Conlang
Advertisement

Rhodian is a Romance language spoken on and around the island of Rhodes in the southeastern Aegean Sea. It is heavily influenced by the Rhodian Doric variety of Greek spoken on the island before the Roman occupation, as well as French and Italian from the Knights Hospitaller who ruled the island until a communist revolution in 1924. Thenceforth, it incorporated many Russian loanwords, until the communist government was overthrown in favour of a more socially democratic one in 1987. For more history, click here. (A/N: this doc is yet unfinished, I have about 50 years left to iron out)

In the modern day, Rhodian is spoken by around 180,000 people worldwide. Of this, some 60,000 are in diaspora, mainly in Germany, Italy, the UK and the USA. There used to be a large Rhodian minority in Russia, but since the 1980s it has largely disappeared. The remaining population live in the traditional language area; around 102,000 of those in Rhodes and the surrounding islands that make up the Republic of Rhodes, just under 20,000 in the coastal regions of southwest Turkey (Caria in Rhodian), and 1,000 in Greece - mainly in some villages on the Greek side of Karpathos (Cárpatu), and on Kasos (Casu).

Rhodian
« Rodia »
Type
Fusional
Alignment
Accusative
Head direction
Head-initial (flexible)
Tonal
No
Declensions
Yes
Conjugations
Yes
Genders
Yes
Nouns decline according to...
Case Number
Definiteness Gender
Verbs conjugate according to...
Voice Mood
Person Number
Tense Aspect


Romance Languages Map

A map of European Romance languages, with Rhodian highlighted in bold

Classification and Dialects[]

Rhodian is typically not classified as a Romance Language, as it derives from a variant of Classical Latin with a heavy Greek substrate. Thus, it is typically labelled as a Para-Romance isolate. The true Romance language with which Rhodian bears the most similarity is said to be either Sardinian or Vlach/Romanian, though this is largely an aesthetic similarity, as the grammar of Rhodian is highly divergent, both in its vocabulary and morphology. It retained cases from Latin, of which there are between 5 and 7 depending on whom you ask and how you count, and has a much more robust morphological tense system than other Latinate languages - including the simple aorist, a convention borrowed from Greek.

Rhodian within the Republic of Rhodes has lost much of its dialectal variation since the communist era, during which time the use of Standard Rhodian was enforced to create a sense of national cohesion. The Standard Rhodian language was based on the dialects of Lindos (Lindu) and the central highlands of Rhodes, as these were deemed to be the most 'pure', though much influence from the vernacular language of the capital, Vastu Rodie, seeped in. The dialects that exist on the island of Rhodes nowadays are primarily distinguished by accent, with the dialects of the outer islands varying a little more.

Rhodian-Dialects-3

A map of the various extant dialects of Rhodian

Phonology and Writing System[]

Rhodian is written with the Latin alphabet, although it has been written in the Greek and Cyrillic scripts in the past. Letters can stand for many different sounds due to rampant lenition that took place in the language.

On top of the 24 base letters, there are three diacritics used in Rhodian.

The acute accent is used on all vowel letters (á é í ó ú ý) and marks irregular stress. Many words are distinguished by stress, such as ancon (javelin) and ancón (elbow). Stress is usually on the penult in Rhodian, and while lenition and reduction can often blur the lines between syllables in speech, in writing it's quite transparent.

The grave accent is used on most vowel letters (à è ì ù ỳ) in two ways. First, it's used to distinguish between certain words that are spelt the same, a convention adopted from French. Contrast words like trans (across) with tràns (interesting). The second and much more common purpose is to de-emphasise some vowels. For example, the word piscìs (fish) is pronounced /pɪʃ/, but would be pronounced /ˈpɪ.ʃɪs/ without the diacritic.

The final diacritic and arguably the most unique in its function is the circumflex, used again on all vowel letters (â ê î ô û ŷ). It has two functions also, the first being to separate vowels in different syllables that may otherwise be construed as diphthongs, such as the word eâr (spring - /ˈɛ.aɾ/). This function hearkens back to the origin of this diacritic, the ancient Greek trema, which was written more like a diaeresis but on Rhodes was simplified to require only one stroke to write - a convention which some printhouses still neglect. The second more important use of the circumflex in Rhodian is to transform 'soft vowels' into 'hard vowels'. Some Rhodian consonants lenite in the presence of the vowels <e>, <i> and <y>, dubbed 'soft vowels', with the other vowels being by extension named 'hard vowels'. Occasionally, a consonant is not lenited before a soft vowel - usually due to etymological shenanigans - and this is marked by putting a circumflex over the vowel. Compare the words ce (to, up to, against - /ʃɛ/) and (also, and - /kɛ/). It is worth noting, however, that the circumflex does not affect the lenition of some consonants, notably <g> and <l>.

