Saltar al contenido principal

1.6. Theory ✅

info

This page contains an adaptation of information from the document "Grammàtica de base neolatina" [Neolatin Base Grammar], available on the page "Lengua" on the official Neolatin website. It explains the theory behind why the spellings exist as they are in Neolatin. For the sake of learning, you may wish to proceed to other lessons.

1.6.1. Multiple Romance Spellings​

The Romance spellings not only match the use of the Latin alphabet1, but also in the will to continue the traditional Latin spelling. For this reason, words like campo, cera, civetate, coppa et curare are written, in all the Romance languages, with <c> (Latin. CAMPUM, CERAM, CIVITATEM, CUPPAM, CURARE) despite Romance languages having different pronunciations depending on the following vowel: a, o, u vs. e, i (in Latin, in turn, always [k])2.

  • ACCENTUATION: The writing of Romanian does not allow the reader to know on which syllable falls the stress (mobilă, mobilă aut mobilă?). Italian spelling, in turn, uses diacritical signs over vowels to mark oxytonic words (città, perché, però, etc.) but it does not distinguish between paraoxytonic words (a word with stress on the penultimate syllable) and proparoxytonic words (a word with stress on the antepenultimate syllable): (rubrica aut rubrica?). Finally, Portuguese, Spanish and Catalan possess accurate and efficient systems that transcribe the accentuation in every word, using and omitting diacritics.
  • VOWEL OPENING: The diacritical signs are also used to transcribe the opening of the vowels e and o in Romance languages that distinguish between /e, ɛ/ and between /o, ɔ/. French has, for example, <é> for [e] (hétérogénéité) and <è> for [ɛ] (koinè), but on the other hands, it represents both [e] and [ɛ] with <e, ë, ef,...> (essayer, erre; canoë, noël ; clefs, chefs;...). Portuguese has a less complex spelling (it does not have 56 different spellings for [ɛ] like French), but it lacks the opening of these vowels only in part of the words, not for example in bestia with [e] ‘animale] or with [ɛ] ‘arma’? The Catalan and Italian spellings make the distinction in words that already have a diacritic for the accent of intensity (more cases in Catalan than in Italian)3. Finally, Occitan spelling has the most accurate system. It allows one to know the opening in all words, with diacritics.
  • VALUE OF THE LETTER <J>: The letter J represents different sounds in Romance spelling. Originally it was a decorative variant of <i>, but it became a new letter in the 16th century, at the proposal of the Italian humanist Gian Giorgio Trissino. In Italian, it traditionally represents the sound [j] (like in Latin, and like so in Slavic languages written in the Latin alphabet and Germanic languages, except for English), although Italian conserves that use now in a limited manner (especially in some nouns particular of Italian like Jacopo, Jesolo, etc.), generally substituted by <i> (iena, notaio, Savoia, etc.). In other Romance spellings it represents the sound that in the respective languages continues the Latin initial i (for example in Latin. IOCUM): [ʤ > Ʒ] (Portuguese. jogo, Catalan. joc, French. jeu, Romanian. joc) and [ʒ > χ] (Spanish. juego). Italian also has [ʤ], but it also uses the digraph <gi> (gioco).
  • PALATAL N: Almost all Romance languages, including the six largest Romance languages, present (or previously presented) the sound [ɲ]. However, the spellings used to represent that sound are multiple: <nh / ñ / gn / ny>. Portuguese uses the silent letter H as a “note” to indicate that the previous consonant does not have the usual sound. This spelling, borrowed from the Occitan minstrels (“trobadores”) is conventional, but it does not have a phonetic basis. In turn, the other languages are based on old sounds from which emerged [ɲ]:

  1. The old sequence [nn] explains <ñ> in Spanish (abbreviation of <nn>): Latin ANNU(M) > Spanish año. This option could also have been chosen for Catalan.

  2. [gn] explains <gn> in French and Italian: Latin. PUGNU(M) > Italian. pugno. This option could also have been chosen for Portuguese, Spanish and Catalan.

  3. [nj] explains Catalan. <ny>: Latin. ARANEA(M) > Catalan. aranya. This option could have been chosen for other languages, since in all of them the sequence [nj] produces [ɲ] (currently [j] in Romanian: râie). An equivalent digraph, and more etymological, could have been <nj>, at least in Italian, where <j> is read [j]4.

