The Romance languages are a subfamily of the Italic languages, specifically the descendants of the Vulgar Latin dialects spoken by the common people evolving in different areas after the break-up of the Roman Empire. Latin itself is treated as an Italic but not a Romance language.

The modern Romance languages differ from Classical Latin in a number of fundamental respects:

  • No declensions (except Romanian which has five cases)
  • Only two grammatical genders, rather than the three of Classical Latin (except Romanian, and verbs which are used as nouns in Spanish)
  • Introduction of grammatical articles, based on Latin demonstratives
  • Latin future tense scrapped, and new future and conditional tenses introduced, based on infinitive+present or imperfect tense of habere, fused to form new inflections.
  • Latin synthetic perfect tenses replaced by new compound forms with be or have + past participle (except Portuguese, where the Latin plusquamperfect tense has been retained).

Roughly, from west to east, the Romance variants, or dialects, form a continuum. Portuguese, French, and Romanian are the three extreme deviations, though this does not imply that they are totally distinct. Sardinian is the most isolated and conservative variant. Languedocian Occitan could be tagged as the central "Western Romance by default".

There are many local varieties spoken in the Romance-language countries, and there is no clear differentiation between a 'language' and a 'dialect'. Roughly speaking, those varieties that are definitely separate languages include the main national languages (French, Italian, Spanish, Portuguese, and Romanian), plus Catalan, Occitan (or Provençal), Sardinian, and Rumansh. After that it becomes questionable: is Galician, for example, yet another full language, or is it a variety of Portuguese with strong influence of Spanish? Or rather a language, with Portuguese as a dialect of it (as some argue it is?) Naturally, political and cultural and local pride issues play a role in these debates.

Historically, the first split was between Sardinian and the rest. Then of the rest, the next split was between Romanian in the east, and the others in the west. The third major split was between Italian and the Gallo-Iberian group. This latter then split into a Gallic group, which became French, Occitan, and Rumansh, and an Iberian group which became Spanish, Catalan, and Portuguese. Catalan is considered by many specialists as a transition language between the Gallic group and the Iberian group, since it shares characteristics from both groups (just for an example, among many others: 'fear' is 'medo' in Portuguese, 'miedo' in Spanish, but 'por' in Catalan - compare with 'peur' in French). See also Iberian Romance Languages.

The most spoken Romance language is the Spanish language (Castellano or Español), followed by Portuguese, French and Italian.

Latin and the Romance languages also give rise to numerous constructed languages, both International Auxiliary Languages (well-known examples of which are Esperanto, Ido, Interlingua and Latino sine flexione) and languages created for artistic purposes only (like Brithenig and Wenedyk).

Here is a more detailed listing of languages and dialects:

The classification below is largely based on the analysis provided at The ISO-639-2 code roa is applied by the ISO for any Romance language that does not have its own code. The Ethnologue classification (produced by the SIL International) is at one extreme of linguists, who divide into 'splitters' and 'lumpers'. Ethnologue produce a very detailed classification, which is more precise than many other linguists would accept, but it is valuable as a description of varieties.

The Southern group

  • Sardinian Four versions recognized; all are included in ISO 639-1 code, sc; ISO 639-2 code, srd)
  • Corsican - (SIL Code, COI; ISO 639-1 code, co; ISO 639-2 code, cos)