There is one more marginal diacritic, which is a combination of the circumflex and acute used in certain situations when words' declensions with soft vowels would alter the pronunciation of the preceding consonant sound. For example, the word for 'lifestyle' is psucá, and in the Genitive case that would be psucé, but this would change the consonant from /k/ to /ʃ/, which is not correct as this particular <c> comes from an aspirated Greek chi (χ) and does not undergo softening. Thus, the circumflex and the acute are stacked together and the word comes out to psucế. When the circumflex is (uncommonly but not negligibly commonly) written instead as a diaeresis, it does not manifest as a Hungarumlaut but rather as a stacked diaeresis and acute, i.e psucë́.

Letter Aa Bb Cc Çç Dd Ee Ff Gg Hh Ii Ll Mm
Sound /a/ /b/ ~ /β/ /w/ 1 /k/ /x/ /ʃ/ /ç/ 2 /ʃ/ /d/

/ʒ/ /ð/ ~ /θ/3

/ɛ/ /j/4 /f/ /ɣ/ /j/ /w/ 5 /h/ /ɪ/

/j/6

/l/ /j/ /w/ /ɮ/ 7 /m/
Letter Nn Oo Pp Qq Rr Ss Tt Uu Vv Xx Yy Zz
Sound /n/ /ŋ/ /ɔ/ /o/ 8 /p/ /kʷ/ /k/ 9 /ɾ/ 10 /s/ /t/ /θ/ /ʃ/11 /u/ /ʊ/

/w/12

/v/ /ks/ /s/ 13 /i/ /z/
  1. <b> is pronounced as /b/ ~ /β/ word-initially only (except before <u>), elsewhere it is /w/.
  2. <c> is pronounced /k/ before hard vowels and /ʃ/ before soft vowels. It is often realised as /x/ in coda position after hard vowels but /ç/ after soft vowels, particularly /i/ and /ɪ/, as well as the glide /j/. These non-sibilant fricative realisations are most common in rapid and/or informal speech, and occur most often intervocalically, such as in the above example psucá /p͡su.ˈxa/.
  3. <d> is pronounced /ð/ intervocalically and syllable-finally (occasionally devoiced to /θ/ word-finally) and /d/ ~ /d̪/ elsewhere. It is also pronounced /ʒ/ (see below)
  4. <e> is pronounced /j/ in the sequences <ea> and <eo>, with the exception of the first-person singular nominative pronoun, in which it is pronounced /ʒ/.
  5. <g> is pronounced /ɣ/ around hard vowels except <u>, around which it is pronounced /w/. It is pronounced /j/ around soft vowels.
  6. <i> is pronounced /ɪ/, except from in the verb ending <ire>, where it is /i/. It is also pronounced as a glide /j/ when followed by another vowel, unless it softens a preceding consonant in which case it becomes silent. For example, acia /ˈa.ʃa/.
  7. <l> is pronounced /l/ around hard vowels, /j/ around soft vowels, /w/ before fricatives and /ɮ/ before stops (both assuming it is not already lenited by a soft vowel).
  8. <o> is pronounced /ɔ/ or /ä/ in stressed syllables and /o/ in unstressed syllables.
  9. <q> is only found in the digraph <qu>, where it is usually pronounced /kʷ/ before hard vowels (except <u>) and /k/ elsewhere. However, it is pronounced /k/ in all environments by many speakers, particularly in informal speech.
  10. <r> is pronounced as a tap /ɾ/, but when doubled as <rr> it is usually pronounced /r/ or /r̥/. It is also often devoiced, particularly around unvoiced consonants.
  11. <t> is most often pronounced /t/ or /t̪/, but occasionally at the ends of words it can become softened to /θ/. It is also pronounced /ʃ/ (see below)
  12. <u> is typically pronounced /ʊ/ in stressed syllables and /u/ in unstressed syllables, however this can often vary and the two are sometimes analysed as existing in free variation. It is also pronounced as /w/ in diphthongs.
  13. <x> is pronounced as /ks/ intervocalically, and as /s/ elsewhere. There is a growing tendency to pronounce intervocalic <x> as a geminated /s/, probably from Italian influence.