  • ETYMOLOGICAL <H>: The sound /h/ disappeared from the Latin language during the Roman era, but back then the writing with <h> had already been consolidated and, therefore, it was conserved as a silent letter (with some hesitation, like in HUMERUS / UMERUS). The Romance spellings continue this old tradition, but to varying degrees, as shows the following table which contains 10 inherited Pan-Romance words that had <h> in Latin:
LATINPORTUGUESESPANISHCATALANFRENCHITALIANROMANIAN
ECCUM HIC, ECCE HICaquíaquíaquí, acíciquiaci
HABEREhaverhaberhaveravoiravere, ho, haavea
HEDERAMherahiedraheuralierreederaiederă
HORAMhorahorahoraheureoraoară
HOSPITEMhospedehuéspedhostehôteospiteoaspete
HOMINEM, HOMOhomemhombrehomehomme, onuomoom
HOC DIEhojehoyavui, huihuioggiazi
HERBAMervahierba, yerbaherbaherbeerbaiarbă
HERI(eire)ayerahirhierieriieri
HIBERNUMinvernoinviernohivernhiverinvernoiarnă
100 %60 %65 %85 %70 %5%0%

As the table shows, Catalan and French are the languages that present the most silent Hs in inherited vocabulary, Spanish and Portuguese have fewer of them, Italian much fewer (only when the letter is useful to distinguish homophones: o, ho; a, ha), and Romanian none5. The results are similar in borrowed Greco-Latin vocabulary6, but here Romanian presents 50% of (sounded) Hs and Italian dispenses with this letter completely.

On the other hand, some users also have different spelling preferences. Among some people interested in Pan and Inter-Latin communication, some love the etymological spellings (typo, theatro, etc.), which call back to Latin and Greek; others, in turn, support a more regular, phonemic spelling, which does away with old distinctions that are no longer relevant in current pronunciation (tipo, teatro, etc.).

Keeping in mind the existent plurality regarding spelling, not lacking in arbitrariness, Vía Neolatina proposes a writing system for Neolatin that can be descript, summarized using the following qualifications. It is (or proof of being):

  • FLEXIBLE and
  • INTUITIVE

These two characteristics or traits of the spelling (flexibility and intuitivity) correspond to Vía Neolatina’s two objectives of codification (see section 4.2 of the “Design Principles of Neolatin”): To develop a linguistic model with

  1. MAXIMUM POTENTIAL FOR ACCEPTANCE
  2. MAXIMUM COMMUNICATIVE VALUE

1.6.2. An Intuitive Spelling​

On one hand, the spelling proposed for Neolatin attempts to be maximally intuitive, in a way that may allow users not only to identify words (in the case of Romance users) but also to pronounce them well. This is sought out keeping multiple criteria in mind: the logic of the phonological processes that produce the sounds, the graphical accuracy, the forms and majority uses in Romance languages, etc. Here are some examples:

Spelling of Palatal Sounds​

An example of intuitive spelling is that of palatal sounds, which did not exist in Latin. As has been seen above, to represent the sound [ɲ] the Romance languages use on one hand <nh> and on the other <ñ / gn / ny>. The first spelling does not have a phonetic basis: the sound [h] does not palatalize a preceding consonant. As a result, whoever is not familiar with the convention cannot intuitively pronounce the sound, nor approximate it. This is not advisable in Neolatin.

The other spellings do have a phonetic basis, convenient in Neolatin. Of the three options, <ny> is the most adequate, since it is based on the sequence [nj], which is the origin of [ɲ] in more Romance languages, is the only origin of the sound in Neolatin and is also the closest sequence phonetically to [ɲ] ([sɔnnjo] is a good approximation of [sɔɲɲo]). In fact, it is frequent in the pronunciation of foreign speakers7.

However, the spelling <y> is not used for other palatal sounds (they do not existen –at least currently– the digraphs <ly, cy, gy, sy> for the sounds [ʎ, ʧ, ʤ, ʃ]8). As a palatalizing spelling, the letter <i> is used more in Romance languages: in <gli> (Italian. foglia), <ill> (French. mouiller Old French. [ʎ]); <ci> (Italian. crociare, Romanian. crucia); <gi> (Italian. gioco, Romanian. giuvaier); <sci> (Italian. sciame). And the vowel i, as well as e9, palatalize some consonants in practically all the Romance languages: ci, gi (vs. ca, ga). Furthermore, the digraphs with <y> are not etymological (Latin. SOMNIUM, FOLIAM, MOLLIARE had <i>). For all these reasons, Vía Neolatina proposes the use of the spelling <i>.

However, it is necessary to somehow distinguish those words from the many Latin borrowings that conserve the sound [i] without palatalizing the preceding consonant. Vía Neolatina proposes the use of a circumflex accent:

NON-PALATALIZED SOUNDPALATALIZED SOUND
decènniosònnîo
rebellionemollîare
edificiocrocîare
religionefugîa
visionecamisîa

The function of the circumflex in Neolatin can be easily deduced by the majority of Romance speakers that are familiar with these sounds in their languages. And if the sign is ignored, the result is still a very similar sequence to the desired sound. In any case, the addition of this diacritical sign is the minimum innovation that allows having a maximally convenient spelling in Neolatin according to multiple codification criteria. And it is rarely needed.

