline for case, number, and gender, as do nouns. This table covers the basic neuter forms; for a complete rundown, see Project:Pronoun chart.

Person Type
Subject Object Indirect Possessive Reflexive Reflexive Emphatic
Determiner / Pronoun Adjective Verb Emphatic
1st Singular mié madé mozé mozi mozar sebmozé sebmé séblimé
Plural més miés madés mozés möutozi möutozar sebmozés sebmés séblimés
2nd (Familiar) Singular tié tadé tozé tozi tozar sebtozé sebté séblité
Plural tés tiés tadés tozés tüetozi tüetozar sebtozés sebtés séblités
2nd (Formal) Singular usté ustié ustadé ustozé ustozi ustozar sébustozé sébusté séblusté
Plural ustés ustiés ustadés ustozés ustüetozi ustüetozar sébustozés sébustes séblustés
3rd Singular lumé lumié lumadé lumozé lumozi lumozar séblumozé séblumé séblilumé
Plural lumés lumiés lumadés lumozés lumöutozi lumöutozar séblumozés séblumés séblilumés
4th Subject selbé selbie selbade selboze selbozi selbozar sësselboze sësselbe sébliselbe
Object selbème selbemie selbemade selbemoze selbemozi selbemozar sësselbemoze sësselbème sebliselbème

Referential

Relformaide employs a special pronoun, selbé/selbo/selba, to denote the subject last referred to. Derived from German, it acts the same way as its Lojban influence, ri.

Jaurad lumo vat, selbo prendat poabrëanslés. (When he goes, he takes the keys.) (The selbo is the lumo the sentence refers to, and no one else.)

When the subject is also the object, selbié/selbio/selbia is used:

Jaurad luma vat, toutiène véyat selbia. (When she goes, everyone sees her.)

If another sentence refers to the object, then selbème/selbemo/selbema is used:

Jauno véyat l'aibũre. Selbème pódalerimat. (John sees the tree. It is very tall.)

When two or more items are stated, a reference number follows selb*/selbem*, as demonstrated in this translation from Herman Melville's Moby-Dick (Chapter 135, "The Chase — Third Day"):

«Plovarda! Moarttenière!—touƒtème moarttenière!» pékrivat Âhabo selboanadi; «selbtrigozé drève seulÁmerikaik!»
"The ship! The hearse!—the second hearse!" cried Ahab from the boat; "its wood could only be American!"

Here, selboanadi refers to the ship Captain Ahab himself is on, and selbtrigozé refers to "the second hearse".

Relative

In addition, the language possesses several relative pronouns, all of which are counterparts to the interrogatives and begin with j.

Pronoun Interrogative Parent morpheme English meaning
jiène kiène -íen who
jíenie kíenie -íen whom
jíenoze kíenoze -íen whose
jière kière -íer which, that

Nouns

Gendered nouns

In Relformaide, nouns given to members of the taxonomic kingdom Animalia are assigned masculine (male) and feminine (female) forms, ending in -o and -a respectively. This is the case for human beings, as shown below:

aumbro (man; male human) corresponds to aumbra (woman; female human). Similarly, edo (young male) corresponds to eda (young female).

Plurals are formed by adding an -s at the end:

aumbros (men)/aumbras (women); edos (young males)/edas (young females).

The above rules also apply to non-human animals:

chono(s)/chona(s) (cat[s]); nâyo(s)/nâya(s) (dog[s]); báleano(s)/báleana(s) (whale[s]).

Neuter nouns

When the subject's gender is unknown, undetermined, unspecified, or irrelevant, the neuter form (-e) is assumed; this also applies to groups and crowds with members in both genders:

aumbré(s) (person[s]); ríante(s) (parents); bíenède(s) (young person[s]); choné(s) (cat[s]); nâye(s) (dog[s]); báleane(s) (whale[s]).

Although they are living organisms, plants and trees are considered neuter:

plantile (plant), l'aibũre (tree), flũrile (flower), tovallé (garden/swamp weed).

All other nouns are considered inanimate, and also end with -e in most cases:

chairile (chair; concrete object), voasenile (summer; abstract concept), l'ästrettánule (astronomy; study/field).

If a neuter word's last three letters are e + C + e, then the penultimate e receives a grave accent (è):

moshelo (male meerkat), moshela (female meerkat), moshèle (meerkat).

Several dozen neuter nouns retain the original -o or -a of their original etymons, marking them with an underdot. This is the only scenario where le(s) and une(s) can be used as real words.

une kasínọ (a casino), plũrime mangạ (several manga volumes), le piezạ (the pizza), les toamạtọs (the tomatoes).

Masculine nouns

Relformaide boasts a small set of inherently masculine words rooted in Christianity and astronomy:

  • Paipo (the Pope),[29] Dómeno (the Lord), Jesúyo (Jesus), Kristo (Christ);
  • Júpito (Jupiter), Mạrzo (Mars), Mokilo (Mercury), Neptuvo (Neptune), Pluto, Satũrno (Saturn), Yureno (Uranus).

Feminine nouns

While inanimate in nature, Relformaide nonetheless follows English tradition with the feminine plova (boat)/plovarda (ship), and all other words for water transportation. Vênauza, the language's name for the planet Venus, is also feminine.

Occupations

For words in this category, the same rules for gender apply:

  • póliezíeno (policeman)/póliezíena (policewoman) — formed from pólieze (the police [force])
  • pîloto/pîlota (pilot)
  • flâmerivíeno/flâmerivíena (firefighter)
  • livrusontíeno/livrusontíena (librarian)
  • ódígíeno/ódígíena (leader [of a country]; driver [of a car])
  • kimo (king)/kima (queen)
  • kimódro (prince)/kimódra (princess)
  • dúko (duke)/dúka (duchess)
  • Senoro (Sro., Mister/Mr.)/Senora (Sra., Missis/Mrs.); Eusenora (Eusa., Miss/Ms.)
  • Aultado/Aultada (Highness)
  • (lo) Paipo (the Pope)

Proper nouns

Proper nouns are converted per Relformaide's spelling conventions. All first names of people are given -o/-a endings, depending on the subject's gender. Names of languages are treated as neuter, ending with -e. In similar fashion to Ido , place names can either end with -e or inherit the -o or -a from the original word, but remain neuter.

Examples include Jauno (John; male name), Samanfa (Samantha; female given name), Karibine (Caribbean; place name), Tokýo (place name with -o), Somalida (Somalia; place name with -a), and Mandạrine (Mandarin; language). For more, please see § Words and phrases.

Imported words

Surnames, terms such as QWERTY, and names of species (such as Homo sapiens) are mainly borrowed from English, and neither change their spellings nor inflect in Relformaide. This is the only instance where the language permits the use of c and q. (To mark pronunciation, c is converted into either ć [for /k/] or ç [for /s/].)

Verbs

All verbs in Relformaide end with -ar in their infinitive forms, and possess regular conjugation throughout. Unlike in the Romance languages, they are not marked for person, but rather for tense, mood, and sometimes gender plus number (as well as the perfect and prospective aspects).

Affixes

Suffixes

Infinitive verbs end with -ar, as many do in Spanish. Conjugation involves

  • -at for indicative forms;
  • -aik for the conditional and subjunctive moods;
  • -ait for the imperative;
  • -aijat for the prospective aspect;
  • -ant for the progressive aspect (present participle);
  • -asant for the continuative aspect;
  • -aid for past participles.

The -av- affix is also used for perfect forms, resulting in:

  • -avar, perfect infinitive;
  • -avat, perfect indicative;
  • -avaik, perfect conditional;
  • -antavat, progressive perfect;
  • -aijavat, prospective perfect.

Verbal nouns, meaning "an act of...", can be formed by adding at the end. Resultative nouns, meaning "the result of doing...", are formed with -ain followed by an e.

In certain advanced instances, verbs can inflect for gender by simply adding -o/-a after the tense/mood suffix:

Chuzlumato (He cooks), Troavlumata (She works), Plétozata (Used to be yours [addressing a female listener]).

These forms can also inflect in the plural, which is mandatory for the neuter -es:

Flegmatos (We are nurses; male speakers), Benkantábiltatas (You can sing well; to females), Dosellumatés (They are teachers; of males and females).
Prefixes

There are five tense markers (or augments) which are added before indicative forms, as well as conditional/subjunctive and continuative/progressive forms; the present tense has none. These are the only prefixes in Relformaide to end with vowels.

  • nupé-/(pre-vowel) nuper- — Recent Past
  • pé-/(pre-vowel) per- — Simple Past
  • plé-/(pre-vowel) pler- — Discontinuous Past
  • fé-/(pre-vowel) fer- — Future
  • péfé-/(pre-vowel) péfer- — Future in the Past

Negation is expressed with the nal- prefix, which always precedes the tense slot (and is the leftmost possible morpheme in a Relformaide word). Due to morphological processes, nal- always turns into naŕ- unless before a beginning vowel, b, h, or l.

Inflection types

Finite

Relformaide's two finite markers are -at (indicative/INF) and -ait (imperative/IMP).

duerat (SBJ love[s]), duerait (love!).
Nonfinite

The nonfinite markers are -ar (infinitive/INF), -ant (present participle/PR.PTCP), -aid (past participle/PST.PTCP), and -aik (conditional/subjunctive / COND/SBJ).

duerar (to love), duerant (loving), dueraid (loved [by an entity]), dueraik (could/would/should love; that SBJ love).

TAM system

Tense

As mentioned earlier, Relformaide expresses tense with nupé- (recent past/PST.REC), pé- (simple past/PST), ple- (discontinuous past/PST.DISC), fe- (future/FUT), and péfé- (future in the past/PST-FUT); the present is unmarked (Ø).

duerat (love[s]), nupéduerat (just loved), péduerat (loved), pléduerat (used to love), féduerat (will love), péféduerat (would love).
Aspect

Aspect is expressed with -ant (progressive/PROG), -asant (continuative/CONT), -avat (perfect/PERF), and -aijat (prospective/PRSP).

duerant (is/are loving), duerasant (is/are still loving), dueravat (has/have loved), dueraijat (is/are about to love).
Mood

Mood is expressed with the realis -at (indicative/IND), and the irrealis -ait (infinitive/INF) and -aik (conditional/subjunctive / COND/SJV).

duerat (love[s]), duerait (love!), dueraik (could/would/should love; that SBJ love).

-ait is also used for optative statements, formed with VSO:

Aubogait-suertiles fímondu tió. (May the odds ever be in your favour.)
Tüvvait-lumo, Díevómistu! (Let him go for Heaven's sake!)

Aktionsart

Aktionsart, German for lexical aspect, concerns the structure of verbs in relation to time.

Active

Active verbs in Relformaide are always transitive in SVO/OVS sentences, and intransitive in their passive forms.

Brîănto duerat Ashlêya. (Brian loves Ashley.)
Brîănto duerat Ashlêyieba. (Brian is loved by Ashley. = Ashley loves Brian.) (-ieb is the ergative case marker.)
Brîănto dueraidat. (Brian is loved. = Someone loves Brian.)
Stative

Intransitive verbs are stative, along with forms stemming from nominal, adjectival, and adpositional roots.

Esines toutwobu hódiezat. (These things always happen.) (Intransitive Verb)
Jauno plépîlotat. (John was once a pilot.) (Noun)
Esan'aumbro pódalat. (That man is tall.) (Adjective)
Wilmintaune proximat kortileń Ătlăntiki. (Wilmington is near the Atlantic shore.) (Adposition)

Unless expressed by the analytic avat -aid, statives do not assume passive conjugation:

Tés avat tristaid wob touti razile! (You guys have been sad the whole time!)

Valency

As in English, Relformaide contains five levels of verbal valency: impersonal, intransitive, transitive, ditransitive, and tritransitive. Valency determines the number of arguments that a verbal predicate takes.

Impersonal

With impersonal verbs, a dummy subject such as "it" precedes the predicate. In Relformaide, they do not conjugate in the passive voice at all; words dealing with weather conditions are the most common examples, among them pleuvar (to rain) and nívinar (to snow). Whereas English uses "It rains", Relformaide drops the subject entirely and simply uses Pleuvat.

Intransitive

In sentences with intransitive verbs, the subject (S) is the only argument involved. In Relformaide, intransitives also lack passive conjugation; examples include bar (be), hódiezar (happen/occur/take place), and adjective-/noun-/adposition-based constructs.

EsineS pléhódiezat. (This used to happen.)
PastelléS shoulat. (The cake is white.)
Mạriya SkautiS pékimat. (Mary of Scotland was a queen.)
Transitive

Transitive verbs normally take a subject (agent; A) and an object (O) (patient; P) in sentences; Relformaide examples include dezar (say), audar (hear), kantar (sing), toakar (play music/perform), skríbar (write), véyar (see), mitar (send), tenar (have/hold), and ódígar (drive/lead).

LumoS/A véyat l'ourème.O/P (He sees the city.)
LumasS/A kantat kantaidune.O/P (They [the women] sing a song.)

Transitives also behave like intransitives when no object is present, as in:

BobinoS troavat bromu. (Bobby works hard.)

Verbs of motion, such as var (go) and hendar (walk), are only transitive when referring to distance covered:

Nend méS/A hendaik kinsente pouzardesO/P—nend mé hendaik selbème shobrèles... (And I would walk 500 miles—and I would walk 500 more...)

In all other cases, they are nominally intransitive:

Lumozés móbileS plévat lesmu. (Our car was once a slowpoke.)

At least one verb, vũrdar (consider), uses a non-core argument (oblique; Ob) that governs the essive case:

LumésS/A pévũrdat lumiaO/P pualnandu.Ob (They took her for a fool.)
Ditransitive

Ditransitive verbs take a subject and two objects, one direct and the other indirect (I). Relformaide marks the indirect object with the ad adposition, as shown in this example involving óbrar (give):

MáfeuloS/A feróbrat l'aumbredadaI livriles.O/P (Matthew will give the books to the girl.)

Verbs whose stems end in -óbr govern the dative case:

RíantoS/A sidóbrardat fraulizällé.I (Daddy strongly approves of the union.) (Sidóbrar literally means "give a yes [sid] to".)
Tritransitive

Some sources classify English bet as tritransitive (taking three objects). The Relformaide equivalent, gealar, requires an indirect object, a direct object, and a relative or prepositional clause at most:

MaS/A gealat tadoI touve zanauris pasteltimesO/P jíer Bellwether férëvválimaidat.RCl (I bet you two slices of carrot cake that Bellwether gets re-elected.)
ShĩrlêyaS/A gealavat selboz'amitadeI deykíloge daulèresO/P aup suikolozlo bouŕálistesèle.PCl (Shirley's bet her friend $10,000 on the traitor's deserved possible abdication.)

Conjugation overview

The names of Relformaide's aspect and mood classes are named after the order in which the affixes are placed, and differ from standard terms used in other resources and publications.