There are also many sequences of letters that have perhaps unexpected pronunciations:

  • <di> + any vowel = /ʒ/, <dì> anywhere.
  • <de> (unstressed) + <a, o> = /ʒ/, <dè> anywhere.
  • <ndi> + any vowel = /ⁿd͡ʒ/
  • <nde> + <a,o> = /ⁿd͡ʒ/
  • <si>/<ti> + any vowel = /ʃ/.
  • <se>/<te> (unstressed) + <a, o> = /ʃ/.
  • <li> <ly> <le> <gi> <gy> <ge> are often reduced to simply /j/ in unstressed syllables.
  • <cî>/<tî> + any vowel = often /ç/, especially among younger speakers.
  • Final <t> and <d> sometimes are softened to /θ/ after vowels - particularly in informal or dialectal speech.
  • <mb> and <nd> are pronounced as long nasals rather than clusters or prenasalised stops. Sometimes there is a more forceful, stop-like release, but this is minimal.
  • Coda <n> before voiceless consonants is often deleted.
  • <tc> + high vowel = /t͡ʃ/, <tç> anywhere
  • <zc> + high vowel = /ʒ/, <zç> anywhere

Rhodian also has a system of velarisation, where liquids that are next to historic /g/ sounds are velarised. For example, the word gregalis (common, usual) is pronounced /ɾɣɛ.ˈjajs/. This applies to <r>, but also <l>, <m> and <n>. It is worth noting that among younger speakers this velarisation is disappearing and the consonants are pronounced plain.

Grammar[]

Nouns[]

Declensions, Gender and Cases[]

Nouns in Rhodian are divided into two genders, Masculine and Feminine. Within these genders, there are six declension patterns that determine how nouns change when they decline into each of Rhodian's five cases, which it retained from Latin - thanks in part to the Greek influence upon the language. A noun's declension is independent of its gender, and while there are some declensions that have a particular gender associated with them (1st declension feminine nouns, 2nd and 3rd declension masculine nouns) there are many exceptions to that rule. Furthermore, there are a small subset of noun endings that do not correspond to these declensions, as they are common derivational suffixes that diverged slightly, the most common example being -siu (from Latin -tiō, related to English/French -tion, Spanish -ción, Portuguese -ção, etc).

(A/N: I would highly recommend expanding the page view and hiding the sidebar so as not to be bamboozled by the formatting of these tables)

Singular i iia iib iiia iiib iv v vi Plural i iia iib iiia iiib iv v vi
nom -a -u -u -on -es / -is / -e ~ ~ nom -e -y -a -y -a -es -e / -a / -es -(e)s
acc -au -on -eu ~ / -a acc -as -os -os -as
gen -e -ie -ou -es -(o)s -a gen -aru -oru -on -u -on -e
dat -e -o -y -e -u dat -y -oìs -ivu -ase
loc-voc -a -e* -u -e -o -e ~ -îe loc-voc -e -y -a -oi -a -e -ac

NB: ~ refers to a very variable ending. | * the -e ending can also be -y if the nominative form of the noun ends in -iu

Rhodian has a case called the Locative-Vocative, which was formed from the merger of the Latin ablative and vocative cases. It is used as a vocative and also to mark the objects of many prepositions of location. It is also sometimes called the Prepositional-Vocative, and while being technically more accurate, it has less of a ring to it, so that name is less frequently used.

There are also many verbs in Rhodian that assign cases other than the Nominative and Accusative to their subjects and direct objects respectively, a phenomenon known as quirky subject.

The six declensions of Rhodian come from both Greek and Latin noun declensions that have collapsed into one another. However, the case of the sixth declension is interesting. During the period of standardisation that took place after the communist rise to power, the sixth declension was devised as a way to decline the new Russian loanwords in the language that did not easily fit into one of the pre-existing patterns. This is why the endings closely resemble the Russian masculine animate case endings. Older loanwords that did not fit easily into the case patterns, such as many from French and English, were reanalysed with these new endings so as to more easily be used in documents and official publications in Rhodian.

Finally, an interesting aspect of both nouns and verbs (as discussed later) in Rhodian is stem alteration, where the dictionary form of a word does not correspond to its stem. Examples include words like aner (man - whence pronoun ner, more below) becoming andra (acc); the stem of this noun is not aner- but andr-. Stem alteration is a phenomenon most common in the fifth declension but pervasive throughout all categories of nouns, much to the chagrin of bright-eyed learners.

Personal Pronouns[]

The Rhodian third-person pronouns are dissimilar to almost all of the modern Romance language pronouns, with the exception of Romanian. This is due to the fact they come from a different set of Latin demonstratives, notably 'ea/is/id' rather than 'illa/illus/illud'. They also notably decline for a different number of cases to most nouns, as they have kept the locative and vocative forms distinct rather than collapsing them together. The Genitive and Possessive pronouns are also identical, but are understood to be different due to their placement either before or after the relevant noun. The Comitative case comes from the old Latin forms 'mecum, tecum, etc' and was analogised and standardised as a separate case.