Geminated Consonants by Yod​

Romance languages present double consonants inherited from Latin (Lat. VACCAM, ILLAM > Neo. vacca, ella) and some new ones. Among these, the group of double consonants originating from a [j] stand out, such as in Lat. PLATĔAM, PŬTĔUM > Lat. vul. PLAT[j]A, PŬT[j]U > Romance pla[tʦ]a, po[tʦ]o. In these cases, Via Neolatina recommends graphically representing the duplication in Neolatin, as it generally done in Italian:

LATINNEOLATINITALIANSOUND
FACĬAMfaccîafaccia[tʧ]
CORRIGĬAMcorreggîacorreggia[dʤ]
HODĬEhòjjeoggi[dʤ]
FILĬAMfillîafiglia[ʎʎ]
PINĔAMpinnîapigna[ɲɲ]
PLATĔAMplattsapiazza[tʦ]

In reality, the graphic duplication of [ʎʎ] and [ɲɲ] is not strictly necessary, since these sounds are always double in an intervocalic position. In fact, Italian does not write neither figglia nor piggna. However, in any case, it is necessary to distinguish between double consonants [tʧ, dʤ] (in faccîa, correggîa) and the simple consonants [ʧ, ʤ] (in crocîare, fugîa, de croce, fugere). Therefore, out of systemacity and to avoid confusion, it is convenient to also write double consonants [ʎʎ, ɲɲ, tʦ], as well as [dʤ] in cases like hòjje. The double spelling avoids exceptions and is intuitive. Furthermore, there are some cases of simple [ʎ] and [ɲ] in Neolatin in an initial position (such as lîama or nîam).

In any case, it is advisable to write the suffix -izzare with double spelling, which presents a double sound (Ita. organizzare) despite the simple spelling in Latin (Lat. medievale ORGANIZARE), which imitates the simple Greek spelling. The pronunciation in Koinè Greek was [zz].

Use of the Letter <j>​

In an initial position, Latin [j-] (like in IAM) is confused with [dj-] (DEŌRSUM) in all Romance languages, resulting in [j-], from which [ʣ-, ʤ-]; in an intervocalic position, [-jj-] (like in MAIUM, with double yod despite the simple Latin spelling) is confused with [-gj-] (FAGEUM) in all Romance languages, with the results [-dʣ-, -dʤ-, -j- (simple)]10. Neolatin respects these Pan-Romance convergences, but which result (and spelling) is presented?

In Romance languages, the main solutions are [(d)ʤ] and [j] which, as the following table shows, has a very complex distribution, with variation even between the majority of languages: Portuguese, Spanish, Italian and Romanian (in different proportions, if other words are studied).

[ʤ] (current or old)[j] (current or old)
IOCUM, IAMPortuguese. já, jogo; Spanish. juego ([ʒ > χ]); Catalan. ja, joc; French. (dé)ja, jeu,; Italian. già, gioco, Romanian. jocCas. ya
DEŌRSUMPortuguese. juso; Old Catalan. jus, Old French. jus; Italian. giuso, Romanian. josCas. yuso
MAIUM, MAIOREM, CUIUM , MAIALEM, IEIUNAREPortuguese. cuju; Catalan: maig, major, Italian. maggio, maggiore, Romanian: ajunaPortuguese. maio; Spanish. mayo, mayor, French. mai, Old French. maiour, Italian. maiale, Romanian. maiu
CORRIGIAM, FAGEUM/FAGEAMCatalan. corretja, faig; Italian. correggia, faggioPortuguese. Correia, faia; Spanish. haya 11; French. courroie, Old French. faie

In an initial position, [ʤ] clearly prevails and in an intervocal position, [j] lightly prevails. Altogether, [ʤ] prevails, although lightly.

On the other hand, while [j] is simple in all Romance languages12, the result [(d)ʤ] conserves the double pronunciation (CORRIGIAM > Italian. correggia13) in an intervocalic position, which is systematic with the result of other groups of consonants father by a semiconsonant (Neolatin. faccîa, fillîa, vinnîa, etc.).

Furthermore, [-gj-] (CORRIGIA) > [dʤ] (correggîa) has a result parallel to the voiceless analogue, [-kj-] (FACIA) > [tʧ] (faccîa).

Finally, [(d)ʤ] is the only option compatible with an etymological spelling in Neolatin, since it can be represented as <(j)j> or <ggî> according to the etymon: Neolatin. ja, josso, majjo vs. faggîo. This is coherent with the majority use in Romance language an initial position, where the spelling is distinguished from the successors of Latin. [j-] et [g-]:

  • IŎCUM > Neolatin. jòco (Portuguese. jogo, Spanish. juego, Catalan. joc; French. jeu, Romanian. joc vs. Italian. gioco14)
  • GĔLUM > Neolatin. gèlo (Portuguese., Italian. gelo, Catalan., French. gel, Romanian. ger vs. Spanish. hielo, with yod).

In any case, <(j)j>, <ggî> et <g> + e/i (in words of Latin origin) can be pronounced in Neolatin alternative with a semiconsonant: IAM > ja ([ʤ]a / [j]a); DEŌRSU(M) > josso [ʤ]osso / [j]osso), HŎDĬE > hòjje (hò[dʤ]e / hò[j]e), FAGEU(M) > faggîo (fa[dʤ]o / fa[j]o15), GĔLU > gèlo ([ʤ]èlo / [j]èlo).