Aspect/Mood Tense
Present Recent Past Simple Past Discontinuous Past Future Future in the Past
Simple R-at
do(es)
nupé-R-at
just did
pé-R-at
did
plé-R-at
used to do
fé-R-at
will do
péfé-R-at
would do
Progressive R-ant
is/are doing
nupé-R-ant
was/were just doing
pé-R-ant
was/were doing
plé-R-ant
used to be doing
fé-R-ant
will be doing
péfé-R-ant
would be doing
Continuative R-asant
is/are still doing
nupé-R-asant
was/were just still doing
pé-R-asant
was/were still doing
plé-R-asant
used to be still doing
fé-R-asant
will be still doing
péfé-R-asant
would be still doing
Progressive Perfect R-ant-av-at
has/have been doing
nupé-R-ant-av-at
has/had just been doing
pé-R-ant-av-at
had been doing
plé-R-ant-av-at
used to have been doing
fé-R-ant-av-at
will have been doing
péfé-R-ant-av-at
would have been doing
Continuative Perfect R-asant-av-at
has/have been still doing
nupé-R-asant-av-at
has/had just been still doing
pé-R-asant-av-at
had been still doing
plé-R-asant-av-at
used to have been still doing
fé-R-asant-av-at
will have been still doing
péfé-R-asant-av-at
would have been still doing
Progressive Prospective R-ant-aij-at
is/are about to be doing
nupé-R-ant-aij-at
was just about to be doing
pé-R-ant-aij-at
was/were about to be doing
plé-R-ant-aij-at
used to be about to be doing
fé-R-ant-aij-at
will be about to be doing
péfé-R-ant-aij-at
would be about to be doing
Progressive Conditional/Subjunctive R-ant-aik
ought to be doing;
...that SBJ be doing
nupé-R-ant-aik
was just ought to be doing
pé-R-ant-aik
was ought to be doing
plé-R-ant-aik
should have used to be doing
fé-R-ant-aik
shall be doing
péfé-R-ant-aik
should be doing
Continuative Conditional/Subjunctive R-asant-aik
ought to be still doing;
...that SBJ still be doing
nupé-R-asant-aik
was just ought to be still doing
pé-R-asant-aik
was ought to be still doing
plé-R-asant-aik
should have used to be still doing
fé-R-asant-aik
shall still be doing
péfé-R-asant-aik
should still be doing
Perfect R-av-at
has/have done
nupé-R-av-at
has/had just done
pé-R-av-at
had done
plé-R-av-at
used to have done
fé-R-av-at
will have done
péfé-R-av-at
would have done
Perfect Progressive[30] R-av-ant
having done
nupé-R-av-ant
having just done
pé-R-av-ant
having done previously
plé-R-av-ant
having done at one point
but not anymore
fé-R-av-ant
having done soon
péfé-R-av-ant
having done soon in the past
Perfect Prospective R-av-aij-at
is/are about to have done
nupé-R-av-aij-at
is just about to have done
pé-R-av-aij-at
was/were about to have done
plé-R-av-aij-at
used to be about to have done
fé-R-av-aij-at
will be about to have done
péfé-R-av-aij-at
would be about to have done
Perfect Conditional/Subjunctive R-av-aik
ought to have done;
...that SBJ have done
nupé-R-av-aik
just ought to have done
pé-R-av-aik
was ought to have done
plé-R-av-aik
should have used to have done
fé-R-av-aik
shall have done
péfé-R-av-aik
should have done
Passive Simple R-aid-at
is/are done
nupé-R-aid-at
is/are just done
pé-R-aid-at
was/were done
plé-R-aid-at
used to be done
fé-R-aid-at
will be done
péfé-R-aid-at
would be done
Passive Progressive R-aid-ant
is/are being done
nupé-R-aid-ant
is/are just being done
pé-R-aid-ant
was/were being done
plé-R-aid-ant
used to be being done
fé-R-aid-ant
will be being done
péfé-R-aid-ant
would be being done
Passive Continuative R-aid-asant
is/are still being done
nupé-R-aid-asant
is/are just still being done
pé-R-aid-asant
was/were still being done
plé-R-aid-asant
used to be still being done
fé-R-aid-asant
will be still being done
péfé-R-aid-asant
would be still being done
Passive Perfect R-aid-av-at
has/have been done
nupé-R-aid-av-at
had just been done
pé-R-aid-av-at
had been done
plé-R-aid-av-at
used to have been done
fé-R-aid-av-at
will have been done
péfé-R-aid-av-at
would have been done
Passive Progressive R-aid-aij-at
is/are about to be done
nupé-R-aid-aij-at
was/were just about to be done
pé-R-aid-aij-at
was/were about to be done
plé-R-aid-aij-at
used to be about to be done
fé-R-aid-aij-at
will be about to be done
péfé-R-aid-aij-at
would be about to be done
Passive Conditional/Subjunctive R-aid-aik
ought to be done;
...that SBJ be done
nupé-R-aid-aik
was/were just ought to be done
pé-R-aid-aik
was/were ought to be done
plé-R-aid-aik
should have used to be done
fé-R-aid-aik
shall be done
péfé-R-aid-aik
should be done
Prospective Simple R-aij-at
is/are about to do
nupé-R-aij-at
was/were just about to do
pé-R-aij-at
was/were about to do
plé-R-aij-at
used to be about to do
fé-R-aij-at
will be about to do
péfé-R-aij-at
would be about to do
Prospective Perfect R-aij-av-at
has/have been about to do
nupé-R-aij-av-at
has/have just been about to do
pé-R-aij-av-at
had been about to do
plé-R-aij-av-at
used to have been about to do
fé-R-aij-av-at
will have been about to do
péfé-R-aij-av-at
would have been about to do
Prospective Conditional/Subjunctive R-aij-aik
ought to be about to do;
...that SBJ be about to do
nupé-R-aij-aik
was/were just ought to be about to do
pé-R-aij-aik
was ought to be about to do
plé-R-aij-aik
should have used to be about to do
fé-R-aij-aik
shall be about to do
péfé-R-aij-aik
should be about to do
Conditional/Subjunctive R-aik
ought to do; could/would/should do; ...that SBJ do
nupé-R-aik
was/were just ought to do
pé-R-aik
was ought to do
plé-R-aik
should have used to do
fé-R-aik
shall do
péfé-R-aik
should do
Imperative/Cohortative R-ait!
do!
fé-R-ait!
let's do!

Conjugation examples

1: sujar

Equivalents of this verb are extremely irregular in various natural languages. Its conjugation in Relformaide is an exception, along with the Esperanto equivalent esti and Ido esar.

It is ungrammatical in English to precede be's past participle, been, with any other form of that verb. The same applies in Relformaide, where sujar lacks passive conjugation. The participle is retained in the perfect form, sujavat (have/has been).

Sujar is only used to introduce subjects in encyclopedia/news articles and biographies.

Infinitive: bar (to be)
Category Form Meaning
Affirmative Negative
Present sujat naŕsujat (I) am, (we/you/they) are, (he/she/it) is
Recent Past nupésujat naŕnupésujat (I/he/she/it) recently was, (you/they) recently were
Simple Past pésujat naŕpésujat (I/he/she/it) was, (you/they) were
Discontinuous Past plésujat naŕplésujat ...used to be
Future fésujat naŕfésujat ...will be
Future in the Past péfésujat naŕpéfésujat ...would be
Conditional
Subjunctive
sujaik naŕsujaik (that) SBJ be
Imperative sujait! naŕsujait! be!
Cohortative fébait! naŕfébait! let's (= let us) be!
Present Participle
(Progressive Active)
sujant naŕsujant be being
Continuative Active sujasant naŕsujasant still be being
Past Participle sujaid naŕsujaid been
Prospective sujaijat naŕsujaijat about to be, going to be
Passive Infinitive [31] [31]
Progressive Passive Infinitive [32] [32]
Past Participle Infinitive
(Perfect Active)
sujavar naŕsujavar (to) have been
Perfect Passive Infinitive [33] [33]
Verbal Noun sujé naŕsujé (the) being/existing
2: mankar

This root conjugates in both the active and passive voices. The French counterpart, manger, possesses regular conjugation (albeit with a few spelling adjustments where applicable). The English past tense form (ate) and past participle (eaten) deviate from the regular eat(s)/eating.

Infinitive: mankar (to eat)
Category Form Meaning
Affirmative Negative
Present mankat naŕmankat (I/we/you/they) eat, (he/she/it) eats
Recent Past nupémankat naŕnupémankat ...just ate
Simple Past pémankat naŕpémankat ...ate
Discontinuous Past plémankat naŕplémankat ...used to eat
Future fémankat naŕfémankat ...will eat
Future in the Past péfémankat naŕpéfémankat ...would eat
Conditional
Subjunctive
mankaik naŕmankaik (that) SBJ eat
Imperative mankait! naŕmankait! eat!
Cohortative fémankait! naŕfémankait! let's (= let us) eat!
Present Participle
(Progressive Active)
mankant naŕmankant be eating
Continuative Active mankasant naŕmankasant still be eating
Past Participle mankaid naŕmankaid eaten
Prospective mankaijat naŕmankaijat about to eat, going to eat
Passive Infinitive mankaidar naŕmankaidar be eaten
Progressive Passive Infinitive mankaidantar naŕmankaidantar be being eaten
Past Participle Infinitive
(Perfect Active)
mankavar naŕmankavar (to) have eaten
Perfect Passive Infinitive mankaidavar naŕmankaidavar have been eaten
Verbal Noun manké naŕmanké (the) eating
3: foartar

Adjectives can also be used to form verbs (meaning "to be ADJ"), but do not conjugate in the passive voice (as explained in Example 1). The noun form describes something that possesses the attribute in question.

Infinitive: foartar (to be strong)
Category Form Meaning
Affirmative Negative
Present foartat naŕfoartat (I) am strong, (we/you/they) are strong, (he/she/it) is strong
Recent Past nupéfoartat naŕnupéfoartat (I/he/she/it) was just strong, (you/they) were just strong
Simple Past péfoartat naŕpéfoartat (I/he/she/it) was strong, (you/they) were strong
Discontinuous Past pléfoartat naŕpléfoartat ...used to be strong
Future féfoartat naŕféfoartat ...will be strong
Future in the Past péféfoartat naŕpéféfoartat ...would be strong
Conditional
Subjunctive
foartaik naŕfoartaik (that) SBJ be strong
Imperative foartait! naŕfoartait! be strong!
Cohortative féfoartait! naŕféfoartait! let's (= let us) be strong!
Present Participle
(Progressive Active)
foartant naŕfoartant (be) being strong
Continuative Active foartasant naŕfoartasant still being strong
Past Participle foartaid naŕfoartaid been strong
Prospective foartaijat naŕfoartaijat about to be strong, going to be strong
Passive Infinitive
Progressive Passive Infinitive
Past Participle Infinitive
(Perfect Active)
foartavar naŕfoartavar (to) have been strong
Perfect Passive Infinitive
Adjectival Noun foarté naŕfoarté a strong thing; a fort
4: intar

Verbs can also be formed from adpositions; again, these forms do not conjugate in the passive voice. The past, past habitual, and future preverbs respectively become nuper-, per-, pler-, and fer- before this vowel-initial root.

Infinitive: intar (to be in)
Category Form Meaning
Affirmative Negative
Present intat nalintat (I) am in, (we/you/they) are in, (he/she/it) is in
Recent Past nuperintat naŕnuperintat (I/he/she/it) was just in, (you/they) were just in
Simple Past perintat naŕperintat (I/he/she/it) was in, (you/they) were in
Discontinuous Past plerintat naŕplerintat ...used to be in
Future ferintat naŕferintat ...will be in
Future in the Past péferintat naŕpéferintat ...would be in
Conditional
Subjunctive
intaik nalintaik (that) SBJ be in
Imperative intait! nalintait! be in!
Cohortative ferintait! naŕferintait! let's (= let us) be in!
Present Participle
(Progressive Active)
intant nalintant (be) being in
Continuative Active intasant nalintasant still being in
Prospective intaijat nalintaijat about to be in, going to be in
Past Participle intaid nalintaid been in
Passive Infinitive
Progressive Passive Infinitive
Past Participle Infinitive
(Perfect Active)
intavar nalintavar (to) have been in
Perfect Passive Infinitive
Adpositional Noun inté nalinté something that is indoors
5: skríbar

This is based on the rundown found in a late 19th-century book detailing the grammar of Manchu, an almost-extinct language of China.[34] (The progressive forms of this verb stand in for the translations marked as "Gerunds".)

Paradigm of skríbar (to write)
(Original verb: arambi)
Category Form Meaning
Manchu Relformaide
Imperative ara skríbait! write!
Present Tense arambi skríbat I write
Infinitive arame skríbar to write
Preterite araha péskríbat I wrote
Future arara féskríbat I shall write
Conditional araci skríbaik-mé should I write
Subjunctive Present araki skríbaik-lumo may [that] he write
Past Gerund arafi péskríbavantu having written
Imperfect arambihe péskríbant I was writing
Indefinite Past arahabi skríbavat I have written
Pluperfect arahabihe péskríbavat I had written
Past Conditional arahabici sif mé péskríbavaik if I had written
Adversative aracibe mólenoip lumo skríbaik although he may write
Concessive aracina skríbait-lumo may he write
Optative arakini seulsif lumo skríbaik would that he write
Gerund I arambime skríbwobu while writing
Gerund II arambifi skríbavantu having written
Gerund III aranggala skríbprevu before writing
Passive arambumbi skríbaidat it is written
Causative or Passive arambubumbi skríbaidinzat I cause to be written
Verbal Noun arahangge, ararangge skríbé, skríbiène the writing, the writer
Indefinite arahale, ararale jíenoipe skríbat whoever writes
Adverbial araralame skríbu in the manner of writing

Adjectives

Nominal

Adjectives and adjectival phrases in Relformaide end in -i, and come either before nouns (as in English) or after (as in French):

beli neklo (the beautiful tomcat), beli nekla (the beautiful queen cat), beli jalanile (the beautiful road), vèyaidune frolausomi (a sight most delightful).

Elision is also observed, as in bel'aumbredla (the beautiful girl).

As in the Romance languages, adjectives end in -is if the nouns they describe are plural:

friedifis aumbredlas (the gentle girls), foartis bomiles (the strong walls), nekedlos rouradis (the kittens in the countryside).

In a sentence of analytic SVD (subject–verb–description) syntax, the adjective is converted into a verb and eliminates zero copula:

droiti(s) jalanile(s) (the straight road[s]) — BUT jalanile(s) droitat (the road is/the roads are straight).
l'obenaubré (the new house), obenaubriles (the new houses) — BUT l'obène ńaubrat (the house is new), les obènes bat aubri (the houses are new).

Comparatives and superlatives

Comparative forms of adjectives are made by adding -ausm at the end of a root; superlatives are formed with -ausom. Suppletion is thereby avoided in those forms.

joali/joalausmi/joalausomi (good/better/best);
beni/benausmi/benausomi (well/better/best);
máli/málausmi/málausomi (bad/worse/worst);
osivi/osivausmi/osivausomi (ill/worse/worst);
aulti/aultausmi/aultausomi (high/higher/highest).

To express "less"/"least", -eusm and -eusom are used:

boulaidi/boulaideusmi/boulaideusomi (important/less important/least important).

Adverbs

Nominal

Adverbs, and adverbial phrases, end with -u. This is equivalent to the English -ly, French -ment and Spanish -mente in most given cases. Examples include:

belu (beautifully), fásilu (easily), óbrantu (generously), vúbegu (at all/somewhat), fuiwobetu (in a short while; adverbial phrase).

Comparatives and superlatives

Comparative and superlative adverbs are formed in the same manner as adjectives:

véyu/véyausmu/véyausomu (visually/more visually/most visually);
audaibilu/audaibileusmu/audaibileusomu (audibly/less audibly/least audibly).

Relative

Several relative adverbs are also used:

Pronoun Interrogative Parent morpheme(s) English meaning
jíeradu kíeradu -íer-ad where
jauradu kauradu maurad when
jóvaudu kóvaudu póvaud why
jemeku kemeku emek how

Adpositions

In Relformaide, adpositions are the same as the root forms they derive from, and also serve as standalone prepositions. Many of them are root affixes which double as postpositions after the nouns they modify.