sg impersonal pl reflexive
1 2 3 masc 3 fem 3 epi 1 2 3 masc/mixed 3 fem
nom eo1 tu y ea2 ner nu vu ey e X
acc me te iu3 eau nèu dra4 eos eas se
gen my tui eiu5 nèiu der nostru vestru eoru earu sui
dat tive ei nèi dere noe voe eiey sive
loc me te nove vove yy seise
voc / emph. tu ea ner nui vui istè6 se
com mec tec iec eac nèc X nusc vusc yc sec
  1. Pronounced /ʒɔ/
  2. Pronounced /ja/ (see above)
  3. Originally ýu, often still so in older or more formal texts (pronunciation remains /ju/ rather than expected /ˈi.u/)
  4. In recent years, dra has slowly begun being replaced by ner
  5. The spelling and pronunciation of this pronoun, and its declined derivatives (as genitive pronouns agree in case and number with the possessed argument) is strange. It is pronounced in its base form /jɛw/ and all other forms follow this same pattern, such as eia (feminine nom sg possessee) pronounced as /ja/ and eiie (masculine gen sg possessee) as /ˈjɛ.jɛ/.
  6. From Latin istaec, pronounced /ɪʃ/ sometimes respelled <iç>

The Vocative is also labelled Emphatic, as some argue this label is closer to their actual function; similar to the emphatic or disjunctive pronouns in languages such as Dutch and French. However, this form remains that which the pronoun takes in agreement with certain prepositions that assign a more Vocative meaning, as well as being etymologically descended from the Latin (and in the case of ner, Greek) Vocative cases, so the name Vocative still tends to be used.

Furthermore, the third-person singular epicene pronoun is contentious among some speakers. It was coined in response to the growing frustration of a strange pair: the small Rhodian non-binary gender community and the even smaller Rhodian community of document writers. The former group wished to have an official pronoun by which to refer to themselves and the latter group wanted to cut down on the ink costs from having to print 'ey/e, eoru/earu' and such like on all their forms, not to mention the differences in formatting between choosing to type 'e(y)' or 'ey/e' and the potential biases toward either sex that this presented. Many older and/or less informed speakers combat the joint decision by the Buró Regularu Lengue Rodou (the Bureau of Language Standardisation of Rhodes) and the Lexicon Rodie (Rhodian Dictionary) to include it in their official documentation as of the 2021 reforms.

Articles and Demonstrative Pronouns[]

Rhodian has a definite article but no indefinite article. Occasionally though, the numeral one (ys/mía) is used to specify exactly one of something, but this doesn't carry the same exact meaning as a generic indefinite article. The definite article is very different from that found in modern Romance languages, due to it coming from Greek. It also declines for only four cases; the nominative also expressing the vocative and the dative also expressing the locative - this reduction in case forms being due to the articles' and numerals' Greek origin.

sg pl
m f m f
nom/voc o ha ty te
acc to ta tos tas
gen tas tu tau
dat/loc tu ta toìs tes

Note: mass nouns are ascribed plural articles, not singular ones. For example, 'the grass' = te herba, not ha herba (the latter would imply a specific type of grass)

There are also demonstrative pronouns, derived from Latin. These take slightly different cases than other nouns.

"This" fem masc "That" fem masc
Sg nom/voc he Sg nom/voc ista iste
acc han hun acc istau istu
gen huiu gen iscy
dat hui dat
loc ho loc ista isto
Pl nom/voc hei hy Pl nom/voc iste isty
acc has hos acc istas istos
gen haru horu gen istaru istoru
dat hy dat isty
loc loc


These follow a relatively predictable pattern of declension, contrasting with the personal pronouns.

There are also relative pronouns, used to link relative clauses. These follow a very Greek pattern, as they derive from the Doric ϝός. They are also very similar to the definite article.

fem masc
Sg nom/voc va vo
acc van vu
gen vas vou
dat/loc ve
Pl nom/voc ve vy
acc vas vous
gen voun
dat/loc vyse

Numerals[]

The last set of relevant determiners are the numerals, which up to and including the number four have irregular declension patterns. All cardinal numbers are derived from Greek, as that was the most common trading language in the Aegean Sea, but the ordinal numbers are all from Latin. Because of this, the Greek-derived numbers decline for the same four cases as the definite article, and the Latin-derived numbers for the five of most other nouns. The irregularly declined numbers are as follows.

1 fem masc 2 fem masc 3 fem masc 4 fem masc
nom/voc mía ys nom/voc nom/voc tcia tce nom/voc tetcia tetce
acc en acc acc acc tetcias
gen míe ens gen dyn gen tcion gen tetcion
dat/loc míy ien dat/loc dat/loc tcese dat/loc tetciarse

Verbs[]

Present Tense[]

Rhodian verbs are divided into four conjugations, based on the ending of the infinitive verb. All of these four verb classes can conjugate into all of Rhodian's 5 morphological tenses in the indicative mood and 4 in the subjunctive mood. The regular present tense indicative mood conjugations of verbs are as follows.