Value of the Letter <z>​

In some Greco-Latin borrowings (such as zèta, zoo, zòna), Italian presents the sound [ʣ], written <z>. Other Romance languages, in turn, pronounce these words with [z] (or [s /θ] in Spanish), but they also write them with <z>16. This was, to be precise, the Greek spelling, which was incorporated into the Latin alphabet. Therefore, this is unanimity. But what is the recommended way of pronouncing this letter in Neolatin?

The phonetic value of <z> in Latin and even in Greek is a complex matter:

  • In Ancient Greek, the phonetic value of the letter <Z, ζ> was [zd] or [dz] (there is no consensus among scholars). In Koinè Greek, it represented [zz] and, since the Middle Ages, [z].
  • In Latin, the sound [z] already existed in native words, but only as an allophone of /s/ in contact with a voiced consonant17. In Greek loanwords, <Z, ζ> was adapted into Ancient Latin with an initial <z> (sona) and a middle <ss> (massa), spellings that represented the closest Latin phonemes (Cockdurn. S.d.: 94)18. Subsequently, Greek influence grew and in the first century BC, Latin introduced, from the Greek <Z, ζ>, the letter <z> (zelus, zòna, Zephyrus, zmaragdus, Amazon). There is no consensus among scholars on its pronunciation in Latin:
    • [s] was heard more than [z] (at least outside of academic language), according to Touratier (2008: 16);
    • [z], according to phoneticists Bassols (1983: 35 and 178), Molina (1993: 12) and Cockdurn (S.d.: 147);
    • [z] at the beginning of a word and [zz] inside a word, according to Panhuis (2009: 4);
    • [ʣ] at the beginning of a word and [dʣ] in an internal intervocalic position, according to Morani (2000: 172);
    • [dz] or [z], according to Müller-Lancé (2006: 77)
    • [z] (with the variants [zz], [dz]), according to Canepari (2008: 5)
  • On the other hand, popular Latin developed a different pronunciation: [dj]. A piece of evidence for this is that until the second century, the spellings <-z-> and <-di-> appear interchangeable in inscriptions: oze, azutor, zebus, zabulus for hodie, adiutor, diebus, diabolus (Cockburn. S.d.: 145, 399). The variants -izare (learned) and -idiare (popular) from the Greek verb suffix -ίζειν, has semantic distinctions in belated texts; and in Romance languages, different evolutions:
  • -IDIARE > Portuguese, Spanish, Catalan –ejar, Spanish –ear, –iar, French –eier > -oier, Italian -eggiare;
  • -IZARE > French. –iser, Italian. –izzare (Cockdurn. S.d.: 146-147).

Consequently, these Greco-Latin borrowings have, in Neolatin, the spelling <z, zz> and both the pronunciation [z,zz] (majority) and [ʣ, dʣ], besides [dʤ] in the popular variant of the suffix, written <jj> (cortejjare, guerrejjare, verdejjare).

1.6.3. A Flexible Spelling​

On one hand, the spelling proposed for Neolatin is flexible, adaptable to different sensibilities. It includes two compatible and combinable graphical options: one more etymological and another more regular.

The following table presents the options proposed for the writing of Neolatin, which allows the users to adapt it to their preferences to feel more comfortable:

ETYMOLOGICAL SPELLINGSREGULAR SPELLINGS
lettera, porta, crociareléttera, pòrta, crocîare
lingua, inlengua, en
musca, cunmosca, con
pœnapena
cæcocèco
typotipo
graphíagrafía
theatroteatro
chòrocòro
rhetòricaretòrica
havere, inhumano, deshonesto 19avere, inumano, desonesto
ja, jorno, majjore, mèjjogîa, gîorno, maggîore, mèggîo
kilogrammachilogramma
quattro, quinto 20cuattro, cuinto
sexantasecsanta
orationeoratsione
zòna, organizzarezòna / dzòna, organizzare / organiddzare
et, aut, ad [e, aw, a]21e, au, a22
èst [ɛ / ɛst]è / èst
que, qui, quèn [ke, ki, kɛn]che, chi, chèn

However, the spelling that Vía Neolatina recommends as the standard for Neolatin is a balance combination of the two options (etymological and regular).

The recommended spellings below are considered more intuitive23.