Adposition Case English meaning
Possessive
der GEN of — belonging to
den POSS of — associated with
ten PROP with — having
zol PROP with — owning
Sociative
aseb COM with — accompanied by
auvek CNJ with an associated group
emek INSTR with the help of, aided by, via; on — supported by
aurd UTIL for, per
opon OPP/ANTI against, versus (in opinion)
ódim OPP/ANTI against, versus (in preference)
uslen TSP on behalf of, serving
nímel in charge of
Locative
ad LOC/DAT at, to;
on – by means of (medium)
obenad chez (French loan) – at the house of
int INE in
intad INE-LOC within, inside
intaup LAT/ILL into
inttug INE-ABL without – outside
tug ABL out(side)
tüvv ELA out of, away from
weg DEL off (surface), out of
íak SUPE on the surface of
íakad SUPE-LOC onto, upon, atop
róvinad aboard
ault ALT above, up, over
úlem SUBE below, under(neath), beneath, down
proxim APUD near, next to, near, (near)by
túdelad beside, at the side of, next to
ótrúdel opposite
aup ALL towards
tranz TRANS through, across
touttranz throughout
korttranz along
multint INTRS among a group of, amid
mẽrkint amidst, in the middle of
hout beyond, past
verhout behind, beneath
Quantitative
ond ABE without
nend INCL plus
moin EXC minus
prolad INTR between
tuig QUANT among those in a set, # of
oantuig SEL among, one of
Temporal
wob TEMP during
prev ANTE before, by (a given time)
áprev PCV after, following, since, past a time
ósot TERM till
ósótuit TERM-INTF until
Qualitative
prin REF on – about; concerning
nand ESS as – in the role of
símil FORM like
ótrid NEG.FORM unlike
kuam COMP as...as/than

Conjunctions

As with adpositions, conjunctions also assume their root forms.

Conjunction English meaning
nend and/plus
sed but – rather
moin but – except; minus
jemek how
jían so (that), in order to/that
ẽrgim so – therefore
sif if
oudin or – inclusive; either...or...or both; and/or
ouden or – exclusive; either...or; one or the other
nal...ouden neither...nor
daisen whether
daisen...(nal)ouden whether...or (not)
ausik as – in the same way
kuam as...as/than
nal not
póvaud because, since, for (obs./formal)
jíer that
mólen though
mólenoip although
jíerad where
sedwob whereas
voard whereas — being the fact that... (legal/formal)
sifmoin unless
jaurad when
prev before
áprev after
wob while
depuid since – from the time that...; afterward(s)
kíeroz whereof – of what/which
kíenoz whereof – of whom

Interjections

Depending on the word, interjections either remain unchanged from their root forms, or add an -(a)t to them. Examples include:

  • véyat (look!)
  • dez (say)
  • hinvéyat (now see here!)
  • hin (here), han (there)
  • juet (cool!; all right!)
  • réshan (an expression of surprise)
  • suiban (whatever)
  • áchat (profanity catch-all)
  • kal/bueb (s[...]t/c[...]p)
  • veid (h[...]l)
  • bouf (an expression of annoyance or dismay)
  • blek (signifying disgust; equivalent to English yuck)
  • puaz (this stinks; P.U.)

Special classes

Numbers

Cardinal

Relformaide's number system is akin to those of Romance languages:

# Relformaide English
Multiplicative
(d × n)
Additive
(dx + n)
0 naul naul- zero
1 oan one
2 touv two
3 trig three
4 kaut four
5 kin five
6 siez six
7 sep seven
8 oat eight
9 nev nine
10 dey deiz ten
20 hogel hogoiz twenty
100 sent seiz hundred
1,000 kílog kíloiz (below 1,000,000);
kílog- (above 1,000,000)
thousand
1,000,000 oanard oanard- million
1,000,000,000 kílogard kílogard- billion
1,000,000,000,000 oantéran oantéran- trillion
1 × 1015 kílöttéran kílöttéran- quadrillion
1 × 1018 oanexan oanexan- quintillion
1 × 1021 kílogexan kílogexan- sextillion
1 × 1024 oanyotan oanyotan- septillion
1 × 1027 kílogyotan kílogyotan- octillion
1 × 1030 oanardyotan oanardyotan- nonillion
1 × 1033 kílogardyotan kílogardyotan- decillion

Relformaide has two words each for 10 (dey/deiz), 100 (sent/seiz), and 1,000 (kílog/kíloiz); hyphens separate words for higher values like oanard (1,000,000), kílogard (1,000,000,000), and so forth. This helps distinguish a combination like kindeiśsep (57) from kindeysep (350; literally "fifty sevens"/50 × 7).

The number system is addition- and multiple-based, resulting in the likes of deizoan (11), deiśtouv (12), deiśtrig (13), deiśkin (15), touɖdey (20), touɖdeizoan (21), trïddey (30), käuddey (40), kindey (50), sepdeizoat (78), seiśtouɖdeiśnev (129), kinseiśtrïddeiśtrig (533), touӄkílöiddeiśsiez (2016), kindeiśtrïkkílog (53,000), seiśkäuddeykílog (140,000), and so forth. Complex numbers such as oanard-touƒseiśtrïddeiśkäukkílog-kinseiśsïeddeiśsep (1,234,567) are hyphenated.

Ordinal

Ordinal forms are denoted by the -tem suffix, as in oantem (first/1st), touƒtem (second/2nd), trïttem (third/3rd), deiśsieśtem (sixteenth/16th), touɖdeiśneƒtem (twenty-ninth/29th), and senttem (hundredth/100th). Values up to 31st (trïddeizoantem) are used in expressing simple calendar dates.

Fractional

Fractions are expressed with the -tim suffix, as in touƒtim (one half/½), trïttim (one third/⅓), kauttim (a quarter/¼), oattim (one eight/⅛), deytim (a tenth / 1/10), touɖdeiśkintim (a twenty-fifth / 1/25), and senttim (a hundredth / 1/100). If higher than one, then the dividend precedes the divisor in forms such as touve trïttim (two-thirds/⅔), trige kauttim (three-quarters/¾), kine ńoattim (five-eighths/⅝), sèpe deytim (seven-tenths / 7/10), deiśtouve touɖdeiśneƒtim (twelve in 29 / 12/29), and oanärttim (a millionth; 1/1,000,000). In the case of "sesquitertia" (an obsolete term for the 4:3 ratio), the fraction is converted into a whole word (kautraśtrïttime = 4 × ⅓).

Proximal

Estimated and approximate values are expressed with the -tam suffix, most notably in deuvam (a dozen; about twelve). Other examples include senttam (about 100), trïddeykílöttam (some 30,000), and oanärttam (roughly a million).

The -úpot suffix, meaning "near(ly)"/"almost"/"close to" in non-spatial expressions, is also used for numbers: käuddeyúpot (nearly 40), deiśkinsentúpot (almost 150), sieśkílogúpot (close to 6,000).

Decimal

Decimals can either be expressed with single numbers, or in fractions denoted by 10 to the nth power (10n). Thus, a number such as 0.421876 can be written as either naul point kaut-touv-oan-oat-sep-siez (as in English), or käusseiśtouɖdeizoankílog-oatseiśsepdeiśsieze ńoanärttim (421,876/1,000,000). Similarly, 4.21876 = kaut point touv-oan-oat-sep-siez or kaute nend deizoankílog-oatseiśsepdeiśsieze sënkkílöttim (421,876/100,000), and 18.6357 = deizoate nend siez-trig-kin-sep or seizoatdeiśsieśkílog-trigseiśkindeiśsèpe deykílöttim (186,357/10,000).

Serial

As in English, the dictation of years is split between the first and last duos of numbers, so that 1789 becomes deiśsèpe ńoatdeiśnev; 1854, deizoate kindeiśkaut; and 1945, deiśnève käuddeiśkin. If referring to 2000 or later, then something like 2006 can be interpreted as either touӄkíloiśsiez or touɖdèye naul-siez. Similarly, 2016 becomes touӄkílöiddeiśsiez or touɖdèye deiśsiez, and 2020 = touӄkílöittouɖdey or touɖdèye touɖdèye. (The last example is a rare example of reduplication in Relformaide.)

Telephone numbers are also recited by their digits. For example, 555-7824 = kin-kin-kin sep-oat-touv-kaut, and 867-5309 = oat-siez-sep kin-trig-naul-nev.

Multiplicative

Relformaide uses the -raz suffixoid to denote multiples of given numbers. It forms multiples with an -e, such as oanrazé (single); touvrazé (double); trigrazé (triple); kautrazé (quadruple); kinrazé (quintuple); and deyrazé (decuple). With -u, it corresponds to English "x-fold" or "times x", as in touvrazu (twice); kautrazu (fourfold, times four); seprazu (sevenfold); oatrazu (eightfold); deyrazu (tenfold); trïddeyrazu (thirty-fold); and sepdeiśseprazu (seventy-sevenfold).

Symbols

Relformaide's words for arithmetic operators are nend (plus/+), moin (minus/-), raz (times/×), puem (divided by/÷), and aibem (equal/=).

Dates

Days and months

Words for days of the week, and months of the year, are formed by placing díem (week) or masin (month) after the attribute they are named for.

Days
  • Díeɖdième (Sunday)
  • Loundième (Monday)
  • Hilddième (Tuesday)
  • Mẽrkoṛdième (Wednesday)
  • Joaldième (Thursday)
  • Froldième (Friday)
  • Brouddième (Saturday)
Months
  • Premmasine (January)
  • Duermasine (February)
  • Suertmasine (March)
  • Paummasine (April)
  • Troavmasine (May)
  • Fraulmasine (June)
  • Kainmasine (July)
  • Vinkmasine (August)
  • Eskolmasine (September)
  • Noulmasine (October)
  • Dolchmasine (November)
  • Fimmasine (December)

Notation

Numerical dates are written in either British (29/9/2016) or American (9/29/2016) form, and are fully written out in this matter: le touɖdeiśneƒtème der Eskolmasine touӄkílöiddeiśsiezi (the 29th of September 2016).

Interrogatives

All interrogative words in Relformaide begin with the letter k, and correspond to a core morpheme.

Word Parent morpheme English meaning Concept
kíer -íer what/which thing
kíerad ad where(by) location-related
kíeraup aup whither location-related
kíerint int wherein location-related
kíeresil esil whence location-related
kíerkout kout how much cost
kíerjían jían wherefore specific reason
kíersaum saum how much; how many amount, quality
kíerēmek emek whereby, wherewith means, instrument
kíen -íen who person
kíeni- -íen whom person
kíenoz- -íen whose person
kaurad maurad when time
kíermaurv maurv at what time time (on a clock)
kóvaud póvaud why reason
kemek emek how, wherein manner

Correlatives

With Esperanto, L.L. Zamenhof managed to compile a table of correlatives. Along with the original words (in italics), the Relformaide equivalents are presented in this replica.

Concept Category
Question (?) Indication
(this/that)
Indefinite
(some)
Universal
(each/every)
Negative
(no/not)
Quality kíer
kia
(what)
sohen
tia
(what/such a...!)
unēsen
ia
(some kind of)
toutēsen
ĉia
(every kind)
naulēsen
nenia
(no kind of)
Reason kóvaud
kial
(why)
ẽrgim
tial
(therefore, so)
unpóvaud
ial
(for some reason)
touŕpóvaud
ĉial
(for all reasons)
naulpóvaud
nenial
(for no reason)
Time kaurad
kiam
(when)
puiwob
tiam
(then)
oguil
iam
(sometime)
töurwob
ĉiam
(always)
naulwob
neniam
(never)
Place kíerad
kie
(where)
hinad/hanad
tie
(here/there)
unad
ie
(somewhere)
toutad
ĉie'
(everywhere)
naulad
nenie
(nowhere)
Manner kemek
kiel
(how)
emekēsin
tiel
(thus)
unemek
iel
(somehow)
toutemek
ĉiel
(in every way)
naulemek
neniel
(in no way; no-how)
Association kíenoze
kies
(whose)
esin(íen)oze/esan(íen)oze
ties
(this/that one's)
un(íen)oze
ies
(someone's)
tout(íen)oze
ĉies
(everyone's)
naul(íen)oze
nenies
(no one's)
Thing kíer
kio
(what)
esin/esan
tio
(this/that)
unēsen
io
(something)
toutēsen
ĉio
(everything)
naulēsen
nenio
(nothing)
Amount kíersaum
kiom
(how much)
plũrimēsan
tiom
(that much)
un
iom
(some)
tout
ĉiom
(all of it)
naul
neniom
(none)
Individual kíen
kiu
(who)
esin(íen)/esan(íen)
tiu
(this/that one)
un(íen)
iu
(someone)
tout(íen)
ĉiu
(everyone)
naul(íen)
neniu
(no one)

Relformaide also has its own correlative hierachy, shown below:

Concept Deixis Pronoun Parent morpheme(s)
Proximal Medial Distant Question (?) Relative
Distance proxim
near
proxam
far
proxaum
beyond
kíerproxem
how far
jíerproxem
how far
proxem
Where (Location) hinad
here
hanad
there
haunad
yonder
kíerad
where
jíerad
where
(h*n/-íer)ad
What (Subject/Object) esin
this
esan
that
esaun
yon
kíer
what
jíer
what
esen/-íer
Who (Person) esiníen
this one
esaníen
that one
esauníen
yon one
kíen
who
jíen
who
-íen
When (Time) noutwob
now
puiwob
then
puiwobard
far back
kaurad
when
jaurad
when
noutwob/puiwob/maurad
Why (Reason/Cause) póvaudesin
this reason
póvaudesan
that reason
póvaudesaun
yon reason
kóvaud
why
jóvaud
why
póvaud
How (Manner/Method/Way) emekēsin
this way
emekēsan
that way
emekēsaun
yon way
kemek
how
jemek
how
emek

Determiners

As with nouns, all words classified as determiners end in -o, -a, or -e, and adhere to the elision rules.

  • Both articles (l- and un-)
  • Demonstratives (esine/esane [this/that], esaune [yon])
  • Possessive pronouns (e.g. moz-, toz-, lumoz-)
  • Numerals before nouns (oane, touve, trige, kaute, kine, oate, deye, kindèye, sente, kíloge, oanarde)
  • Quantifiers (kuilibe [any], paulime [much/many], peuve [few], plũrime [several], une [some/a certain...], touƒtoute [both], toute [every/all])
  • Distributive words (aivé [each])
  • Interrogatives kière and kíenoze
  • Relative pronouns jière and jíenoze

Affects

Several Relformaide suffixes and suffixoids serve as affects to regular roots:

Size

  • Diminutive -et, which denotes miniature/dwarf forms of objects. Examples include ékuine (horse) → ékuinète (pony); aumbra (woman) → aumbreta (dwarfette/midget/munchkin); and toapé (stone) → toapète (pebble). -et corresponds to the impersonal free root aubim- (small).
  • Medial -ead, which indicates medium-sized objects. Commonly used with clothes and footwear-related roots, as in chámezeade (medium shirt); hoaseneade (medium pants); and jódeades (medium shoes). Corresponds to the position-specific mẽrk- (middle/central), along with the location-specific prolad- (between).
  • Augmentative -ard, which indicates large objects. Found in such derivations as plovarda (big boat = ship), laufarde (big air = atmosphere), toaparde (large stone = rock), and doarmíenarde (sleeping giant); corresponds to free roots aubaum- (size; impersonal) and eushel- (age). In Relformaide, aubaumiène (someone who is big in size), tuerārde (big creature) and aumbrarde (big human) are synonyms for English "giant".

Age

  • Nascent -eb, which denotes newborn/baby forms of creatures and plants: nèke (cat) → nekèbe (baby kitten); fazole (bean) → fazolèbe (bean sprout); aumbra (woman) → aumbreba (baby girl).
  • Juvenile -ed, which denotes young forms of creatures and plants. Examples include nèke (cat) → nekède (kitten); nâye (dog) → nâyède (puppy); aumbra (woman) → aumbreda (girl); and tuere (creature) → tuerède (cub/whelp). -ed always follows -et in forms such as ékuinétède (young pony).
  • Geriatric -eld, which denotes senior/elder creatures and old objects/customs: aumbra (woman) → aumbrelda (old woman); kima (queen) → kimelda (old queen); eskũro (squirrel) → eskũreldo (old squirrel); strúbaine (building) → strúbainelde (old building); besnèle (custom) → besneŕelde (tradition).

