Romance Language Lexical Similarity Calculator

Romance language lexical similarity: PT-ES 89%, PT-IT 80%, PT-FR 75%, PT-RO 73%

All the Romance languages trace back to Vulgar Latin, so they carry a lot of cognates in common. Measured against Portuguese, the Ethnologue-style lexical-similarity scores come out as Spanish 89%, Italian 80%, French 75%, Romanian 73%, with Catalan around 85%. The score is just shared cognates ÷ total compared lemmas, so 0.89 means roughly 89% of basic vocabulary has a cognate you can recognize on sight (think noite/noche/notte/nuit/noapte). Be careful, though: a high lexical score does not mean speakers actually understand each other. French pronunciation and Romanian's Slavic loanwords pull real comprehension well below what the number suggests.

Applications

Picking up a second Romance language in roughly half the usual time. Building comprehensible-input reading ladders such as PT→ES→IT. Generating cross-language flashcards, mapping out multilingual travel routes, or sizing up how much dubbing and translation work it takes to move between sister languages.

FAQ

If PT-ES is 89%, why don't speakers understand each other immediately? That missing 11% lands on the most common function words, and the sounds diverge a lot (ç, lh, nh vs. ll, ñ). Spoken comprehension usually trails the written number by 20 to 30 points.

Are false cognates a big risk? They are. Learners get caught all the time by pairs like exquisito (ES "exquisite") vs. esquisito (PT "weird"), or embarazada (ES "pregnant") vs. embaraçada (PT "embarrassed").

Which is closest to Latin? Sardinian and Italian hold on to the most Latin features. French has drifted the furthest in terms of sound.

Does Catalan help with both Spanish and French? It does. Both geographically and linguistically it sits between the two, sharing about 85% with Spanish and about 80% with French.