RECOMMENDED SPELLINGS
léttera, pòrta, crocîare
lengua, en
mosca, con
pena
cèco
tipo
grafía
teatro
còro
retòrica
havere, inhumano, deshonesto
ja, jorno, majjore, mèjjo
kilogramma
quattro, quinto
sexanta
oratione
zòna, organizzare
et, aut, ad [e, aw, a]
èst [ɛ / ɛst]
que, qui, quèn [ke, ki, kɛn]

The reasons for these recommendations are the following:

  • The diacritical signs <◌́, ◌̀, ◌̂> (the acute, grave and circumflex accents) proposed for Neolatin are not necessary to understand a written text, obviously. Therefore, they are dispensable, especially in informal contexts24. However, there use is important to know and learn the pronunciation of Neolatin, which is not only a written language. In any case, all the Romance spellings use diacritics (in fact, they were not unheard of in Latin). And the frequency of use of these signs in Neolatin are at the midpoint of the Romance spellings. This is certainly more than in Italian, where only 1% of the letters of a text present diacritics; but in this language, the writing does not represent the phonetic accentuation and the vowel opening well. Portuguese is the opposite case: it uses these signs in 5.4% of letters, but it does not end up specifying all the open vowels. Neolatin presents a middle option: with only 3.4% of modified letters, it can already completely indicate the standard pronunciation. It is a beneficial investment.
  • Writing with <i, u> in words like lingua, in, musca, cun, which had the short vowels ĭ and ŭ in Latin, brings inherited vocabulary from Neolatin closer to its mother language and to multiple Latin borrowings incorporated in Romance languages (as it has been in Neolatin lingua, like in linguística). In fact, inherited words can be pronounced, with [i, u] in Neolatin, since the Latin vocalic timbre has been conserved in Sardinian and Romanian (here, only in the case of u). However, the majority of Romance languages have opened those vowels and present (or presented in a previous phase) the pronunciation [e, o]: l[e]ngua, [e]n, m[o]sca, c[o]n25. This is the preferred pronunciation in Neolatin, and representing it with the Latinized spellings lingua, in, musca, cun complicates writing and is problematic because other words must be read with [i, u], such as vino, cinque, luna, fuste (they had the long vowels ī and ū in Latin). This affects many words. Therefore, the preferred writing is with <e, o>26.
  • The use of both <œ, æ> and <y, ph, th, ch, rh> is not necessary in Neolatin: they are irrelevant phonetically (except <ch> for Romanian, but only in some of the words: Neolatin. anarchía, Romanian. anarhie vs. Neolatin. anachrònico, Romanian. anacronic) and its use is a minority (only in French, and not in a systematic manner: Neo. rhetòrica, French. rhetorique vs. Neolatin. rhythmo, Fra. rythme). However, they can be used to the desire of the Neolatin user. In fact, they are used in other European languages, such as English and German.
  • The conservation of the graph <h> is anachronistic in Romance languages, certainly. In fact, this sound was already lost in latin (not the spelling27), and Italian almost completely dispenses of the letter. However, in Neolatin the variant with <h> (humano, inhumano, deshonesto) is prioritized because it is the majority in Romance languages, especially in loanwords (in inherited words, Romance languages write <h> in approximately 50% of cases). Furthermore, it is a writing that is diasystematic with other Indo-European28 and international families. However, there are exceptions among compound words. While the relation between antehère and hère is evident, other compounds are already not perceived as such. If, beyond this, the writing with <h> is rare in Romance languages, it is already not recommended in Neolatin:
    • Latin. HANC HORAM, HAC HORA, AD ILLAM HORAM, ADHUC > Neolatin. ancora, acora, allora, aduc. Romance languages generally do not present <h> in these words (only Spanish. ahora, which lacks the first H, however). Furthermore, with the writing hanchora, hachora, allhora, adhuc, the words would be unrecognizable in Neolatin.
    • Latin. ECCUM HIC > cui, not *cuhi . Romance languages do not conserve <h> in this compound word (Portuguese. aqui, Spanish., Catalan. aquí, Italian. qui). Furthermore, it favors the current pronunciation with a diphthong.
    • Latin. AD HIC / IBI > adí, non *adhí . Two etymons converge in this word, one with <h> and another without: HIC (in Spanish. ahí) and IBI (in Portuguese. aí). Furthermore, a writing without <h> is systematic with cui.
  • ja, jorno, majjore, mèjjo are preferable to gîam, gîorno, maggîore, mèggîo because the majority (5/6) of Romance languages conserve the spelling <j>, at least in Latin borrowings:
    • JUSTITITA → Portuguese. justiça, Spanish. justicia, Catalan. justícia, French. justice, Romanian. justiţie (vs. Italian. giustizia).
  • In the case of words like kilogramma vs. chilogramma, Romance languages are divided without a clear majority.
    • Portuguese. quilograma, Catalan. quilogram, Italian. chilogrammo vs. Spanish. kilogramo, French. kilogramme, Romanian. kilogram. However, the etymological spelling is more international: English. kilogram, German. kilogramm.
  • quattro, quinto are preferred to cuattro, cuinto because the majority (4/6) of Romance languages conserve the spelling <qu>, at least in Latin borrowings:
    • AQUATICUS → Portuguese. aquático, Catalan. aquàtic, French. aquatique, Italian. acquatico (vs. Spanish. acuático, Romanian. acvatic).
    • FREQUENTE(M) → Portuguese., Italian. frequente, Catalan. freqüent, French. fréquent (vs. Spanish. frecuente, Romanian. frecvent).
  • sexanta is preferred to secsanta because the majority (5/6) of Romance languages conserve the spelling <x>, at least in Latin borrowings:
    • CONVEXUS → Portuguese., Spanish. convexo, Catalan., Romanian. convex, French. convexe (vs. Italian. convesso).
  • The spelling <ts> clearly transcribes the sound [ts]. It is used in Aromanian: ortsã, bratsu, dultse ‘dolce’, etc. Other Romance languages also use it, in exotic loanwords like tsar, tsunami, tsetsé, tsuga, etc. In Neolatin, it is used in both these loanwords and in the relatively few words inherited from Latin that generally present or presented the sound in Romance languages (fòrtsa, altsare, avantsare, etc.), from which comes the Vulgar Latin sequence [-tj-], among other origins. In Latin borrowings (a language where the sounds [ts] did not exist in Classical pronunciation, but indeed in Medieval and later), Romance languages transcribe [ts] with the spellings that in the respective spellings, are read (or were read) [ts]: <c, ç, ţ, z>. Furthermore, French also presents the etymological spelling <t>.
PORTUGUESEtradiçãoterciáriosentençiarionegociador
SPANISHtradiciónterciariosentencianegociador
CATALANtradicióterciarisentèncianegociador
FRENCHtraditiontertiairesentencenégociateur
ITALIANtradizioneterziariosentenzanegociatore
ROMANIANtradiţieterţiarsentinţănegociator