1 sg 2 sg 3 sg 1 pl 2 pl 3 pl meaning
Sperare spero speras sperat speramu speratìs speran "hope"
Rydere rydeo rydes rydet rydemu rydetìs ryden "smile"
Telire telio telis telit telimu telitìs telin "finish"
Gerre gero gers gert germu gertìs gen* "tell"

*/ɾn/ is not a permitted cluster word-finally, so the <r> is deleted.

As seen, each of the conjugations is characterised by its theme vowel - or lack thereof. Rhodian is relatively unique among modern Romance languages for having 4 conjugations rather than the typical 3, although Romanian has between 4 and 11 depending on who's counting. However, it is important to remember the phenomenon of stem change often comes into play, and is considered a regular feature of the language even though the patterns may seem arbitrary to the untrained eye.

Past Tenses[]

There are three past tenses in Rhodian, distinguished by aspect and proximity.

Past Perfect Tense[]

The past perfect tense describes actions that took place at a point in the distant past, but also can fill the role of a more traditional perfect tense such as that in English. It does not follow patterns based on the verb's conjugation and has only one set of endings. However, stem change is very common in this tense, and some conjugations are more prone to one pattern of stem change.

1 sg 2 sg 3 sg 1 pl 2 pl 3 pl
-e -iste -it -imu -istìs -erun


The stem change in this tense follows a few main patterns that are relatively predictable.

perfect present meaning
Coptare coptave copto "choose"
Plere ple pleo "sail, drive"
Strure struxe struo "build"
Quere quese queo "search"
Enere enue eneo "recommend"
Lenire lenie lenio "soften"

(organised in order of rarity, most common at the top) | NB: all featured patterns are considered regular

The patterns are characterised by their infixes between the stem and the ending.

Past Aorist Tense[]

For more recent past events, the aorist is used, characterised by its use of a prefix as well as a suffix. This is usually described in such terms, rather than as a circumfix, as while the suffix part changes with person, the prefix part does not. These also have only one form irrespective of conjugation class,

1 sg 2 sg 3 sg 1 pl 2 pl 3 pl
e- + -sa -sas -se -sams -s(a)te -san


There is sometimes a degree of stem change in the aorist, but it is far less common. Due to stress change brought about by the addition of a prefix, the vowels of a stem may sometimes be regularly simplified or simply deleted, but this is not considered stem change in its purest form. Sometimes, the prefixed e- will interact with the first vowel of a word and change it, but this is predictable. If the vowel is <a> or <i> it will change to <e> or <y>, but if it is any other vowel the e- will become i-, pronounced /j/. Also, the suffix will voice if next to a voiced consonant in most contexts.

aorist present meaning
Flere eflesa fleo "cry"
Orire iorse orio "appear"
Perdre eperdza perdo "lose"
Salutare eslutsa saluto "visit"
Amare emsa amo "like"
Ìplere plesa ìpleo "fill up"

The main possible aorist mutations

Past Imperfect Tense[]

This final past tense is probably the most simple; it undergoes very little stem change and the patterns are very predictable. It does vary based on conjugation, and the forms are displayed below.

1 sg 2 sg 3 sg 1 pl 2 pl 3 pl meaning
Locare locavau locavas locavat locavamu locavatìs locavan "put"
Cratere cratevau cratevas cratevat cratevamu cratevatìs cratevan "conquer"
Mentire mentievau mentievas mentievat mentievamu mentievatìs mentievan "tell a lie"
Crure cruvau cruvas cruvat cruvamu cruvatìs cruvan "bleed"


The imperfect is used to speak about events that were in the process of happening in the past. The degree of proximity does not matter as it does with the aorist vs perfect.

Future Tense[]

Rhodian retains the morphological future tense from Latin, and it is the most variable tense by conjugation. The forms for -are and -ere verbs are actually from the Latin future perfect tense, as the original future forms, as they evolved into Rhodian, were incredibly similar to the past perfect. This grew to be such a problem that the future perfect forms for these conjugation classes began to be used for only a future meaning, and the future perfect began to be expressed with a periphrastic construction (see below). This likely would also have happened to the -ire and -re verbs, as their future tense forms are rather similar to their present subjunctive forms, however the relative rarity of these two classes in comparison to the much more common -are and -ere verbs probably allowed this to slide under the radar. All of this explains the strange dichotomy of suffix forms in the modern Rhodian future tense. The forms are as follows.