Suffixoids -eshoan (adolescent/teenager) and -eushel (adult/mature) can also be appended to animate roots, but are not encountered as often.

Focus

  • -uit/oip is used as an intensifier/differentiator (véyar = see/véyoipar = stare; ludes nend lúduites = games and sports), while -iruj and -erim signify emphasis (Luma peraigirújat = She really did it; Le guerte belerimat = The garden is very beautiful). -evoir translates to English "even" (Mevoires bengeplerimat = Even we can type very well).

Personal

Relformaide boasts eight personal particles, each with different levels of intensity:

  • Pejorative -ach, the strongest and most intensive, is a catchall profanity.
  • Three epithets of contempt—-puaz (literally "rotten"), -eskom (meaning "trash"/"garbage"/"rubbish"), and -úbel ("shoddy"/"of poor/low quality")—substitute -ach in family-friendly speech. Another substitute is -úzaum, or "pitiful"/"poor little...".
  • -uvam indicates that the subject is outstanding, exceptional, impressive, high-class, or brilliant in quality or behaviour.
  • Affectionate marker -yoab takes on various meanings, from "dear", "love", "honey", "sweetheart", and "cutie" in English, to "cher/chère" in French; the similar -yoaf stands for "attractive"/"hot".


Syntax

Constituent order

Subject–verb–object

Relformaide's default constituent order, like that of English, is Subject–verb–object (SVO). This sentence is typical of the SVO structure:

SámoS pémankatV ouranjés.O (Sam ate oranges.)

Subject–object–verb

Many natural languages possess a Subject–object–verb (SOV) order; in Relformaide, this leads to something like

SámoS ńouranjésO pémankat.V

which appears illogical and confusing; some may assume it was Sam's oranges who did all the eating, not Sam himself. To resolve this, the ergative marker -ieb- is employed between the root and the end marker, resulting in:

SámieboS ńouranjésO pémankat.V

Owing to Relformaide's agglutinative capacities, the last two words can resolve into a compound verb that translates into "orange-eating". In the first two examples, pémankat is a transitive verb; here, the resulting form is intransitive since no object follows it.

SámoS perouranjmankat.V

In this case, the subject is an absolutive proper noun that retains its original form, and the past tense augment pé- assumes its pre-vowel form of per-.

Object–subject–verb

Another constituent order, Object–subject–verb (OSV), is associated with Yoda of the Star Wars saga; otherwise, this is extremely rare in natural languages as a default order. Nonetheless, English uses it from time to time, as can Relformaide in certain situations like this one:

OuranjésO SámoS pémankat.V (Oranges Sam ate.)

OSV sentences can also utilise the -lé/-lo/-la and -uno/-una/-une suffixes:

LivrileO l'aumbraS pélezat.V (The book, the woman once read.)
Neu-Yorke,O S mouvveyámelat.V (New York, I always visit.)

Verb–object–subject

English employs a similar pattern, Verb–object–subject (VOS), in various humorous expressions. Relformaide also accommodates it in certain cases, such as:

SkríbatV une livré,O l'aumbredo.S (Wrote a book, the boy.)

or even more accessibly:

SkríbatV une livré,O ńesan'aumbredo.S (Wrote a book, that boy.)

Relformaide requires a comma plus an article, determiner, or demonstrative (esin/esan) before the subject in VOS statements, or else they could appear rather unnatural as well.

PémankatV ouranjés,O esane Sámo.S (He's eating oranges, that Sam.) (Note that esane does not decline into esano, as the subject's identity is not yet immediately known.)

Object–verb–subject

The regular subject and object can be reversed, leading to an Object–verb–subject (OVS) setup like:

OuranjésO pémankatV Sámo.S (Oranges ate Sam.)

Unmarked ergatively, this reads like an excerpt from a science-fiction story. Again, -ieb- must be used to distinguish the subject, as in:

OuranjésO pémankatV Sámiebo.S

which is equivalent to the passive statement "Oranges were eaten by Sam." Pronouns ending in -io/-ia/-ié are exempt from this rule, as demonstrated in the next section.

Verb–subject–object

As in French, the uncommon Verb–subject–object (VSO) is used to form question statements in Relformaide:

PémankatV SámoS lémaunes?O (Did Sam eat lemons?)

If pronouns are involved, they are attached to the verb with a hyphen:

Pémoavvéyat-téVS mio?O (Did you visit me?)

If the object begins with a vowel, then ń- is added for elision purposes.

Pémoavvéyat-lumaVS ńustiés?O (Did she visit you guys?)
PémankatV SámoS ńouranjés?O (Did Sam eat oranges?)

VSO is also found in optative statements, using imperative marker -ait:

Hildait-lumésVS joali hildlé.O (May they fight the good fight.)
NérelaitV DíevoS malaigíenesanes.O (May God forgive those who have commited sin.)
Tenait-jomlapinlaVS benstásénis ódrés.O (May the expectant rabbit have well-behaved kits.)
Vévait-toVS ńósot seiśtrige. (May you live to be 103.)

VSO (without the object) also occurs in "There is/was/were" statements, such as:

PéhanadatV doarmanti nâyunoS ńaibũrproximi...AdvP (There was a sleeping dog next to the tree...)
Note Note:

Advanced structure

With subjects and objects

Pronouns only

Standard Relformaide nouns and pronouns are unchanged in the nominative and absolutive forms. The accusative is only used in pronouns, as seen in the following variations of the simple sentence "She loves him". (Again, the ń- is placed on vowel-initial words if vowel termisons follow them.)

LumaS ńaimatV lumio.O
LumaS lumioV ńaimat.O (= She him loves.)
LumioO lumaS ńaimat.V (= Him she loves.)
LumioO ńaimatV luma.S (= Him loves she.)
AimatV lumio,O luma.S (= Loves him she.)
Aimat-lumaVS lumio?O (Does she love him? = Loves she him?)
Subject pronoun, object noun

If the subject remains a pronoun but the object is a noun, the ergative marker is not needed. Here, "She loves him" becomes "She loves the man", and luma is the same as before:

LumaS ńaimatV l'aumbro.O
LumaS l'aumbroO ńaimat.V (= She the man loves.)
L'aumbroO lumaS ńaimat.V (= The man she loves.)
L'aumbroO ńaimatV luma.S (= The man loves she.)
AimatV l'aumbro,O luma.S (= Loves the man, she.)
Aimat-lumaVS l'aumbro?O (Does she love the man? = Loves she the man?)
Subject noun, object pronoun

The reverse occurs in sentences such as "The lady loves him"; here, lumio is clearly distinguished as the accusative.

L'aumbraS ńaimatV lumio.O
L'aumbraS lumioO ńaimat.V (= The lady him loves.)
LumioO l'aumbraS ńaimat.V (= Him the lady loves.)
LumioO ńaimatV l'aumbra.S (= Him loves the lady.)
AimatV lumio,O l'aumbra.S (= Loves him, the lady.)
AimatV l'aumbraS lumio?O (Does the lady love him? = Loves the lady him?)
Nouns only

Relformaide exhibits split-ergative capabilities when both a sentence's subject and object are standard nouns, thus demanding the ergative marker when necessary. (See also the examples involving "Sam ate oranges" above.)

L'aumbraS ńaimatV l'aumbro.O (The woman loves the man.)
AumbrieblaS l'aumbroO ńaimat.V (= The woman the man loves.)
L'aumbroO ńaimatV l'aumbrieba.S (The man is loved by the woman./The woman loves the man.)
Aimat-aumbrilaVS l'aumbro?O (Does the woman love the man?)
L'aumbroS ńaimatV l'aumbra.O (The man loves the woman.)
AumbriebloS l'aumbraO ńaimat.V (= The man the woman loves.)
L'aumbraO ńaimatV l'aumbriebo.S (The woman is loved by the man./The man loves the woman.)
Aimat-aumbriloVS l'aumbra?O (Does the man love the woman?)

With indirect objects

Below is another example of SVO in Relformaide:

AnyaS proadatV (le) fáloave.O (Anne sells [the] bread.) (As in English, use of le [the] before the object is optional.)

When indirect objects are involved, -ad is placed in the word referring to the receiver:

AnyaS ńóbratV l'aumbrédadoI (le) fáloave.O (Anne gives the boy [the] bread.)
AnyaS ńóbratV nekälléI (le) leché.O (Anne gives the cat [the] milk.)

If the object precedes the indirect, then either -ad is used standalone:

AnyaS ńóbratV fáloave/fáloavléO ńad l'edo.I (Anne gives [the] bread to the boy.)
AnyaS ńóbratV leché/lechileO ńad neklé.I (Anne gives [the] milk to the cat.)

or the case-converted word becomes an adverbial phrase:

AnyaS ńóbratV fáloave/fáloavléO tuomedadu.AdvP (Anne gives [the] bread to the boy.) (The -u precedes the gender marker in masculine or feminine dative nouns.)
AnyaS ńóbratV leché/lechileO nekadu.AdvP (Anne gives [the] milk to the cat.)

With appositive phrases

Apposition involves the use of two phrases, one of which serves to identify the other. For example:

Montserrat,S a volcanic island in the Caribbean...A

Here, Montserrat is the antecedent subject, and volcanic island... is the appositive phrase describing it. In Relformaide, the suffix -erij-, which stands in for jíer- (which/that) and jíen- (who), marks the noun in the appositive:

Maunzerạte,S vaulkenif'ansulunerije ńad le Karibine...A

When the appositive phrase refers to an occupation or role, -erij- is still used:

Moza Ríanto:S JoalerijoA (My Father the Hero, title of a 1991 French comedy and its 1994 U.S. remake)

This example is similar to those in the previous section:

Anyieba,S moza siblerija,A bematV fáloave/fáloavlé.O (Anne, my sister, buys [the] bread.)

This sentence makes use of both apposition and indirect objects:

Luma,S trúmeni lapinedunerija,A peróbratV selboza ríantadaI zanaurunes.O (She, a kind rabbit girl, gave her mother some carrots.)

When the appositive phrase describes the object, -erij- marks the appositive's noun:

LumaS pémoavatV selboza ríanta,O fami kantíenunerija.A (She met her [own] mother, a famous singer.)

This rule also takes effect in questions such as:

Dúbitat-téVS mia,O joalausomi krikítíenlerija yorbu?A (Are you doubting me, the best cricket player around?)

In certain sentences with at least two subjects or objects, one must apply -erij- to tell the apposition apart from other subjects. In an English sentence such as:

She, the queen, and several others were going.

it is hard to tell whether "she" refers to the queen or someone else. In Relformaide, this resolves to:

Luma, la kima, nend plũrimótrés pévant.

At face value, the "luma" refers to someone else who is not the queen. If she really is the queen, one can prevent ambiguity by saying:

Luma kimeriji nend plũrimótrés pévant. (= She [identified as the queen] and several others were going.)

This is helpful in more complex scenarios, such as:

Lumo, toujíenunerijo, nend luma, duerskríbíenunerija, péfraulaijat. (He, an adventurer, and she, a romance writer, were about to marry.)

-erij also translates English of, and French/Spanish de, when they stand for "also known/named as/called":

L'ouremardeS Chíkágerijo...A (The city of Chicago...)
Le vaulkenif'ansuleS Maunzerạterije...A (The volcanic island of Montserrat...)
Note Note:
An example of appositives in conlangs can be found in the grammar of Nåmúþ, a fictional constructed language constituting part of the Akana universe. Among natural languages, Basque provides some specimens involving the ergative case; see "Examples (1943) and (1944)" in Hualde and de Urbina (2003), p. 804https://books.google.com/books?id=nIaPL4kLt6cC.
See also the notes on apposition in Rick Harrison's grammar of constructed language Vorlin (2006 revision).

With relative clauses

-erij- is also found between the stem and termison(s) in the target verbs of relative clauses:

Puerchedlo, ńaulmaderijat vuoli strouzetile, ńäuvvant toargenile. (The little pig who lives down the lane is going to market.)
Mé pévéyat plovärlla, péresterijat eskellés Kastriezis nënzzol pépauzavat aurd un'abimaubré ńágrotvíandi. (I saw the ship, which was docked in Castries harbour and had stopped for a fresh supply of livestock.) (Note the zol after nend; otherwise, the speaker is assumed to go for the livestock, not the ship. Zol's role is discussed in the Suffixaufnahme section.)
Voaŕinealle péchaulferijat intad mozés ouremète nouprev sieze sepèmes teśnanderivaidant. (The plane that crashed into our town six weeks ago is quickly being cleared away.)
Lumo moz'amitat duerṣerijat mośsiblieba. (He is my friend whom my sister loves.) (The last two words literally mean "who is loved by my sister"; note the epenthetic buffer and ergative -ieb.)
Luma mozo siblat aubimausmerijat mikuamuo. (She is my sister, who is smaller than me.)
Luma mozo siblat m'aubimausmerijat selbkuamu. (She is my sister, whom I am smaller than.) (Here, selb- agrees with luma.)
Note Note:

This was inspired by the examples in "TL", a Latin-based conlang by Redditor "cyprinus_carpio" (May 2018 thread).

With passive voice

While various languages can handle passive statements such as "The biscuit is eaten by Anne", Relformaide has no exact equivalent for the word by in this context. OVS order (and -ieb- for the ergative subject) must be used, resulting in:

Guispellé O mankatV Anyieba.S (Anne eats the biscuit.)

As passive voice is generally discouraged in some circles, the same sentence can be simply written as Anya mankat guispellé.

Below are another two sentences in the passive voice:

GuispellésO bematV Anyieba.S (The biscuits are bought by Anne.)
GuispellésO proadatV lẽrníenieblas.S (The biscuits are sold by the girl students.)

When the past participle form of a verb is also a stem, an OVS variant—OV(I)S—also applies:

GuispellésO óbraidatV Anyada.I (The biscuits are received by Anne.) (Óbr-aid literally means "be given", and the dative -ad- indicates whom it was given to. One can further use something like ...noagelíeniebla [...through the lady swimmer] to ergatively indicate the giver.)

With past participles

This example from a late 19th-century edition of McGuffey's Reader demonstrates the use of the past participle:

NedoS paskulavatPPV poulla.O (Ned has fed the hen.)[35]

With adjectives as object complements

In the English sentence "The cards painted the roses red", the adjective "red" is the object complement, and assumes the translative case in its Relformaide counterpart:

KärllésS pésivelaigatV roazilesO roubizu.AdvP

Synthesis

These four versions of a sentence from the same volume, "The cat is on the mat"[36] (with an indirect object, but none direct), demonstrate Relformaide's syntactic flexibility:

  1. Neklé ńíakat gófrile. (Slightly agglutinating, 10 morphemes / 3.333/word)
  2. Neklé gófríakat. (Agglutinating, 6 morphemes / 3/word)
  3. Gófríännèke. (Polysynthetic, 4 morphemes in a single word)

Emphasis increases as more affixes combine to modify the root. The first two forms are more or less written as in English, and the third approaches levels seen in Hungarian, Turkish, and Finnish among others. In all three, the emphasis is placed on gófr- and its location thereof.

The last structure is typical of many an indigenous language of the Americas (such as Greenlandic and Central Alaskan Yup'ik in the Eskimo-Aleut family). Here, the focus shifts to nèke, the subject of the original English sentence. As glossed, it essentially translates to "mat-on_surface-cat-N" (gófr-íak-nek-é).