Applying these same procedures, these same Latin borrowings in Neolatin can be written regularly with <ts> oratsione, sentèntsia, dictsionario) or conserving the Latin spelling (oratione, sentèntia, dictionario)29. The second option, despite being a minority option in Romance languages (it was normal in Old Italian), seems to be preferable in Neolatin, since the spelling <ts> is only used in these loanwords in Aromanian (traditsiunal, dictsiunar)30 and it affects a vast group of words, which are furthermore international in many cases.

  • The spelling <dz> clearly transcribes the sound [dz]. It is used in Aromanian: prãndzu ‘lunch’, ordzu ‘barley’, dzer ‘frost’, etc. Other Romance languages also use it, in exotic loanwords like dzong, dzihunia, etc. In Neolatin, it is used both in these loanwords and in relatively few words inherited from latin that generally present or presented the sound in Romance languages (admòrdzo, verdza, hòrdzo, prandzo, vercondza), where it comes from the Vulgar Latin sequence [dj] after a consonant. In Neolatin, Greco-Latin borrowings with Z have both the pronunciation [z, zz] (majority) and [ʣ, dʣ], besides [dʤ] in the popular variant of the suffix (cortejjare, guerrejjare, verdejjare) (see the reason above):

    • [z, zz] is written without problems with <z> (zèta, zoo, zòna, etc.) or <zz> (in the verbal suffix -izz, like in organizzare)
    • [ʣ, dʣ]
      • This pronunciation can be written with the traditional spelling <z / zz>. However, it collides with the suer of <z> for [z] in some loanwords that are neither Latin nor Greek, like bazar (< Farsi. bāzār), zèro (< Ita zero < Lat zephĭrum, adaptatione de lo àrabo ṣifr). Also in zebra, originating from Ibero-Romance31.
      • An alternative is a regular spelling with <dz> (dzòna, dzèta, dzoo, etc.). However, it is rare in Romance languages: it mainly appears in Catalan -itzar and in the word Portuguese. dzero/Spanish. dsero (the name of the Greek letter).
    • About et, aut, ad, see Syntactic Gemination.

  • The verb èssere –as well as havere– is a very irregular verb and with short forms, because it is used frequently. The singular third person of the present indicative, which was est [ɛst] in Latin, presents in Neolatin two forms:

    • èst [ɛst], a conservative form corresponding to Spanish., Catalan. [es], French. [ɛt] lia. and Romanian. ['jeste];
    • è [ɛ], a reduced form corresponding to Portuguese., French., Italian. [ɛ], Romanian. [je]. It is systematic with the fall of the –t in the third person, forming the majority in Romance languages and preferred in Neolatin. However, it has a minimal body, which make its identification difficult.

    Regardless of the pronunciation [ɛ] or [ɛst], the writing <èst> is preferred in Neolatin, which makes the word more identifiable, thanks to the fact that the Latin form is more known.

  • The [-w-] of the Latin sequences [kw] and [gw] are conserved before e, i in Neolatin, since it is the encompassing form:

    • Neolatin. quercîa, cinque, quíndece, sequire (Italian. quercia, cinque, quindici, seguire)
    • Neolatin., Italian. sangue, anguilla.

However, there are some exceptions in the case of [kw], notably Neolatin. que, qui, quèn, which have lost the [-w-] in all Romance languages. Lausberg (1965: § 345) explains it by analogy with QUOD > Vulgar Latin. [ko]. Therefore, these three words are pronounced in Neolatin preferably without [-w-], and just as well for the compound words qualque, quionque, alquèn32. However, it is convenient to conserve the traditional spelling, since writing these frequent words can be less recognizable for many Romance speakers (the Western languages conserve the spelling <qu>, and only Italian presents <ch>). The result is a maximally intuitive standard spelling.