1 sg 2 sg 3 sg 1 pl 2 pl 3 pl meaning
Postulare postulero postulérs postulért postulermu postulertìs postulér "require"
Têorere têorero têorérs têorért têorermu têorertìs têorér "consider"
Morire moriau mories moriet moriemu morietìs morien "die"
Legre legau leges leget legemu legetìs legen "say"


The future tense describes simple future actions irrespective of proximity.

Subjunctive Mood[]

The subjunctive mood in Rhodian is used much more pervasively than in other Romance languages. It has many functions on its own, and is also used in many verb constructions (see below). It is used after many prepositions and adverbs to express a slightly different shade of meaning, for example, du on its own means during or while, but with a subjunctive verb it means until. It is also used with determiners when giving examples; 'what is x' = qui est x, but 'what x is' = qui sit x/qui x sit. The subjunctive is also used for many generally irrealis sorts of statement, such as expressing doubt or suspicion, as well as in conjunction with the word nonne for expressions of surprise. e.g: nonne y ne veniat hodie? = he's not coming today?! With the indicative, nonne means more like 'right?' or 'isn't it?'. Its forms are quite similar to the present tense, but with some vowel alternation. There are subjunctive forms of all tenses except the future.

(present) 1 sg 2 sg 3 sg 1 pl 2 pl 3 pl meaning
Ydare ydeu ydes ydet ydemu ydetìs yden "sing"
Agnoere agnoeau agnoeas agnoeat agnoeamu agnoeatìs agnoean "ignore"
Calpire calpiau calpias calpiat calpiamu calpiatìs calpian "harden"
Ledre ledau ledas ledat ledamu ledatìs ledan "hurt"


(other tenses) 1 sg 2 sg 3 sg 1 pl 2 pl 3 pl meaning
Peptere Aorist epeptso epeptses epeptse epeptsomes epeptsete epeptsos "concede, give up"
Perfect peptueri peptueris peptuerit peptuerimu peptueristìs peptuerin
Imperfect peptereu pepteres pepteret pepteremu pepteretìs pepteren

These three tenses do not vary their endings based on their conjugation pattern in the subjunctive

Other Verb Constructions and Morphology[]

Pluperfect Tense and Anterior Past Tense[]

Rhodian lost Latin's morphological pluperfect and now uses a periphrastic construction using the supine participle form and the verb avère, 'to have'. This may be confusing to some speakers, especially of Romance the Germanic Languages, as it appears at face value more like a present perfect than a pluperfect. If you put the verb avère in the past tense, then it functions as more of an anterior past, but this is less common.

Xante ne nunquau ante at gleptu tumbon. = Xanthe had never seen a grave before. /ˈsan.tɛ nɛ ˈnuŋ.kʷaw ˈan.tɛ at ˈjɛp.tu ˈtʊm.mon/

Periphrastic Future Tense[]

The periphrastic future uses the future form of the verb esse, 'to be', and the infinitive form of the lexical verb. It tends to be used for proximal (near) future statements in very casual speech.

Eris cocre niemnoga ofáu? = are you going to cook some soup? (more literally: will you be cooking soup?) /ˈɛ.ɾɪs ˈkɔx.ɾ̥ɛ ɲɛ.ˈnɔ.ɣa o.ˈfaw/

Passive Voice[]

The passive uses the present tense form of the verb esse as well as the supine participle of the lexical verb. For the past tense, the usual imperfect form of esse is not used, but rather the perfect form. This replaced the Latin morphological passive, and the supine always agrees in gender, case and number with the subjectivised noun. The agent can be reintroduced in the dative case.

Subito, fuit sectu i meron (eie uxorule). = Suddenly, he was stabbed in the thigh (by his mistress). /ˈswɪ.to fʊjθ ˈsɛk.tu ɪ ˈmɛ.ɾon/ (/jɛ uk.ˈsɔɾ.wɛ/)

Dèy and Udén[]

The verb dèy /ʒi/ is a non-finite verb that has no infinitive form. It means something like 'one must' or 'you should', similar to il faut in French. There is a negative equivalent, udén, which is used for 'you must not', not 'you don't have to'. This is a relatively tricky difference, and is the bane of many learners. Neither dèy or udén conjugate for person, but they can be applied to all persons with an optional pronoun in the Dative Case. They are always paired with an infinitive lexical verb.

Dèy salutare mecrý tau tueu aviau. = You have to visit your grandmother on Wednesday! /ʒi sa.lu.ˈta.ɾɛ mɛ.ˈkɾi taw ˈtu.ew ˈav.jaw/

Udén garrire voe i ta ecclesia! = You (pl) mustn't chatter in church. /u.ˈðɛn ɣa.ˈr̥i.ɾɛ vɔj ɪ ta ɛç.ˈçɛ.ʃa/

Conditional Mood[]

Conditionals use a construction of the subjunctive form of esse as well as the infinitive of the lexical verb. They correspond to statements with 'would' in English.