Emphasis

Topical

A special intrafix in Relformaide, -eun- (from Korean eun [은]), serves the same function as the English phrases "speaking of", "as for", "on the subject of", and "when it comes to". In a sentence such as:

Mariya ńaimat selboza ríanta. (Mary loves her mother.)

the subject can be converted into a topic, resulting in:

Mariyeuna ńaimat selboza ríanta. (As for Mary, she loves her mother.)

-eun- is also employed when the second half of certain sentences discusses an aspect of the first, as shown in:

Stuveuno Hopps sujat holómi zanauragrotíenuno jíenoz'ódra voulbremat póliezíenuna. (Speaking of Stu Hopps, he's a prosperous carrot farmer whose daughter wants to be a police officer.)
Dauruētesinēunu, lẽrníeniles kríonat. (Speaking of this class, the students are smart.)

The intrafix is also seen in cleft sentences, such that the first example can be interpreted in English as "It is Mary who loves her mother". In these translations of samples from the Wikipedia article linked to in this paragraph, the object or focus verb is tagged regardless of word order:

  • Joêyeuno més lorgant. (OSV)/Més lorgant Joêyeuno. (SVO) (It's Joey [whom] we're looking for.)
  • Dínereune m'aimat. (OSV)/M'aimat dínereune. (SVO) (It's money that I love.)
  • Esil Jaunēuno luma peraudat les nívites./Luma peraudat les nívites esil Jaunēuno. (It was from John that she heard the news.)
  • Lumo pévoulbemat une Fiateune. (What he wanted to buy was a Fiat.)
  • Seulu jaurad més peradvenēunat le hótèle, més pémoavat lumia. (It wasn't until we arrived at the hotel that we met her.)

-eun is also used for gnomic statements and proverbs:

  • Soleune premat eastadu, nend fímat westadu. (The sun rises in the east, and sets in the west.)
  • Dinereune naŕkrezat aibũríaku. (Money doesn't grow on trees.)
  • Díeveuno ńauxilat esanes jière sebhaizat. (Heaven helps those who help themselves.)

Negation

This sentence is a normal example of a negative statement in Relformaide:

Mé naŕperäuvvat plájile. (I didn't go to the beach.)

The nal- negator can also be placed before the subject or object for emphatic focus, resulting in:

Naŕmé peräuvvat plájile. (I didn't go to the beach [but someone else did].)
Mé peräuvvat nal plájile. (I didn't go to the beach [but somewhere else].)

Interrogation

Another special intrafix, -iruj-, is used in questions for emphasis on the topic or action. -iruj- is borrowed from the Austronesian Marshallese language, and functions similarly to Bulgarian li (ли). For these variants of the statement Peräuvvat-té le plaje? (Did you go to the beach?), the focus in the English equivalent is underlined.

Peräuvvirújat-té plájile? (Did you go to the beach?) (The respondee may have done something else there, or just stopped over.)
Peräuvvat-tiruje plájile? (Did you go to the beach?) (Some friend of the respondee may have visited instead of them.)
Peräuvvat-té plájiliruje? (Did you go to the beach?) (The respondee may have visited a different beach.)
Peräuvvat-té plájirujile? (Did you go to the beach?) (The asker expected the respondee to be there, but the latter may have decided to call it off or change plans.)

Self-interrogation is accomplished with -ódiv-, which carries the same meaning as Finnish han/hän ("I wonder if...") and is found mainly in verbs:

Äuvvirujódivant-luma Guadloupe? (I wonder if she's really going to Guadeloupe?)
Kière hódiezódivat hanadu? (Hmmm...what's going on over there?)
Tortougesano vuelaijódivat noutadu; mo févinkirújat selbio! (Beats me if that tortoise is gonna run now...I'll beat him for sure!)
Note Note:
These usage examples were inspired by:

Evidentiality

Relformaide has four evidentiality particles used in verbs: -zeg, -zem, -zeng, and -zev.

zeg

-zeg denotes reported statements and indirect speech in sentences. This counterpart to dez- (say) is borrowed from Dutch, and is equivalent to English "they say/it is said (that...)".

  • Lumo pézegat selbo félezat livrile. (He said, "I'll read the book.")
  • Proadune premzegat nodemu. (They say a sale will begin today [literally, "A sale is said to begin today"].)

zem

-zem, derived from the Hungarian szem (eye) and corresponding to vey- (see), denotes visual evidence by a speaker.

  • Lumèdes ludzémant plájadu. (I see kids playing on the beach.)
  • Loaklo pévuelzémat bankintaupu. (The madman was seen running into the bank.)

zeng

-zeng, a portmanteau of the French and Ket words for "feel", corresponds to toag(aid)- (touch[ed]) as the sensory evidence marker.

  • Choulzengat. (Feels cold.)
  • Húminile péziginzengat. (The ground shook [I felt it].)

zev

-zev, based on the Tamil/Telugu/Malayalam cevi (ear), corresponds to aud- (hear) and marks auditory evidence.

  • Nâylés péwoufzevat túgadu. (I heard the dogs barking outside.)
  • Sîgloanile proximäuzzévant miés. (Hear those winds? That hurricane's coming closer to us.)

Possessives

Relformaide employs six morphemes to express alienable and inalienable possession. Alienable items are likely to part from their possessor at any time, while inalienable items are inseparable from them.

Possession type Morpheme type
Possessive intrafix Proprietive intrafix Particle
Alienable -orz- -ten den
Inalienable -oz- -zol der

In Relformaide, English -'s is represented by the genitive/inalienable -oz- and the possessive/alienable -orz-. Either intrafix is always placed between the root and the gender marker in nouns, as in:

Jainoza vaulde (Jane's ball, i.e. the one she owns); mozríantozo kloché (my father's clock, i.e. his heirloom); julgíenozla livrés (the judge's books, i.e. those she needs for her job); Jaurjorzo véyeande (George's glasses, i.e. what he borrowed from a friend).

When a possessed object's first letter is a vowel, the ń- is placed directly before it:

lumozo ńeandes (his tools); mozríantorzo ńansule (my father's island, i.e. the place he calls home).

Whereas Romance languages use phrases such as la maison de mon oncle (French)/la casa de mi tío (Spanish), Relformaide uses either

l'obène der lumoza ríablo (the house of her uncle)

or

l'obène lumoza ríablozońi (her uncle's house)

where a genitive adjective states whom the house belongs to. In certain complex cases, the standalone der/den is employed:

Int chórivune der La Mancha, jíenoze naume mé naŕfévoulchembat... (In a village of La Mancha whose name I hardly wish to recall...)
L'ótrizaine der livré... (The declension of the word livré [book]...)
Lofärllé den Nóaho... (The mountain of Noah [Mount Ararat])

A pronoun stem followed by -oz-/-orz- also forms possessive-based prefixes before nouns:

esine mozobène (this house of mine = this in-my-possession house); torśmóbillé (the car you're renting = the in-your-holding car).

If the possessor's gender is stated, then either tuom- (masculine) or tuam- (feminine) is placed before the target stem. Tuom- and tuam- are chiefly found in complex compounds, substituting the respective -o and -a termisons.

tüallumozlivresane (that book of hers); tuommorśsátrokesine (this hat of mine; male speaker).

Relformaide also utilises the rare proprietive case through another two intrafixes, -ten- (indicating alienable possession or bodily/emotional conditions) and -zol- (indicating ownership). The proprietive denotes an item owned or held by a sentence's referent, and is virtually nonexistent in natural languages outside several indigenous ones from Australia (including Martuthunira and Kayardild).

Mé pémoavat l'aumbra móbilärnzoli. (I met the lady with the big car.)
Dúbitat-tiebé mio, ńaumbrunerijo sohenvalïtteni? (Are you doubting me, a man of such worth?)

Suffixaufnahme

At least eight morphemes in Relformaide—-ieb-, -oz-, -ten-, -zol-, -ad-, -emek-, -aseb-, and -auvek—exhibit traits of the linguistic phenomenon known as Suffixaufnahme, German for "case stacking".

Ergative

-ieb- helps distinguish the agent of a sentence in complex OSV/SOV sentences.

ChónuneO shouliebi mórinuniebeS mankat.V (A white mouse eats a cat.) (For more examples, see § Precision: Cat and mouse.)

In passive-voice sentences, only the noun is tagged:

Dolzile pémankat shouli mórinieblé. (The sweet was eaten by the white mouse.)

-ieb- also represents English both in sentences such as:

Ríkiebo nend Nanzieba vuelat eskolaupu díemaivu. (Both Rick and Nancy run to school every day.)

Possessive

When two or more subjects are associated with an entity, -oz- tags all of them as "joint possessives" or "compound possessives". If the subjects share the same entity, then the comitative aseb conjunction precedes the last subject:

Jaunozo ńaseb Mariyoza pastèle (John and Mary's cake)
Jaunozo, Mariyoza, ńaseb Jóséfozo pastèle (John, Mary, and Joseph's cake)

If the subjects have at least one of the entities described, then nend is used instead; the entity noun must be pluralised.

Jaunozo nend Mariyoza pastèles (John's and Mary's cakes)
Jaunozo, Mariyoza, nend Jósefozo pastèles (John's, Mary's, and Joseph's cakes)

A similar situation happens with -ten- and -zol-:

Luma pémoavat un'animaldauktera, Saint-Îvésenäuvvantzoli sepchonzoli móbiŕzolinti. (She met a lady veterinarian, bound for St. Ives with seven cats in her own car.)
Taumazo perauxilat l'eda nâyteni ńobententúgi. (Thomas helped the girl, who was outside her [relatives'] house, with the dog.)

Dative

In addition to its normal function as an indirect/dative marker, -ad- can also tag attributes associated with the indirect object of a sentence.

Ma peróbrat l'aumbrado Bilbáonesiladi moza dreƒfonaistière. (I gave my xylophone to the man from Bilbao.)
Ma peróbrat l'aumbrado noulsátrokzoladi lumoza vailtenière. (I gave the man with the black hat her suitcase.)

Instrumental

This example replicates a specimen found in Double Case: Agreement by Suffixaufnahme (Plank (1995), p. 400https://books.google.com/books?id=0EDl8_gmmvQC):

Mé péshoubat vuemiles tüossiblozemeku reshilēmeku. (I caught the fish with the brother's net.)

Comitative

As with the instrumental, the comitative also tags associated genitives.

Mé pévat tüoppastorövasebu tuamódrasebu. (I went along with the pastor's daughter.)

Below is another example, based on Plank (1995), p. 84https://books.google.com/books?id=0EDl8_gmmvQC; here the last complex plural agrees with the first modifier.

Més pévéyat elda ńótröutasebi guertesilöutasebi. (We saw a/the old woman with other people from the garden.)

Conjunctive

The conjunctive particle -auvek behaves the same way as -aseb:

Lumés tánulat Baibullé Díeɖdièmaivu Jaunozauveku tírombainauveku. (They study the Bible every Sunday as part of John's congregation.)
Note Note:
The above examples were inspired by this reply by "Valdeut" (from a March 2016 discussion entitled "Basque's Surdéclinaison") at the zompist bboard.
The use of -ieb- to denote "both" was inspired by the West Greenlandic sample at the Leipzig Glossing Rules page.

Quotations and punctuation

Relformaide is written similarly to English; its quotation marks («», wilémètes) are borrowed from French.

«Kiène lumbat?» pézegat l'aumbrelda.
«Mio, Róbairto,» péverjautat selboz'ódro.
"Who is it?" said the old lady.
"It's me, Robert," replied her son.
«Mo naŕpékroyábilat lumé jaurad julgíenlo pézégat, ‹Naule pointes!›» pézégat l'edo.
"I couldn't believe it when the judge said, 'Nul points!'" said the boy.

Otherwise, its punctuation system remains unchanged.

Zẽrdéblo ńäggaunat trig'eldausmis síblas—l'oantema toutrazu triesti, touƒtemla froŕámeli, nend l'aulttema duimétanti.
The fennec kit has three older sisters—one forever sad, another occasionally happy, and the last a slight bore.
Ma naŕperaigat lumié, ma shũrat; ótrune péhaizat!
I didn't do it, I swear; someone else did!
Péhanat seulbíenuna mo péfídeŕábilat: Sebmoz'eldoubranti doslíena.
There was only one person I could trust: My own aging teacher.
Esane droalat... Esine móbile pékoutat madés oanard'íroaves (€1,000,000)?!
That's funny... This car cost us a million Euros (€1,000,000)?!

Precision

Precision is one of Relformaide's main objectives, in that words and expressions (regardless of length) should be as free of ambiguities and semantic baggage as possible.

Filed under Sesquipedalian

p45: Safety not guaranteed

Commonly cited as one of the longest words in English, pneumonoultramicroscopicsilicovolcanoconiosis—defined as "a disease of the lungs caused by inhaling volcanic silica dust"—is actually a coinage devised by the U.S. National Puzzlers' League in 1935. Consisting of 45 letters (hence the code nickname "p45"), it contains the following morphemes:

  • pneumon (Greek for "lung")
  • ultra
  • microscopic
  • silico-
  • volcanic
  • konis (Greek for "dust")
  • -osis

"p45" is an extension of the more concise term pneumoconiosis (14 letters), whose simpler form is just the nine-letter silicosis. Relformaide translates it into the 20-letter silíkomẽrgimosivaine, or "illness caused by silicon" (silíkom-ẽrgim-osiv-ain-e; silicon-CAUS-sick-RSLT-N), or the more concise 11-lettter maŕsilíkome (mal-silíkom-e; bad-silicon-N). Lung disease, on the other hand, is peumaunosivaine (peumaun-osiv-ain-e; lung-sick-RSLT-N; 15 letters). An attempt to capture the essence of the original word results in

silíkomẽrgimpeumaunosivaine

which is 27 letters—more than half the English counterpart's—and six morphemes (silíkom-ẽrgim-peumaun-osiv-ain-e) in length.

Still long in the tooth

Through derivational and compounding methods, and on the basis of their original definitions, long words in various languages can receive Relformaide equivalents.

  • A well-known English example, the 28-letter antidisestablishmentarianism, means "a movement opposing the separation of church and state". This becomes the similarly long
gúdentüvvúzemōponogrile
which comprises seven morphemes in total (gúden-tug-v-územ-opon-ogril-e; church-EXE-go-state-OPP-movement-N).
  • The 29-letter Latin-based coinage, floccinaucinihilipilification (the act of estimating something as worthless), turns into the 15-letter/five-morpheme
vervalnänvvũrde
which consists of "ver-val-nand-vũrd-e" (RVT-worthy-ESS-consider-N).
  • Honorificabilitudinitatibus, a 27-letter Latin loan meaning "the ability to achieve honours", is found in Love's Labour's Lost by Shakespeare. Its Relformaide equivalent, námúzoṛganábilídé, comprises 16 letters and five morphemes (námuz-oṛ=gan-ábil-id-e; honour-BFR.IFX.A=win-ability-ABST-N).
  • Pōkamuṭiyātavarkaḷukkāka (போகமுடியாதவர்களுக்காக), Tamil for "for the sake of those who cannot go", is converted into nemvábilouzómistu (nem-v-ábil-ouz-ómist-u; NEG-go-POTE-CPLX.PL-BEN-ADV).
Note Note:
The Tamil example is taken from Wikipedia's article on Tamil grammar, which also provides a morpheme breakdown.

See also World Wide Words for more information on the terms profiled in the preceding two sections:

Which way to say it?

To have...or not to have

Consider the following sentence, taken from Wiktionary's entry for "have":

I have no German.

On the surface, it can be interpreted as one of several things:

  1. The speaker has no one from Germany living in their house.
  2. The speaker has no German heritage.
  3. The speaker cannot converse in the German language (the "have" in this example uses the Irish English sense of "speak a language").