1.6.4. Associated Bibliography​

TOURATIER, Christian (2008). Grammaire latine. Paris: Sedes.

MOLINA YÉVENES, José (1993). Iniciación a la fonética, fonología y morfología latinas. Barcelona: Publicacions Universitat de Barcelona.

BASSOLS DE CLIMENT, Mariano (1983). Fonetica latina. Madrid: Consejo Superior de Investigaciones Científicas.

PANHUIS, Dirk (2009 [1998]). Latin Grammar. S.c.: University of Michigan Press.

MORANI, Moreno (2000). Introduzione alla linguistica latina. München: Lincom Europa.

MÜLLER-LANCÉ, Johannes (2006). Latein für Romanisten. Tübingen: Narr.

CANEPARI, Luciano (2008). La pronuncia «neutra, internazionale» del latino classico. No city specified: no publisher.

COROMINES, Joan (1980-1991). Diccionari etimològic i complementari de la llengua catalana. Barcelona: Curial Edicions Catalanes. Free access at: https://decat.iec.cat/

REPETTI, Lori (2016). «Palatalization» en Adam LEDGEWAY et Martin MAIDEN (eds.) (2016).

VÄÄNÄNEN, Veikko (2003 [1988, 3a ed.]). Introducción al latín vulgar. Madrid: Gredos.

BADIA I MARGARIT, Antoni (1981). Gramàtica històrica catalana. València: Tres i Quatre.

ROHLFS, Gerhard (2021 [1966]). Grammatica storica della lingua italiana e dei suoi dialetti. Vol. I, Fonetica. Bologna: Il Mulino.

NIEDERMANN, Max (19534). Précis de phonétique historique du latin. Paris: C. Klincksieck.

LOPORCARO, Michele (2011). «Phonological processes» en Martin Maiden et al. (2011), pg. 109-154.

Trésor de la Langue Française informatisé (Dirs. Paul Imbs et Bernard Quémada. Paris: CNRS, 1971-1994), diccionario de francese con informatione etimològica. Free access at: http://atilf.atilf.fr/

Footnotes​

  1. The spellings <ç, ş, ţ> existing in some Romance languages are not particularly new letters, but letters complemented by diacritical signs (compare. <á, à, â, ă, ã>). ↩

  2. A parte of Sardinian conserves the pronunciation [k] before e, i. ↩

  3. For [ɛ, ɔ] The Greek letters <ɛ, ω> were proposted for [ɛ, ɔ] in Italian by the humanist Gian Giorgio Trissino, in his Ɛpistola del Trissinω de le lettere nuωvamente aggiunte ne la lingua Italiana. However, unlike other proposals of his, this was not followed. In turn, the International Phonetic Alphabet adopted <ɛ> for [ɛ]. ↩

  4. In Aromanian, the digraphs <nj> and <lj> for the sounds [ɲ, ʎ]. In Italian, <lj> for [ʎ] (for example, in vollja, normally spelled voglia) was another one of the innovations proposed by Trissino. ↩

  5. Romanian, in fact, does not use <h> as a silent letter because, in this language, the sound /h/ exists in loanwords from other languages (and, consequently, Latin borrowings as well). ↩

  6. French. 95%, Catalan. 70%, Portuguese. 70%, Spanish. 50%, Romanian. 50%, Italian. 0%. These are the results of the 10 Greco-Latin loanwords with <h>: ABHORRERE, PROHIBERE, ADHAERENTEM, HUMANITATEM, HYPOCONDRIAM, HYPOTHECAM, HARMONIAM, HELENA, HIEROGLYPHIKÁ (Greek.), HYACINTHUM. A larger sample would possibly reduce the Portuguese result, since here <h> does not exist in cases like Portuguese. anelar, inabitável. ↩

  7. Also in Sardinian, a Romance language without [ʎ, ɲ], Italian loanwords with those sounds are adapted to Campidanese Sardinian as [llj, nnj] (with the variant [ʎ, ɲ]): Sardìnnia [nnj / ɲ], màllia [llj / ʎ]. ↩

  8. In Old Catalan, <yl> and rarely <ly> (this digraph was used, for example, systematically by the second copyist of the Roussillonese Vides de Sants, as reports Coromines in Lleures i converses d’un filòleg 1983: 289.) ↩

  9. In French and Romanian, the e works also as a palatalizing note: French. Je mangeais, Romanian. ceapa, geam. ↩

  10. According to Loporcaro (2011: 144) -and essentially Repetti as well (2016: 660)-, Proto-Romance [(-j)j-, (-)dj-, (-)gj-] are soon confused (1st century AD?) in [-j(j)-], from where [(-d)ʣ-] (affrication and -possibly after- [(-d)ʤ-] (palatalization), with subsequent evolutions [z] and [ʒ > x], respectively. Currently, a large part of Southern Italo-Romance presents yod. For example, in Sicilian IOCUM, PEIUS > [j]ocu, pe[j]u, HODIE >o[j]i, FUGIO > fu[j]u (also GINESTRAM > [j]ine[ʃ]tra). However, Loporcaro (ibidem. 145) and Repetti (ibidem. 660) report two interpretations (without showing a preference): this Southern Italo-Romance yod has been considered a preservation (Rohlfs 2021 [1966]: section 278, page 393) and a regression from a previous affricate -the Spanish yod would also be it- (Väänänen 2003 [1988]: section 96, page 102). ↩