Eo si magi potiu ércêre à eventu cu Marcelie tesy cu Máxime. = I would rather go to the party with Marcel than Max. /ʒɔ sɪ maj ˈpɔ.ʃu ˈɛɾ.kɛ.ɾɛ a ɛ.ˈvɛn.tu ku maɾ.ˈʃɛ.jɛ ˈtɛ.si ku ˈmak.sɪ.mɛ/

Hortative Mood[]

Hortative statements are also periphrastic, they use the auxiliary verb bet, which does not decline for person, and the correctly declined subjunctive form of the lexical verb. Hortative statements translate to sentences using 'let's ...' or 'shouldn't we ...' in English. Bet comes from the Latin verb 'baetere', which means something like 'go' but was quite lexically weak.

Rhodian Republic

A map of the Modern Rhodian Republic

Ne eco èi sumu sity, sè bet faciamu étiamsy! = I don't know if we're allowed, but let's do it anyway! /nɛ ˈɛ.xo i ˈsʊ.mu ˈsɪ.ti | sɛ bɛt fa.ˈʃa.mu ˈɛ.ʃam.si/

Prospective Aspect[]

An unusual feature carried over from Latin is the morphological (sort of) prospective aspect. The supine form plus the -ru suffix and paired with the copula gives the meaning of 'about to' or 'going to', much like a near future tense but not quite.

Sumu legtury cur casi syfa venuimu. = We're about to say why we came home late (lit. 'in the dark hours of the night'). /ˈsʊ.mu ˈjɛj.tu.ɾi kʊr ˈka.sɪ ˈsi.fa vɛn.ˈwɪ.mu/

Syntax[]

Word order[]

The order of words in a sentence is relatively free in Rhodian. With the exception of pre- and postpositions having to be adjacent to the nouns they modify and the difference in meaning between a genitive pronoun placed before and after a noun. The 'default' sentence structure is either SVO or SOV, but others such as VSO are also quite common. The word order is most often shifted for focus, with the focus of the sentence coming at the beginning.

Questions[]

Polar (yes-no) questions in Rhodian are usually formed by some more specific word inversion and by placing a question mark at the end of a sentence, marking a rise in tone. Compare a question with a statement:

Filipa mordet carneu. = Filipa eats meat. /ˈfij.pa ˈmɔɾ.ðɛ(t) ˈkaɾ.nɛw/

Mordet Filipa carneu? = Does Filipa eat meat? /ˈmɔɾ.ðɛ(t) ˈfij.pa ˈkaɾ.nɛw/

However, these inversions may also be present anyway in a statement. Thus is the question mark always necessary, though it is usually easy to tell from context whether something is a question despite lack of appropriate punctuation.

For content questions, i.e. questions containing 'who, what, when, etc', the question word is most often initial. The question words are as follows:

Why cur
What qui
Who quis
When quû*
How quom
How many quantu
How long quosque
To what extent quau
Where uy


When you use these in statements like 'this is how much flour you need' or 'I told him where I went', the subjunctive is used for the verb. However, quû is a special case as when it is used in this context it changes to cu (+subj still). This is pronounced identically or almost identically depending on accent, and is simply a relic of old Latin prepositional constructions.

Imperatives[]

There is a specific imperative form for the ends of verbs, with unique singular and plural forms. These vary based on conjugation:

sg pl meaning
Parare pare parate "prepare"
Tumbere tumbe tumbete "bury, entomb"
Spedire spedîe spedite "ship, dispatch"
Scandre scande scandète "climb"


These imperatives are used for direct commands, not requests, and thus may be considered rude if used inappropriately. There are some verbs, like esse, that have irregular imperative forms - in that case - but usually the form is very transparent.

Negatives[]

Negatives work very simply, just by putting the negative particle ne before the verb you want to negate. There are some derivational affixes with a negative meaning (e.g. à-, anti-) but these cannot serve as replacement for ne, i.e. one could not say "te àvideo", rather than "te ne video", as the former would mean "I un-see you".

Lexicon[]

RhodianGraph

A comparative pie chart of all the major sources of Rhodian core vocabulary

As stated before, the vocabulary of Rhodian is largely Latin-based (55%), with around 32% from Greek, mainly the Rhodian and Laconian Doric varieties. The remainder of the vocabulary is made up of French (~5%), Russian (~4%), Italian (~2%), with others such as English, Turkish and Arabic accounting for the rest.

Example texts[]

Tectasiu Universalis Drate Ántroporu, Statîé I à III[]

Statîá I

Toty átomy humany nascun lybres e equales axitaton e drate. Daty eiey sun logicu e sunydesis, cê sin ei devre facère intersive cu spyrite fraterne.