Translated straight into Relformaide, this becomes:

Mé naŕtenat Alémane.

Since the last word primarily refers to the country (and in some contexts, the language itself), one is led to believe that the speaker is missing the realm of Germany—perhaps as a piece of a map puzzle, or part of a souvenir collection. To clear things up, one can say instead:

Mé naŕtenat Alémaniènes. (I have no German people.)

Simply put, the speaker is declaring that no German people are at their house, or in their family tree. To be more specific, one can use:

Mé naŕtenat Alémaniènes mozaulmadu/aulmad madé. (I have no Germans at my [own] house.) (The speaker is referring to citizens of Germany. Also, note the indirect/dative pronoun at the end of the second form.)
Mé naŕtenat Alémanesiles mozaulmadu/aulmad madé. (I have no one from Germany at my house.) (The speaker is referring to those who come or hail from Germany, but were not necessarily born there.)

This can also be rendered as:

Alémaniènes/Alémanesiles naŕmozaulmadat/naŕaulmadat mié. (No Germans live at my house.)

or the more morphologically complex:

NaŕmozaulmadAléman(esil)aulmiènes.

which consists of 10 (or 11) uninterrupted morphemes à la Greenlandic et al. (nal-m-oz-aulm-ad-Aléman-[esil-]aulm-íen-e-s), and retains the capital letter denoting a proper noun. Owing to their sheer length, these kinds of one-word sentences are all but infrequent in Relformaide.

When speaking of one's lineage or heritage, this can be used:

M'ondat Alémangaunés./M'Alémangaunondat. (I am without German lineage/descent. / I have no German ancestors.)

Finally, when referring to fluency, this statement captures the essence of the original Irish usage:

Mé naŕparŕábilat Alémane/l'Alémanlinge. (I cannot/can't speak German.)

Food from the sky?

In section 19c of his "Ranto" essay, Justin B. Rye posits this conundrum:

"...Does la fiŝoj estas bongustaj sed pluvas mean 'the fish are tasty but it's raining' or 'the fish are tasty but are falling as rain'?"

The same confusion carries over to the Relformaide equivalent, Les vuemes joalgustat, sed pleuvat. To differentiate, one should write Vuemiles joalgustat, sed pleuvlé toambat. (Pleuvat remains acceptable as a one-word weather statement.)

Dogs and cats

Wikipedia's Esperanto grammar page provides the English sentence, The dog chased the cat in the garden, as another example of ambiguity. In Esperanto and Relformaide respectively, this becomes the equally ambiguous

La hundo ĉasis la katon en la ĝardeno. (EO)
Nâylé péstauzat neklé ńint guertile. (RFM)

It is not easily evident whether the dog or cat resides in the garden, or the dog began chasing the cat somewhere else. Thanks to its case inventory, Relformaide resolves the context problem in one of several ways:

  1. Nâylé guertesili péstauzat neklé. (The dog comes from the garden itself.)
  2. Nâylé péstauzat neklé guertesili. (The cat came from the garden.)
  3. Nâylé péstauzat neklé guerttranzu. (The chase began somewhere else, and the garden is part of the way.)
  4. Guertintádu, nâylé péstauzat neklé. (The chase is taking place within the garden.)

Cat and mouse

This next example, based on an April 2016 answer by Olivier Faurax to this Quora question, demonstrates the effectiveness of the Relformaide ergative.

While Esperanto uses the accusative to mark objects in sentences, Relformaide only does so with pronouns and leaves nouns unmarked in standard SVO sentences. In certain cases like the one below, the language employs ergative -ieb- for clarity purposes, and always places it after the attached article clitic.

Esperanto Relformaide English
Kato blanco muso manĝas. NekuneO shouliAdj mórínuneS mankat.V
OSV: A mouse eats a white cat. (Adjective describes object)
or A white mouse eats a cat. (Adjective describes subject)
Katon blanka muso manĝas. Nekune shouliebi mórínuniebe mankat. OSV: A white mouse eats a cat.
Katon blankan muso manĝas. Nekune shouli mórínuniebe mankat. OSV: A mouse eats a white cat.
Kato blanka muson manĝas. Nekuniebe shouliebi mórínune mankat. SOV: A white cat eats a mouse.
Kato blankan muson mangâs. Nekuniebe shouli mórínune mankat. SOV: A cat eats a white mouse.

The stage is set

On the surface, the Relformaide coinage belkanthustrika (bel-kant-hustrik-a; beautiful-singing-porcupine-FEM) immediately means "a beautiful porcupine songstress".

  • If she performs songs that are beautiful, then one must place -oz- before the penultimate root, thus resulting in belkanthustrika (beautiful-song-GEN-porcupine-FEM).
  • If she sings beautifully, -ant- substitutes it in belkantanthustrika (beautiful-singing-CONT-porcupine-FEM).
Note Note:
With special thanks to "Kloudmutt" at DeviantArt.com and Inkbunny. This section is inspired by "Ashin", a February 2017 fan work based on Ash the porcupine in Universal/Illumination's Sing (2016).

Not quite Belmont

According to one of my correspondents, who seemed very sure of the point, [Esperanto's] accusative case can only replace je [a stand-in preposition]; [surely] somebody is fibbing somewhere! The confusion between the accusative case and je, which is officially blessed in rule 14 [of Esperanto's grammar], gives rise to a curious ambiguity. A commonly mentioned example of the use of je is veti je chevaloj "to bet on horses", which can also be veti chevalojn. So, since veti monon is correct for "to bet money", veti monon chevalojn is quite reasonably both "to bet money on horses" and "to bet horses on money"!

—Geoff Eddy, "Why Esperanto is not my favourite language" (ca. 1998; June 2002 update)

With its well-stocked inventory of adpositions, Relformaide resolves this problem with the straightforward gealar aup ékuines/gealar dínère ńaup ékuines, or literally "bet (money) towards horses". It can also be written as gealar ékuinaupu/gealar dínère ńékuinaupu, or ékuinäuggealar/ékuinäuddínergealar.

A friend in need...

Another ambiguous English sentence, I helped the boy with a spoon, is an audience-friendly version of an example discussed in this Linguistics Stack Exchange question from August 2015. Did the speaker help just the boy, or did they help him with the utensil? Instead of

Mé perauxilat l'edo ńaseb spounune.

Relformaide clears things up with:

  1. Mé perauxilat l'edo spounzoli. (The boy is holding the spoon; proprietive intrafix.)
  2. Mé perauxilat l'edo spounēmeku. (The speaker is feeding the boy, presumably a toddler; instrumental intrafix.)

An antonymous variation, I helped the boy without a spoon, is immediately translated as Mé perauxilat l'edo ńond une spoune. It can be further interpreted in one of two ways (with -ond as a case suffix), depending on the termison:

  1. Mé perauxilat l'edo spounondi. (The boy, and not the speaker, lacks a spoon. If the speaker is without one, then Mé spounondi perauxilat... is used instead.)
  2. Mé perauxilat l'edo spounondu. (The speaker is doing the job without any spoons.)

All the king's horses...

In the same Stack Exchange question, a demonstration of Czech cases provides several forms of this statement:

Král Uher daroval koně. (The King of Hungary donated horses.)

In both Czech and Relformaide, the context can change depending on the case in question; the focus here is on Uhry/Magyạre/Hungary.

Case Czech Relformaide English
Ergative (SOV) + Dative Král Uhry daroval koňům. Lo Kimibo Magyạre peróbrat ékuinades. The King [of some other country] gave Hungary to the horses.
Genitive Král Uher daroval koně. Kimlo Magyạrozi peróbrat ékuines. The King of Hungary donated horses.
Dative Král Uhrám daroval koně. Kimlo Magyạrade peróbrat ékuines. The King donated horses to Hungary.
Instrumental Král Uhrami daroval koně. Kimlo Magyạrēmeku peróbrat ékuines. The King used Hungary's assistance to donate horses.
Case Relformaide English
Comitative Kimlo Magyạrasébu peróbrat ékuines. The King teamed up with Hungary to donate horses.
Causal Kimlo Magyạrẽrgimu peróbrat ékuines. The King donated horses thanks to Hungary.
Utilitive Kimlo Magyạraurdu peróbrat ékuines. The King donated horses so that Hungary could use them.
Benefactive Kimlo Magyạrómistu peróbrat ékuines. The horses were the King's gift to Hungary.
Antessive Kimlo Magyạrprévu peróbrat ékuines. Before Hungary pitched in, the King donated his share of horses.
Postcursive Kimlo Magyạráprévu peróbrat ékuines. After Hungary pitched in, the King donated his share of horses.
Egressive Kimlo Magyạräppremu peróbrat ékuines. The King donated horses across Central Europe, starting with a supply to Hungary.
Apudessive Kimlo Magyạrproximu peróbrat ékuines. The King donated horses near the Hungarian border.
Inessive Kimlo Magyạrintu peróbrat ékuines. The King donated horses while in Hungary.
Exessive Kimlo Magyạrtúgu peróbrat ékuines. The King donated the horses outside Hungarian territory.
Adessive Kimlo Magyạrondu peróbrat ékuines. With no help from Hungary, the King donated the horses himself.
Exceptive Kimlo Magyạrmoinu peróbrat ékuines. The King donated the horses when Hungary wouldn't.
Inclusive Kimlo Magyạrnendu peróbrat ékuines. The King donated horses, as did Hungary.
Immediate Kimlo Magyạroantemu peróbrat ékuines. The King donated the horses to Hungary first.
Exclusive Kimlo Magyạrseulu peróbrat ékuines. Only Hungary received the herd of horses the King donated.
Aversive/Evitative Kimlo Magyạrévitu peróbrat ékuines. The King donated horses, but couldn't cross the Hungarian border while carrying them.
Contrastive Kimlo Magyạrpótalu peróbrat ékuines. Since Hungary couldn't care less, the King donated the horses himself.
Concessive Kimlo Magyạrmólenu peróbrat ékuines. Hungary or no Hungary, the King went ahead with donating the horses.
Postulative Kimlo Magyạrsifu peróbrat ékuines. The King donated the horses as long as Hungary agreed to help.
Considerative Kimlo Magyạromilu peróbrat ékuines. Officials in Hungary said the King donated horses.

Pretty Little Girls' School

This section's title, taken from section 5.16 of John Woldemar Cowan's The Complete Lojban Language (CLL), has several different interpretations in English. Is it a school where pretty little girls attend; a pretty-looking school for dwarfettes; or a small and beautiful school for girls? While Lojban offers 40 ways to translate it (using melbi cmalu nixli ckule), Relformaide gets by with 17; the tenth variant in the following table carries the most sensible connotation.

The first letters of each word in the phrase, PLGS, form the base of the code in the table; bolded letters signify the focus of the translated variants. As grammatically valid long words, they demonstrate how agglutinative Relformaide is, although most of them are unlikely to see general use. (Simplified alternatives are given in parentheses.)

# Relformaide term(s) English meaning Code
1 tuameduezómistēskoletöbbelide
(= belide der eskolète ńómistoṛzol edas)
the beauty of the small school for girls PLGS
2 tuameduezómistoṛbeŕeskolośaubimide
(= aubimide der beŕeskole ńómistoṛzol edas)
the smallness of the beautiful school for girls PLGS
3 beŕeskoletozedas the pretty little school's girls PLGS
4A tüabbelaubimedüetöteskole
(= belaubimedozas eskole)
the school of the small, beautiful girls PLGS
4B tüabbelaubimeduezómistēskole
(= eskole nómist belaubimedas)
school for girls who are beautiful and small PLGS
5 tuameduezómistēskolöbbelétide
(= belétide der eskole ńómistoṛzol edas)
the slight beauty of the girls' school PLGS
6 eskoletöbbeledas the pretty girls of the little school PLGS
7A tuamaubimeduezöbbeŕeskole
(= aubimedozas beŕeskole)
the little girls' pretty school PLGS
7B tuamaubimeduezómistoṛbeŕeskole
(= beŕeskole ńómist aubimedas)
pretty school for small girls PLGS
8 beŕeskolozaubimedas the pretty school's little girls PLGS
9A tüabbeledüetöteskolète
(= eskolète der beledas)
the pretty girls' little school PLGS
9B tüabbeleduezómistēskolète
(= eskolète ńómist beledas)
small school for pretty girls PLGS
10 tuameduezómistoṛbeŕeskolète
(= beŕeskolète ńómist edas)
small, beautiful school for girls PLGS
11 eskolöbbelaubimedas
(= eskoloze belaubimedas)
the school's pretty little girls PLGS
12 tüabbeleduezómistēskolète
(= eskolète ńómist beledas)
small school for pretty girls PLGS
13 tuameduezöbbeŕeskolète
(= edozas beŕeskolète)
the girls' pretty little school PLGS
14A tüabbelaubimeduezómistēskole
(= eskole ńómist belaubimedas)
school for girls who are beautiful and small PLGS
14B tüabbelaubimedüetöteskole
(= belíenétedozas eskole)
the school of the pretty little girls PLGS

Wind speeds

The augmentative -eg- and its opposite -et- reduce many possible degrees of size to just three. Thus the triplet vento, vent-eto, vent-ego "wind, breeze, gale" replaces the entire Beaufort Scale...

[Also:] Can you use -et-eg-a and -eg-et-a to make finer distinctions of size?

—Geoff Eddy, "Why Esperanto is not my favourite language" (ca. 1998; June 2002 update)

With its size suffixes and -uit among other morphemes, Relformaide can form words corresponding to that scale's various levels as shown below:

BF# Description RFM term Affixes Notes
0 Calm nauŕsúflé naul-
1 Light air súflétuite -et-uit
2 Light breeze súflète -et
3 Gentle breeze súfléteade -et-ead
4 Moderate breeze súflétarde -et-ard
5 Fresh breeze súfleadète -ead-et
6 Strong breeze súfleade -ead
7 High wind, moderate/near gale súfleadarde -ead-ard
8 (Fresh) gale súfleaduite -ead-uit
9 Strong/severe gale súflardète -ard-et
10 Storm/whole gale súflardeade -ard-ead Also kamège
11 Violent storm súflarde -ard Also kamegarde
12 Hurricane force súflarduite -ard-uit Also sîgloane


Words and phrases

Note Note:
This section is based on Mathews (1909), p. 281-284https://archive.org/details/jstor-659469, whose contents have been localised and updated for the Relformaide grammar. Several less audience-friendly terms mentioned in the original article, as well as some obscure ones, have been omitted here.

What follows below adheres to the principle stated by Mathews in his introduction to the word list of Dhudhuroa, a now-extinct Aboriginal language of Australia:

Words of a similar kind are placed under similar headings, in the hope that this classification will be found more convenient for reference than if arranged in alphabetic sequence. It is the equivalence of English words which will be most frequently required and therefore they are placed first.

General vocabulary

Articles

  • le/lo/la, les/los/las – the (def.)
  • une/uno/una – a/an (indf. sg.)
  • unes/unos/unas – some (indf. pl.)