  11. There has been loss in Spanish in some cases (correa) and in Romanian (curea). ↩

  12. Also in Southern Italo-Romance (see the note below), despite generally conserving consonant quality, like in Italian. ↩

  13. Catalan. corretja seems to be another case of double pronunciation, but correja is the normal variant; also, platja. (Badia i Margarit 1981: section 85, see below) ↩

  14. According to Rohlfs (2021 [1966]: 213), in old Tuscan texts, it is not rare that words like this are written with "j in the Latin style": joco, judice, jurare, Jovanni. ↩

  15. In Spanish, ha[j]a < FAGEA(M). ↩

  16. In Italian, <z> represents both [ʣ] and [ʦ]. ↩

  17. In ancient times, the sound [z] also emerged from intervocalic sounding, but it disappeared after as a consequence of rhotacism, turned into r. The corresponding letter (a predecessor of <z>) disappeared from the Latin alphabet in the 4th century B.C. (Cockdurn, S.d.: 97) ↩

  18. In turn, the ss of the variant -issare of the verb suffix -ίζειν reflects a dialectal pronunciation of Greek from Italy (-ίσσειν), according to Cockdurn (S.d.: 100). ↩

  19. Also in achora, allhora. But ancora/hanchora (etymological), aduc/adhuc (etymological), because no Romance language writes h. ↩

  20. But cuesto, cuesso, cuello, with etymological <c>, conserved in French and Romanian. ↩

  21. For the pronunciation of et, aud, ad in a nonisolated position (before a vowel and after a consonant), see "Consonant Assimilations". ↩

  22. For the variants of e, *au, a in a nonisolated position (before a vowel and after a consonant), see "Consonant Assimilations". ↩

  23. It is the option recommended by Vía Neolatina, which appears in the Neolatin dictionary. The user interested in the option can deduce it knowing the Latin etymology or the correspondence from the previous table. ↩

  24. In Italian, some accent marks are optional. And they do not distinguish (neither phonetically) between crociare and edificio. ↩

  25. These Romance languages can preserve subsequente evolutions. And in cases like Italian. in, lingua, the i or u present currently is not a preservation of the Latin vowel, but rather a closure of a previous e or o. In aurecla, sete, etc. Italian does present its regular solution. ↩

  26. There is an alternative, proposed by Francisco Piquer: lingua, mùsca, în, cûn (<◌̀> in stressed vowels, <◌̂> unstressed). This system has an important advantage: it allows both the majority reading and the Sardinian reading (it is, therefore, an encompassing spelling). However, it would complicate the spelling for all users and it would multiply the use of diacritical signs, things that surely will not please anyone. ↩

  27. Based on the Latin spelling, the cultured Romans tried to reestablish the sound: not pronouncing it was considered, in the Classical age, a sign of poor education and humble origins. The resulting uncertainty, even in learned people, explains that determining the extension of the aspiration of the tongue (arena or harena?) was one of the main worries of the Roman grammarians (Niedermann 1953: 99-100). ↩

  28. Germanic words like English. have, German. haben are not cognates of Neolatin. havere, but of càpere. ↩

  29. Latin already conserved the traditional spelling when, in the final era of the Western Roman Empire, the affricate pronunciation of [tj] became normalized: etiam or Titius, for example, continued to be written when the grammarians already considered the letter by letter pronunciation of those words incorrect (Loporcaro 2011). ↩

  30. Sardinian presents a comparable spellings, <tz>: dictzionàriu, giustìtzia. ↩

  31. From Portuguese. zebra (Old Portuguese. ezebro, ezevro, ezevra), cognate of Spanish. cebra (Old Spanish. zebra, ezebra, enzebra, ezebrera) and Old Catalan. atzebra, “prob. issu [...] d'un lat. vulg. *eciferus, du lat. equiferus « cheval sauvage », comp. de equus « cheval » et ferus « sauvage ». Les Portugais ont appliqué le nom d'une espèce d'âne sauvage abondamment répandue dans la Péninsule ibérique jusqu'au XVIe s., aut zèbre, animal sauvage qu'ils découvraient en Afrique à cette époque.” (Trésor de la Langue Française informatisé). ↩

  32. It is also the case of other compound words: jaqué, perqué, pòsqué. But here, writing separately is possible and more convenient: ja qué, per qué, pòs qué. It avoids confusion with other words ending in -que that are not compounds resulting from que and, therefore, are pronounced [-kwe]: cinque, donque, qualonque, neque. ↩