Statîá II

Ty drates e lybertata dixty i hà tectasiene dèy esse myrablies omnie, cars alique distincsiens; tanquau gene, colure, sexe, lengua, trescyia, plera publices an allou, génese gostarstvie an socialis, proprieta, lasiene an site alle. Porró, cars distincsiu sé factu perinde sitie publices, iuridicie an internationales gostarstvie an áreê uy-è veniat quispiau; étiamsy lyber, su custodia, ne guvernans se, an cu aliqui orvasiene lybertas.

Statîá III

Omnia at drà vyte, lybertas e siria sui.

TRANSLATION:

Universal Declaration of Human Rights, articles 1 to 3

Article 1

All human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood.

Article 2

Everyone is entitled to all the rights and freedoms set forth in this Declaration, without distinction of any kind, such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status. Furthermore, no distinction shall be made on the basis of the political, jurisdictional or international status of the country or territory to which a person belongs, whether it be independent, trust, non-self-governing or under any other limitation of sovereignty.

Article 3

Everyone has the right to life, liberty and security of person.

Ha Vea Procera à mía Ctona Parvau e Yratau[]

Díe 128, Annu Canonicu CG-ie 306

SPEDISIU

Donec civevat ea se, mnaovit tce reos. Primie, itinerevit trans to prostranstvu. Altres, erat incoaturu cu lavore novele vu ne potuit perdre. Tertiie, at corruptu preseu hodegou pre iu facère dare ei zapyscas identite noveles. Nyl huiu informasiens erat novel, sè ne erat saluber veniens ciendie se.

Undievat esse iau vigil, certe ne pret adú en díeu, hun myrat ea ast quiá reservandu avecsieneu srani. Avecsiu srani valet celnoc srani petans cu casto sraniie, e medìcina srani pre te anestesiere. Ea erat citu i sunydeseu niemnoga volta – oriens i ere, cadens apó momentu ea quispiau taxare potuit. O celnoc erat syfennu; ne gegerun aliquas ecranes guverneses. Ne gegit veau heuriscre quantu tempu essu fuit intre unusc orindu, quosque erat itineratu, an èi vel itinereret ea. Ha ennea eau fecit egra cremánandie.

Eia vorasis psylevit se satìs pre attendre ta fenestrau. Te cremante eran tracte à alte e arcevan omnie géneses potes luces. Ecvat nyl gegerun. Ea erat nunc certe i cenoma. Cars planetaru homylennaru, cars rute avecsionu, cars circulatru lucentu. Monu caû, caû atróx, ìpletu nyls an eia e quondau saxie.

O dvigateli torbavat du hetymavat y magi ien salto sustromas. Ha medìcina veneit poi, eau traêns i somnu molleu. Du explanavat ea, doxit rursu lavorie, mendaciie, andrié i prospo preses du cêvevat ea credites i eiu forsiereu. Mermirixit èi satìs fuerit. Dèy satìs esse, tau dèy. Emit iau nimiu errorivu vous ne fecit ea.

Eiy ocy clauserun. Ha medìcina eau cepit. O celnoc, sperose, perrexit.

TRANSLATION:

The Long Way to a Small, Angry Planet

Day 128, GC Standard 306

TRANSIT

As she woke up in the pod, she remembered three things. First, she was traveling through open space. Second, she was about to start a new job, one she could not screw up. Third, she had bribed a government official into giving her a new identity file. None of this information was new, but it wasn’t pleasant to wake up to.

She wasn’t supposed to be awake yet, not for another day at least, but that was what you got for booking cheap transport. Cheap transport meant a cheap pod flying on cheap fuel, and cheap drugs to knock you out. She had flickered into  consciousness several times  since  launch —  surfacing in confusion, falling back just as she’d gotten a grasp on things. The pod was dark, and there were no navigational screens. There was no way to tell how much time had passed between each waking, or how far she’d traveled, or if she’d even been traveling at all. The thought made her anxious, and sick.

Her vision cleared enough for her to focus on the window. The shutters were down, blocking out any possible light sources. She knew there were none. She was out in the open now. No bustling planets, no travel lanes, no sparkling orbiters. Just emptiness, horrible emptiness, filled with nothing but herself and the occasional rock.

The engine whined as it prepared for another sublayer jump. The drugs reached out, tugging her down into uneasy sleep. As she faded, she thought again of the job, the lies, the smug look on the official’s face as she’d poured credits into his account. She wondered if it had been enough. It had to be. It had to. She’d paid too much already for mistakes she’d had no part in.

Her eyes closed. The drugs took her. The pod, presumably, continued on.

Advertisement