Pronouns

Personal
  • /mo/ma – I
  • mié/mio/mia – me
  • més/mos/mas – we
  • miés/mios/mias – us
  • /to/ta – you (fam. sg. sbj.)
  • tié/tio/tia – you (fam. sg. obj.)
  • tés/tos/tas – you (fam. pl. sg.)
  • tiés/tios/tias – you (fam. pl. obj.)
  • usté/usto/usta – you (frml. sg. sbj.)
  • ustié/ustio/ustia – you (frml. sg. obj.)
  • ustés/ustos/ustas – you (frml. pl. sbj.)
  • ustiés/ustios/ustias – you (frml. pl. obj.)
  • lumo – he
  • lumio – him
  • luma – she
  • lumia – her
  • lumé – it (sbj.)
  • lumié – it (obj.)
  • lumés/lumos/lumas – they
  • lumiés/lumios/lumias – them
Possessive
  • mozé/mozo/moza – my
  • mozés/mozos/mozas – our
  • tozé/tozo/toza – your (fam. sg.)
  • tozés/tozos/tozas – your (fam. pl.)
  • ustozé/ustozo/ustoza – your (frml. sg.)
  • ustozés/ustozos/ustozas – your (frml. pl.)
  • lumozo – his
  • lumoza – her
  • lumozé – its
  • lumozés/lumozos/lumozas – their
Indirect
  • madé/mado/mada – me
  • madés/mados/madas – us
  • tadé/tado/tada – you (fam. sg.)
  • tadés/tados/tadas – you (fam. pl.)
  • ustadé/ustado/ustada – you (frml. sg.)
  • ustadés/ustados/ustadas – you (frml. pl.)
  • lumado – him
  • lumada – her
  • lumadé – it
  • lumadés/lumados/lumadas – them
Relative/Interrogative
  • jière, kière – what (sbj.)
  • jíerie, kíerie – what (obj.)
  • jíeroze, kíeroze – whose (inanim.)
  • jiène/jíeno/jíena, kiène/kíeno/kíena – who
  • jíenie/jíenio/jíenia, kíenie/kíenio/kíenia – whom
  • jíenoze/jíenozo/jíenoza, kíenoze/kíenozo/kíenoza – whose (anim.)
Demonstrative
  • esine/esino/esina – this
  • esane/esano/esana – that
  • esaune/esauno/esauna – yon

Verbs

  • bar – to be, to exist
  • var – to go, to travel
  • aigar – to do, to act
  • inzar – to cause
  • orjar – to force (upon an entity)
  • tenar – to have, to hold
  • zolar – to own, to possess
  • óbrar – to give
  • poadar – to offer, to propose
  • dostar – to get, to obtain, to acquire, to fetch
  • borgar – to borrow
  • borginzar – to lend
  • véyar – to see, to watch, to look (at)
  • audar – to hear, to listen
  • pomar – to smell
  • toagar – to touch (active/SVO), to feel (passive/OVS)
  • gustar – to taste
  • dezar – to say
  • addezar – to tell (someone)
  • parlar – to speak
  • krívar – to cry, to shout
  • mólitar – to pray
  • námúzar – to honour, to praise
  • kantar – to sing
  • vieslar – to whistle
  • jautar – to ask
  • verjautar – to answer, to reply, to respond
  • prósitar – to beg
  • lezar – to read
  • skríbar – to write
  • geplar – to type
  • hendar – to walk
  • vuelar – to run
  • danzar – to dance
  • sautar – to jump
  • voalar – to fly
  • sliedar – to slide
  • sliepar – to slip
  • slievar – to slither
  • plaunjar – to dive
  • noagelar – to swim
  • streavar – to climb
  • äulvvar – to rise, to go up, to ascend
  • toambar – to fall
  • úlemvar – to descend, to go down, to sink
  • rester – to stay
  • shimar – to put upon a surface, to place something on
  • prendar – to take
  • poartar – to carry, to bring
  • bemar – to buy
  • proadar – to sell
  • mitar – to send
  • aimar – to love, to adore
  • ódímar – to hate, to detest
  • aubogar – to support, to advocate
  • oponar – to oppose
  • vevar – to live, to be alive
  • moartar – to die
  • nendar – to add
  • moinar – to subtract, to take out, to reduce, to decrase, to mitigate
  • puemar – to divide, to split
  • razar – to multiply, to copy, to duplicate, to imitate
  • ótrinzar – to change, to alter
  • erivar – to erase, to delete, to wipe out, to trash
  • premar – to start, to begin, to commence
  • pauzar – to pause, to stop
  • térúzar – to continue, to resume
  • fímar – to finish, to end, to conclude
  • aubelar – to cease ([doing] an action) completely
  • aisómar – to accomplish, to fulfill
  • aistar – to make, to produce
  • teremar – to yield
  • lanzar – to throw
  • doalar – to hurt, to harm, to injure, to damage, to ruin
  • verdoalar – to fix, to repair
  • yẽrtar – to hit, to beat, to slap, to strike
  • taugar – to break
  • gontar – to chop
  • hildar – to fight
  • kuibar – to steal
  • stauzar – to hunt, to chase, to pursue, to follow
  • paufar – to shoot (at)
  • moarzar – to kill, to murder
  • moarjar – to execute
  • progónar – to prosecute, to discriminate
  • kastigar – to punish, to scold, to chastise
  • ódígar – to lead, to guide, to drive (transportation)
  • nímelar – to control
  • gómar – to hide, to conceal
  • vergómar – to reveal, to expose
  • podálar – to stand (up)
  • duekar – to sit
  • doarmar – to sleep
  • hoalmar – to dream
  • velmar – to strive, to aspire, to aim for
  • lorgar – to look for, to search for, to seek
  • trouvar – to find
  • sauvar – to save, to rescue, to deliver from
  • pogar – to try
  • prouvar – to prove, to verify
  • holómar – to succeed
  • haulemar – to fail
  • joalinzar – to improve, to ameliorate
  • joalausmar – to improve (by oneself), to get/become better
  • málinzar – to worsen (something), to deteriorate (something), to botch up
  • málausmar – to worsen (by oneself), to deteriorate (by oneself), to get/become worse
  • sabar – to know
  • penzar – to think
  • chembar – to forget
  • verchembar – to remember
  • spírar – to breathe
  • súflar – to blow
  • flérar – to cry/shed tears, to weep, to sob
  • suigar – to suck (up)
  • soustar – to bite
  • cholgar – to spit
  • puxelar – to vomit, to throw up
  • soamar – to make love (to), to copulate (with), to mate (with)
  • tuelar – to give birth, to bear children
  • puadar – to hug, to embrace
  • suavar – to kiss
  • lúzar – to shine
  • ziginar – to shake
  • wulobar – to shiver

Adjectives

  • joali – good
  • mali – bad, evil
  • froli – happy, joyful
  • nïffroli – sad, unhappy
  • beni – well
  • osivi – sick
  • broigi – angry
  • krúli – cruel
  • houtanti – greedy, excessive
  • grandi – big, large, huge
  • aubimi – small
  • mîkróli – tiny
  • aulti – high, up
  • úlemi – low, down
  • proximi – near, short
  • proxami – far, long
  • proxaumi – beyond
  • podáli – tall (in height)
  • touti – full, complete, whole
  • aibemi – equal
  • oidi, ausiki – similar
  • ótri - different
  • ondi – empty, barren
  • nidhini, nidhani – absent
  • jouti – sudden, unexpected
  • vasiti – deliberate, intentional
  • tasivi – accidental
  • tezi – fast, quick
  • vertezi – slow
  • foarti – strong
  • troafi – weak
  • naulvéyi – blind
  • naulaudi – deaf
  • nauŕhendi – lame
  • vasimi – tired, weary, sleepy
  • hugeli – brave, courageous
  • foabeuzi – afraid, scared, fearful
  • drani – right, correct
  • nïddrani – wrong, incorrect
  • voardi – true, genuine
  • seudi – false, fake
  • meavi – fat, thick, dense
  • nidmeavi – lean, thin, meagre, sparse
  • jomi – pregnant, with child
  • holómi – successful
  • haulemi – unsuccessful, failed
  • velmi – ambitious, aspiring, zealous
  • húviti – interested, eager
  • váli – worthy
  • válondi – worthless, good-for-nothing

Adverbs

  • hinu – here
  • hanu – there
  • haunu – yonder, over there
  • mesmu – even, especially

Colours

  • roube – red
  • blouve – blue
  • flauvé – yellow
  • rógèle – green
  • roubblouve – purple
  • ouranjé – orange
  • roubeusmé – pink
  • broute – brown
  • shoule – white
  • noule – black

Family terms

  • aumbré – human
  • aumbro – man
  • aumbra – woman
  • èbe – baby, newborn
  • ebo/eba – baby boy/girl
  • ède – youngster, kid, tyke, lad
  • edo – boy
  • eda – girl
  • eshoane – teenager, adolescent
  • eshoano/eshoana – teenage boy/girl
  • eushèle – adult, grownup
  • eushelo/eushela – man/woman (of any species)
  • elde – old person, elder, senior
  • eldo/elda – old man/old lady
  • gauné – relative
  • gauntruze – family
  • fraule – spouse
  • fraulo – husband
  • fraula – wife
  • ríante – parent
  • ríanto – father, dad
  • ríanta – mother, mom
  • tepríante – grandparent
  • tepríanto – grandfather, granddad
  • tepríanta – grandmother, grandma
  • ódré – child, offspring, progeny
  • ódro – son
  • ódra - daughter
  • siblé – sibling
  • siblo – brother
  • sibla – sister
  • siblódré – nibling
  • siblódro – nephew
  • siblódra – niece
  • ríablé – parent's sibling
  • ríablo/ríabla – aunt/uncle
  • ríablódré – cousin
  • ríablódro/ríablódra – male/female cousin
  • fraulríante – parent-in-law
  • fraulríanto – father-in-law
  • fraulríanta – mother-in-law
  • frauŕsiblé – sibling-in-law
  • frauŕsiblo – brother-in-law
  • frauŕsibla – sister-in-law
  • amite/amito/amita – friend
  • aimamite – significant other
  • aimamito – boyfriend
  • aimamita – girlfriend

Body parts

  • tũrgene – body
  • yovèle – bone
  • górane – blood
  • nũrite - nerve
  • mausèle – muscle
  • mũrite – organ
  • pièle – skin
  • kepale – head
  • élogkepale – forehead
  • kromele – beard
  • vorite – eye
  • porite – nose
  • lúvane – neck
  • trakèle – throat
  • dorite – ear
  • morite – mouth
  • sorite – tooth
  • gorite – tongue
  • suorite – lip
  • foaraxe – chest
  • mame – bosom, mammae
  • mindrạ – navel
  • staumake – stomach, belly, tummy
  • jurèke – heart
  • peumaune – lungs
  • grúfane – liver
  • vúdele – back
  • voaleande – wing (of a bird or bat)
  • jásène – arm, limb
  • gúvite – elbow
  • spạdane – shoulder
  • mène – hand
  • troade – leg
  • aulttroade – thigh
  • gaskote – calf
  • jenoule – knee
  • chevile – ankle
  • kuode – foot
  • hiele – heel
  • jírite – digit
  • menjírite – finger
  • kuodjírite – toe
  • trèpe – tail (of an animal)
  • muge – bottom
  • kale – excrement

Natural world

Entities

  • Tairoke – nature, ecosystem
  • Taire – Earth
  • Solé – the Sune
  • Loune – the Moon
  • zièle – sky
  • laufe – air
  • laufarde – atmosphere
  • astrèle – star
  • hîdrole – water
  • brúlière – fire
  • fumé – smoke
  • holège – shadow
  • húmine – ground, soil
  • toapé – stone
  • saiblé – sand
  • brullesmé – lava

Landforms

  • poaldère – polder
  • róvine – plain
  • aultróvine – plateau
  • lófé – hill
  • lófarde – mountain
  • vaulkène – volcano
  • fálaife – cliff

Weather

  • framzère – weather
  • brúli – hot
  • chouli – cold
  • waurmi – warm
  • súflière – wind
  • kamege – storm
  • trúmèle – thunder
  • trúmelluze – lightning
  • sîgloane – tropical cyclone, hurricane, typhoon)
  • zíelvóraune – tornado, twister
  • pleuve – rain
  • pleuvète – shower
  • tuarène – fog
  • nívine – snow
  • siklète – frost
  • hoagèle – hail
  • onikkũrve – rainbow

Living things

Animals
  • animale – animal
  • tuere – creature
  • tuerède – cub, whelp (of mammals); chick (of birds)
Mammals
  • mamiftuere – mammal
  • nâye – dog
  • loupe – wolf
  • vaulpe – fox
  • zẽrde – fennec
  • dingone – dingo
  • nèke – cat
  • simbale – lion
  • tigère – tiger
  • dumale – cheetah
  • balame – jaguar
  • youlane – genet
  • ródente – rodent
  • mórine – mouse
  • kualike – rat
  • krisèle – hamster
  • poargé – guinea pig
  • eskũre – squirrel
  • voaleskũre – flying squirrel
  • jidale – chipmunk
  • sínome – prairie dog
  • aurvíkole – water vole
  • herisone – hedgehog
  • hustrike – porcupine
  • sorèxe – shrew
  • mefite – skunk
  • bausone – badger
  • gúlove – wolverine
  • dúbime – bear
  • chusèle – panda
  • lapine – rabbit
  • gulèpe – hare
  • síbane – raccoon
  • doabrane – otter
  • mostèle – weasel
  • miegale – ferret
  • boavé – cattle
  • boavo – bull
  • boava – cow
  • búfale – buffalo
  • kabrite – goat
  • óvile – sheep
  • puerché – pig, swine
  • chíroté – bat
  • hafile – elephant
  • protévone – primate
  • simiane – monkey
  • pífike – ape
  • chimpanzé – chimpanzee
  • orangutane – orangutan
  • kangaroune – kangaroo
  • waularoune – wallaroo
  • waulabine – wallaby
  • mákrote – bilby
  • kuokale – quokka
  • kóale – koala
  • maníkoule – North American opossum
  • díluge – Australian possum
  • yagule – bandicoot
  • daurumbé – kangaroo rat
Birds
  • noave – bird
  • poale – chicken
  • koulombe – dove
  • píjaune – pigeon
  • híraunde – swallow
  • jaurike – pelican
  • signé – swan
  • koarvé – crow
  • soroké – magpie
  • aguile – eagle
  • kũruane – crane
  • blouvnoave – bluebird
  • noulnoave – blackbird
  • lipsnèle – robin
  • ordèke – duck
  • nẽrlèke – goose
  • múrove – ostrich
  • gourane – curlew
  • míoure – emu
  • choike – rhea
  • kiwine – kiwi
  • kasúaire – cassowary
  • kaukatouve – cockatoo
  • suarive – parrot
  • baujerigé – budgerigar
Fishes
  • vueme – fish
  • cháfoine – sardine
  • pajaule – shark
  • jóroune – tuna
  • penfrèle – cod
Reptiles
  • reptile – reptile, scalie
  • nakauze – snake
  • lusẽrte – lizard
  • gekone – gecko
  • lieguané – iguana
Invertebrates
  • insekte – insect
  • pagite – larva
  • mouché – housefly
  • chitime – mosquito
  • yúkane – louse
  • chápoale – grasshopper
  • maŕchápoale – locust
  • sënkkuode – centipede
  • kílökkuode – millipede
  • pilipale – butterfly
  • pilipagite – caterpillar
  • eskạrgé – snail
  • lesmé – slug
Plants
  • plante – plant
  • poazène – grass
  • flũre – flower
  • tovale – weed
  • yéronbe – currant
  • wũrite – peppermint
Trees
  • aibũre - tree
  • drevabode – wooded area
  • foareste – forest
  • fóline – tree leaf
  • paume – palm tree
  • tusame – pine
  • chesnèle – oak

Food and drink

  • mongaurde – food and drink
  • mankar – to eat
  • mankaurde – food
  • bevar – to drink
  • bevaurde – drink, beverage
  • chúzar – to cook
  • chúzaine – meal
  • oajẽrde – juice
  • soadạ – soda, soft drink
  • leché – milk
  • húyène – egg
  • fáloave – bread
  • wistèle – pancake
  • esome – fruit
  • drúvène – grape
  • drauvène – raisin
  • sakũre – cherry
  • boabule – berry
  • baunane – banana
  • aniane – pineapple
  • tapaule – apple
  • fazole – bean
  • zelène – vegetable
  • bresike – cabbage
  • mekuane – pumpkin
  • dumèle – honey
  • korzène – spice
  • gustinzeande – seasoning
  • hẽrbe – herb
  • saulte – salt
  • pímente – pepper