The Italo-Western group
The Western sub-group
. .Gallo-Iberian division
. . .Ibero-Romance sub-division
. . . .West Iberian section
  • Asturo-Leonese
    • Asturian - (SIL Code, AUB; ISO 639-2 code, ast)
    • Miranda dl Duoro - (SIL Code, MWL; ISO 639-2 code, roa)
  • Castilian
    • Spanish - (SIL Code, SPN; ISO 639-1 code, es; ISO 639-2 code, spa)
    • Spanish, Loreto-Ucayali - (SIL Code, SPQ; ISO 639-2 code, roa)
    • Judaeo-Spanish (Ladino) - (SIL Code, SPJ; ISO 639-2 code, lad)
    • Extremaduran - (SIL Code, EXT; ISO 639-2 code, roa)
    • Caló - (SIL Code, RMR; ISO 639-2 code, roa)
  • Portuguese-Galician
    • Portuguese - (SIL Code, POR; ISO 639-1 code, pt; ISO 639-2 code, por)
    • Galician - (SIL Code, GLN; ISO 639-1 code, gl; ISO 639-2 code, glg)
    • Fala - (SIL Code, FAX; ISO 639-2 code, roa)
. . . .East Iberian section
. . . .Oc section
  • Occitan (langue d'oc) - Six versions recognized; all are included in ISO 639-1 code, oc; ISO 639-2 code, oci) - all are from France
. . .Gallo-Romance sub-division
. . . .Gallo-Rhaetian section
  • Rhaetian
    • Friulian - (SIL Code, FRL; ISO 639-2 code, fur)
    • Ladin - (SIL Code, LLD; ISO 639-2 code, roa)
    • Romansh - (SIL Code, RHE; ISO 639-1 code, rm; ISO 639-2 code, roh)
  • Langues d'Oïl
    • French (langue d'oïl)
      • Standard French - (SIL Code, FRN; ISO 639-1 code, fr; ISO 639-2(B) code, fre; ISO 639-2(T) code, fra)
      • Cajun French - (SIL Code, FRC; ISO 639-2 code, roa)
      • Picard - (SIL Code, PCD; ISO 639-2 code, roa)
      • Zarphatic - (SIL Code, ZRP; ISO 639-2 code, roa) - extinct
    • Franco-Provençal - (SIL Code, FRA; ISO 639-2 code, roa)
. . . .Gallo-Italian section
    • Emilio-Romagnolo - (SIL Code, EML; ISO 639-2 code, roa)
    • Ligurian - (SIL Code, LIJ; ISO 639-2 code, roa)
    • Lombard - (SIL Code, LMO; ISO 639-2 code, roa)
    • Piemontese - (SIL Code, PMS; ISO 639-2 code, roa)
    • Venetian - (SIL Code, VEC; ISO 639-2 code, roa)
. .Pyrenean-Mozarabic division
  • Pyrenean
    • Aragonese - (SIL Code, AXX; ISO 639-1 code, an;ISO 639-2 code, arg)
  • Mozarabic
    • Mozarabic - (SIL Code, MXI; ISO 639-2 code, roa) - Extinct for common speech
The Italo-Dalmatian sub-group
    • Italian - (SIL Code, ITN; ISO 639-1 code, it; ISO 639-2 code, ita)
    • Napoletano-Calabrese - (SIL Code, NPL; ISO 639-2 code, roa)
    • Sicilian - (SIL Code, SCN; ISO 639-2 code, roa)
    • Judeo-Italian - (SIL Code, ITK; ISO 639-2 code, roa)
    • Dalmatian - (SIL Code, DLM; ISO 639-2 code, roa) - extinct in 19th century.
    • Istriot - (SIL Code, IST; ISO 639-2 code, roa)
The Eastern group
    • Romanian - (SIL Code, RUM; ISO 639-1 code, ro; ISO 639-2(B) code, rum; ISO 639-2(T) code, ron) - Includes Daco-Romanian.
Also as Moldovan - (ISO 639-1 code, mo; ISO 639-2 code, mol) Here are some criteria that distinguish subgroups of the Romance languages:

Formation of plurals
Some Romance languages form plurals by adding "s" (derived from Latin accusative case), while others form the plural by changing the final vowel - "o"/"e" to "i", or "a" to "e" (derived from Latin nominative case).

  • Plural in "s": Portuguese, Spanish, Catalan, Occitan, French
  • Vowel change: Italian, Romanian.

Omission of final Latin vowels
Some Romance languages omit the final unstressed vowels from the Latin roots - for example: Latin LUPUS, LUNA become Italian LUPO, LUNA or Spanish LOBO, LUNA but French LOUP, LUNE.
  • Final vowels retained: Portuguese, Spanish, Italian
  • Final vowels retained in feminine only: Catalan, Occitan, Romanian
  • Final vowels dropped: French.

Word for "more"
Some languages use a version of PLUS, others a version of MAGIS.
  • PLUS-derived: French, Italian
  • MAGIS-derived: Portuguese, Spanish, Catalan, Occitan, Romanian

Sedecim vs. Decem-et-sex
In some languages the word for 16 is morphematim "sixteen", like 11-15; in others it is "ten-and-six", like 17-19.
  • Sedecim: Catalan, Occitan, French, Italian, Romanian.
  • Decem-et-sex: Portuguese, Spanish.

To have and to hold
The words "habere" and "tenere" are used differently for "to hold", "to have", "to have (done)", and "there is".

For instance, in French, je tiens, j'ai, j'ai fait, il y a: these are respectively derived from "tenere", "habere", "habere", "habere". Thus "THHH".

  • TTTT: Portuguese (Brazil).
  • TTTH: Portuguese/Galician.
  • TTHH: Spanish, Catalan.
  • THHH: Occitan, French.
  • THHE: Romanian, Italian (E for "Esse", 'to be')

To have or to be
Some languages use "have" as an auxiliary verb to form the perfect forms (e.g. French: passé composé) of all verbs; others use "be" for some verbs, generally those of motion or becoming.
  • "Have" only: Portuguese, Spanish, Romanian.
  • "Have" and "be": Catalan, Occitan, French, Italian.