Time

  • maurve – time
  • kloché – clock
  • menkloché – watch
  • klochlezar – to tell time
  • sekite – second
  • minite – minute
  • haurite – hour
  • dième – day
  • froditime – morning
  • mẽrkdième – noon, midday
  • sẽrtdième – afternoon
  • fronochime – evening
  • nochème – night
  • mẽrknochème – midnight
  • sẽrtnochème – late night
  • houtnochème – overnight
  • tídène – week
  • sefime – weekend
  • masine – month
  • anove – year
  • deyānove – decade
  • sentanove – century
  • kíloganove – millennium
  • maurvarde – era, epoch
  • nodemu – today
  • nouchemu – tonight
  • vüëiddíemu – yesterday
  • suiɖdíemu – tomorrow

Transportation

  • poartière – vehicle
  • houbène – highway
  • jalane – road, path, way
  • strouze – street
  • lójale – railway
  • móbile – car, automobile
  • bauze – bus
  • kamíaune – truck, lorry
  • ségaune – wagon
  • voaŕineale – airplane
  • lójile – railroad car
  • plova – boat
  • plovarda – ship
  • kanova – canoe

Weapons

  • fúsile – gun
  • spafạ – sword
  • tuekière – spear
  • nábute – club
  • boumerange – boomerang

Given names

Male

  • Adamo – Adam
  • Jauno – John
  • Wilemo – William
  • Máfeulo – Matthew
  • Mărko – Mark
  • Lúko – Luke
  • Jaimzo – James
  • Jâkaubo – Jacob
  • Éduardo – Edward
  • Róbairto – Robert
  • Bobino – Bobby, Bob

Female

  • Anya – Anne, Anna
  • Eliezabefa – Elizabeth
  • Emiliya – Emily
  • Kaita – Kate
  • Mariya – Mary
  • Judiya – Judy
  • Samanfa – Samantha
  • Sofiya – Sophia, Sophie

Place names

Continents

  • Áfrika – Africa
  • Aiźa – Asia
  • Austrailída – Australia
  • Ámerika – America
    • NoardÁmerika – North America
    • SudÁmerika – South America
  • Íroape – Europe
  • Ăntărtika – Antarctica

Oceans

  • Ătlăntike – the Atlantic
  • Pásifike – the Pacific
  • Indíloze – the Indian Ocean

Regions

  • Karibine – the Caribbean
  • Ărtike – the Arctic

Countries

Americas
  • Oanizaidis Sivittimes – United States
  • Kánaida – Canada
  • Mexiko – Mexico
  • Kaustarika – Costa Rica
  • Kŭba – Cuba
  • Jămăika – Jamaica
  • Dauminíka – Dominica
  • Bărbăidos – Barbados
  • Trinidade nend Tobaigo – Trinidad and Tobago
  • Brasile – Brazil
  • Aujenabode – Argentina
  • Chilé – Chile
Europe
  • Oanizaidi Kimúzème – United Kingdom, Great Britain
  • Gálige – Ireland
  • Franze – France
  • Maunako – Monaco
  • Espane – Spain
  • Poartugale – Portugal
  • Alémane – Germany
  • Helvẹte – Switzerland
  • Magyạre – Hungary
  • Ítalida – Italy
  • Chèke – Czech Republic
  • Slovakida – Slovakia
  • Slovênida – Slovenia
  • Kroâzida – Croatia
  • Dăinmărke – Denmark
  • Suidène – Sweden
  • Suomé – Finland
  • Noarwâye – Norway
  • Îslande – Iceland
  • Rauźa – Russia
  • Kazakhé – Kazakhstan
Asia
  • Tũrkiede – Turkey
  • Isráyèle – Israel
  • SaudozẠraibida – Saudi Arabia
  • Írake – Iraq
  • Írane – Iran
  • Indíla – India
  • Pakistane – Pakistan
  • Chîna – China
  • Korieva – Korea
    • NoardKorieva – North Korea
    • SudKorieva – South Korea
  • Japaune – Japan
  • Filipines – Philippines
  • Mălăive – Malaysia
  • Singapoare – Singapore
  • Indonieza – Indonesia
Africa
  • Êjipte – Egypt
  • Soudane – Sudan
  • Aljerida – Algeria
  • Morauko – Morocco
  • Senegale – Senegal
  • Găna – Ghana
  • Nîjairida – Nigeria
  • Nîjaire – Niger
  • Kameroune – Cameroon
  • NoardKaungo – Congo-Brazzaville
  • SudKaungo – Democratic Republic of the Congo
  • Kenya – Kenya
  • Tanzanida – Tanzania
  • Angola – Angola
  • Bautsuana – Botswana
  • SudÁfrika – South Africa
  • Malgashèle – Madagascar
  • Mauriezabode – Mauritius
Oceania
  • Austrailída – Australia
  • NouƒZêlande – New Zealand
  • Fíjine – Fiji
  • ÁmerikośSamoava – American Samoa
  • WestoṛSamoava – Samoa

Languages

  • Ingile – English
  • Gálige – Irish
  • Franze – French
  • Espane – Spanish
  • Poartugale – Portuguese
  • Alémane – German
  • Ítalé – Italian
  • Rauźé – Russian
  • Magyạre – Hungarian
  • Súomé – Finnish
  • Tũrkiede – Turkish
  • Ạraibe – Arabic
  • Hiebrouve – Hebrew
  • Hindine – Hindi
  • Mandạrine – Mandarin Chinese
  • Japaune – Japanese
  • Indonieze – Indonesian
  • Suahiline – Swahili
  • Malgashèle – Malagasy
  • Guaranive – Guarani
  • Esperlinge – Esperanto
  • Ĭdo – Ido
  • Laujiklinge – Lojban

Expressions

  • Sálout – Hello/Hi (there)
  • Salout, maundté!Hello, world!
  • Benvenat – Welcome
  • Joaldíem – Good day
  • Joalfroditim – Good morning
  • Joalsẽrtdíem – Good afternoon
  • Joalfronochim – Good evening
  • Joalnochem – Good night
  • Joaltüvv – Bye/Goodbye
  • Joaleriv – Good riddance
  • Sẽrtat – Later
  • Benat-té? – How are you (doing)?
  • Mé benat. – (I'm) fine.
  • Joalausikat. – Sounds good.
  • Kière tośnaumat? – What is your name?
  • Mozé naume bat... – My name is...
  • Kière ńaigat-to? – What are you doing?
  • Kabodu ńäuvvat-te? – Where are you going?
  • Prósim – Please
  • Dolchat – Thank you
  • Guefóraid – You're welcome
  • Mozé kroagineale ńanguileuzat.My hovercraft is full of eels.


Sample texts

Dezaidètes/Quotations

Mé penzat, ẽrgim mé bat.
Renâyo Descartes

I think, therefore I am.

—Réné Descartes

Seulēsenile més boulfoabat, séblifoabidat.
Franklino Delano Roosevelt

The only thing we have to fear, is fear itself.

—Franklin Delano Roosevelt

Jautait nal jíerie ńustozés úzème ńaigábilat ómist ustés...sed jíerie ńustés aigábilat ómist ustozés úzème.
Jauno F. Kennedy

Ask not what your country can do for you...but what you can do for your country.

—John F. Kennedy

Naumdezait madoń Ishmáyelo.
Hẽrmano Melville, Moby-Dick

Call me Ishmael.

—Herman Melville, Moby-Dick

Premwobu, Díevo pébinzat Zíelle nend Tairlé.
Jénésieze 1:1

In the beginning, God created the Heaven and the Earth.

—Genesis 1:1

Póvaud Díevo peraimardat Tairlé, ńẽrgim Lumo peróbrat selbemade Lumozo seul'adríantaid'Ódro, jían kíenoipe ńäukkróyat Selbtouvio naŕfémoartat, sed fétenat fímönvvévide.
Jauno 3:16

For God so loved the world, that He gave His only begotten Son, that whosoever believeth in Him should not perish, but have everlasting life.

—John 3:16

Saulme/Psalm 23

Une Saulme ńesil Dăivĭdo.
1 Dómeno moz'Óvilhĩrtíenat; mé naŕferondat.
2 Lumo broudinzat mié ńíak rógelis pastũres: Lum'ódigat mié korttranz mouvondis hîdrollés.
3 Lumo rënnoutinzat mozé zelme: Lum'ódigat mié tranz jalaniles dráneuzis ómist Lumozo naume.
4 Sid, mólen mé tranśhendat lúgenile der moarthóleglé, mé naŕféfoabat malé: póvaud T'asebat mié; Tozo proute ńaseb auxilière ńättrúlat mié.
5 To taglat tévolune mélogaupu, proxim mozés ódimes: To frótat mozé tuede ńemek óleave; mozé taséninte tüfflúyardat.
6 Shũru joalide ńaseb nérelide ferasebat mié wob toute díemlés der mozé veuxème: nend mé feraulmadat l'obène der Dómeno fímondu.
Aimène.

Máfeulo/Matthew 6:9-13

9 ...Mozés Ríantto, ńinterijat Zièle,
Saintinzaidat Tozo Naume.
10 Tozo Kimúzème venavat.
Tozo voule ńaigaidat Tairadu,
ńausik selbé haizat Zíelintu.
11 Óbrait miés nodemadu mozés Díemozi Fáloave;
12 nend nérelait miés mozés malaiges,
wob més nérelat esanes
málaigerijat aup més.
13 Nend ódigait miés nal aup málaigaije,
sed sauvait miés malévitu.
Póvaud Tozato Kimúzemile,
Pótestlé nend Gloarïllé,
fímondu nend toutwobu.
Aimène.

Oantemi Time/Article 1

Tout'aumbribiènes tuelaidat povelu nend aibemu ńint dignité nend lugés. Lumés abuamat póvaude nend móralsábantide, nend shoaraupaigaik int une kaisile siblidi.


Notes and references

  1. The past participle noun form, and not the past tense form.
  2. 2.0 2.1 A morpheme (or voabtime in Relformaide) is the smallest unit of language; the study of morphemes is called morphology (voabtánule).
  3. Obsolete in English since the 15th century, the word termison (or fimättime in Relformaide) has been adopted for use in this language's documentation. Derived from the French terminaision, or "termination", it is listed in the Oxford English Dictionary's Second Edition (1989), as well as the University of Michigan's Middle English Dictionary.
  4. Unless they are anthropomorphised, in which case the -o/-a endings apply. -a is also used for words pertaining to water transportation.
  5. Þ is otherwise represented by Ƥ/ƥ, which does not display on some browsers or operating systems.
  6. A special variant, -uez, precedes roots in complex case-oriented inflections.
  7. Comprising 4 roots, 6 affixes, and 0 affixoid; 3 of them are case markers.
  8. Equivalent to English SBJ has PT-VB.
  9. Equivalent to English ... (is) going to/(is) about to VB; followed by a compound root, or gender/verbal termison.
  10. Equivalent to English ...that SBJ VB, or could/would/should.
  11. This pattern also occurs in Pitjantjatjara, a dialect of the Aboriginal Western Desert language of Australia.
  12. Several neuter nouns in Relformaide inherit the last -o/-a of their original etymons; the resulting termison is underdotted.
  13. Taking a cue from Ido (an Esperanto spinoff), some Relformaide place names inherit the last -o/-a/-e of their original etymons; otherwise, -e follows the root in question.
  14. See also section 4.1 in "Proposed Guidelines for the Design of an Optimal International Auxiliary Language" by Richard K. Harrison, 2001 (9th draft):
    "The language should be designed so that all compounds can be unambiguously divided into their constituent morphemes, and so that no single morpheme can be mistaken for a combination of several morphemes. Confusion can occur if such auto-analysis is not designed into the language; for example, the Dutch word kwartslagen can mean 'quarter beats' (kwart + slagen) or 'quartz layers' (kwarts + lagen); the Esperanto word sukero might mean 'sugar' (suker + o) or 'a drop of juice' (suk + er + o). While it is true that context can usually indicate which meaning is intended, there is no reason for such morphemic ambiguity to exist in an optimal language design."
  15. Only found after avar/avat &c. (have/has) in analytic constructs.
  16. 16.0 16.1 Ède and édide are among Relformaide's very few palindromic words; a handful of palindromic roots also exist.
  17. In Relformaide, "knowledge" translates to sabantide or sabaide (depending on the context).
  18. Used in certain sentences with two or more subjects/objects.
  19. See Haspelmath (2006), p. 9https://www.scribd.com/document/118102864/Case-grammar.
  20. Cf. French chez.
  21. Several directional suffixoids substitute the generic túdel when necessary:
    • -eask (left) — obeneaskad/obenöuteaskad (at the left side of the house)
    • -zusk (right) — obenzuskad/obenöunzuskad (at the right side of the house)
    • -élog (front) — obenélogad/obenouzélogad (at the front side of the house)
    • -vúdel (back) — obenvúdelad/obenöuvvúdelad (at the back side of the house)
    • -noard (north) — obennoardad/obenouśnoardad (at the north side of the house)
    • -sud (south) — obensúdad/obenouśsúdad (at the south side of the house)
    • -east (east) — obeneastad/obenöuteastad (at the east side of the house)
    • -west (left) — obenwestad/obenöutwestad (at the west side of the house)
  22. This differs from -ausom, the superlative comparison suffix.
  23. Plũrim- can be substituted by any of the other number roots, or compounds thereof: obentrïttuig (three of the houses), obendeytuig (ten of the houses), obenoatdeiśkauttuig (84 of the houses).
  24. This differs from -ausm, the comparative comparison suffix.
  25. 25.0 25.1 25.2 As Justin B. Rye jokes in "Onomastics", "If Esperanto participles were fully regular, the word esperanto would mean a temporary outbreak of hope, and Zamenhof's pen‐name would have been la Doktoro Esperantulo!"
  26. As Jespersen also noted,
    "Ido has here the verb mariajar 'to marry' with the derivatives mariajo or mariajeso 'marriage,' mariajatulo 'married man,' mariajatino 'married woman,' with the variants mariajitulo, mariajitino; mariajo-festo 'wedding'; further, the independent words spozo, spozulo, spozino for husband and wife. - Occ has maritagie, marito, marita, for 'married couple' maritates, for wedding (eheschliessung) maritagie, maritantie, and for 'married state' (ehestand) matrimonie. Novial thus gets off cheaper than either of these languages."
  27. Also referred to as Haulandlé (Holland).
  28. As Mark Rosenfelder notes in this Quechua primer, "'You' and 'they' are regular plurals, formed by adding the plural suffix to the singular pronouns. That's a regularity that didn't occur to the inventor of Esperanto!"
  29. In spite of medieval European legend, no female Pope has ever headed the Catholic Church.
  30. Only used in adverbial form, as in Trouƒfímavantu, lumo pétüvvat l'oveale. (Having finished his work, he left the office.) This is also known as the perfect participle.
  31. 31.0 31.1 *Sujaidar is ungrammatical in Relformaide.
  32. 32.0 32.1 *Sujaidantar is ungrammatical in Relformaide.
  33. 33.0 33.1 *Sujaidavar is ungrammatical in Relformaide.
  34. von Möllendorff (1892), p. 10https://books.google.com/books?id=KgkQAAAAYAAJ
  35. "Lesson VII" in McGuffey (1896), p. 13https://archive.org/stream/mcguffeysfirstec00mcgu_1
  36. "Lesson II" in McGuffey (1896), p. 8https://archive.org/stream/mcguffeysfirstec00mcgu_1