Can Portuguese speakers understand Spanish?
Portuguese speakers often say they can follow Spanish; Spanish speakers often say Portuguese is easy to read and hard to follow. That lopsidedness is real, and it has been measured. This post puts the measurements side by side — two listening studies thirty years apart — and then checks the obvious explanation, that Portuguese speakers simply recognise more of the words. They do not, and that is the interesting part.
What the listening studies found
The large European study. The best numbers come from the MICReLa project at the University of Groningen (Gooskens and colleagues). Young, university-educated listeners heard short texts read aloud in a related language, with some words replaced by a beep, and had to choose the missing word from a list. Their Portuguese was Portugal's and their Spanish was Spain's.
| Listeners | Heard | All listeners | Listeners with minimal exposure |
|---|---|---|---|
| Portuguese | Spanish | 77.4% | 62.0% |
| Spanish | Portuguese | 37.2% | 35.7% |
| Portuguese | Italian | 49.4% | 44.1% |
| Portuguese | French | 34.3% | not measured |
Portuguese listeners placed 77.4% of the missing Spanish words — the highest score for any pair of Romance languages in the study. Spanish listeners hearing Portuguese managed 37.2%, roughly half as well. The "minimal exposure" column is the fairer one: it keeps only listeners who said they had had very little contact with the other language and had never studied it at school. The gap narrows, but Portuguese listeners still understood Spanish far better (62.0% against 35.7%).
The authors suggest about 40% as a rough threshold for two people to hold a conversation each in their own language. Portuguese listeners with almost no exposure to Spanish clear it comfortably; Spanish listeners with almost no exposure to Portuguese fall short of it. Spanish was the easiest language to understand for every group of Romance listeners in the study, at an average of 57.2%, and Portuguese listeners were the best at understanding other languages, at an average of 47.2%.
The older Latin American study. In 1989 John Jensen tested college students in the Americas — native speakers of Portuguese and of Spanish, none of them expert in the other language — who listened to passages and answered questions about them. In the words of the study's own summary, the two languages were "mutually intelligible at a 50 to 60 percent level", and the Portuguese speakers understood Spanish better than the Spanish speakers understood Portuguese. Two different continents, two different test designs, the same direction.
Is it the words? Measured, no
The tempting explanation is vocabulary: perhaps Spanish is full of words a Portuguese speaker knows, and not the other way round. We can test that on the same kind of material both studies used, everyday speech, in the film and television subtitle corpus (OpenSubtitles 2018).
Take the 1,000 most frequent word forms of each language and keep those spelt identically in both — the same letters and the same accents. That set is the same in both directions, by definition: 223 forms for Brazilian Portuguese and Spanish (names, English and interjections removed). What differs is how much of each language's speech those shared forms make up, because a word can be common in one language and rare in the other. Leaving out the false friends (Spanish no is Portuguese "in the", and so on):
| Shared spellings, as a share of… | Against Brazilian Portuguese | Against European Portuguese |
|---|---|---|
| everything said in Portuguese | 24.0% | 25.5% |
| everything said in Spanish | 22.1% | 23.0% |
The two rows are close, and if anything they lean the wrong way: identical spellings make up slightly more of Portuguese speech than of Spanish speech. On this measure a Spanish speaker listening to Portuguese should have the easier time, which is the opposite of what both listening studies found. Shared vocabulary cannot explain the asymmetry. (Counting the false friends as shared, the figures rise to 28.8% and 26.8% — still no gap in Portuguese speakers' favour.)
Identical spelling is a narrow test, and many more words are close enough to guess (noite and noche, muito and mucho). But that kind of resemblance runs both ways too. Which shared spellings mislead, and how the overlap compares with Italian and French, is in Portuguese vs Spanish.
What does explain it
Sound. The MICReLa authors point to the sound systems. Spanish has 5 vowels, clearly pronounced; Portuguese has a larger vowel system, 5 nasal vowels among them, and reduces its unstressed vowels, especially in Portugal, where many nearly disappear. They also cite work by Voigt and Schüppert showing that Portuguese speakers reduce more syllables in spontaneous speech, which makes it hard for a Spanish listener to tell where one word ends and the next begins. A Portuguese listener hearing Spanish gets the opposite: five clear vowels, and fewer syllables swallowed.
Practice. Brazil borders seven Spanish-speaking countries and Portugal borders only Spain, and Spanish is the bigger language, with 519 million native speakers — so a Portuguese speaker is far likelier to have heard Spanish than the reverse. The minimal-exposure figures show that this is not the whole story — the gap survives among people with little contact — but it widens it.
Who is listening. Every figure above is for educated young adults, under test conditions, hearing clear recorded speech. A fast conversation in a bar is harder for everyone.
So can they?
- Portuguese speakers hearing Spanish: mostly yes. In the European study, even listeners with almost no exposure to Spanish placed 62.0% of the missing words.
- Spanish speakers hearing Portuguese: much less — 35.7% for listeners with minimal exposure, below the level the researchers consider workable for conversation.
- Reading, either way: easier than listening — mutual intelligibility between the two is greater written than spoken — because spelling shows the shared words that the sounds disguise.
For a Spanish speaker, then, the work in learning Portuguese is mostly in the ear: the vowels, the nasals and the reduced syllables, plus the false friends. Is Portuguese hard to learn covers it from an English speaker's side, Brazilian vs European Portuguese sets out how differently the two sound, and the same question for Spanish and Italian is answered in can Spanish speakers understand Italian.
The Portuguese course teaches Brazilian Portuguese from its most common words up, and the Portuguese vocabulary test tells you in two minutes how many of them you already recognise.
Sources
- Gooskens, C., van Heuven, V. J., Golubović, J., Schüppert, A., Swarte, F. & Voigt, S. (2018). Mutual intelligibility between closely related languages in Europe. International Journal of Multilingualism 15(2); published online 2017. tandfonline.com, open access PDF. The figures are from its Table 3, the spoken cloze test in the Romance group; the explanation from sound, and the reference to Voigt and Schüppert (2013) on syllable reduction, are from its sections 2.2 and 4.1.2.
- Jensen, J. B. (1989). On the mutual intelligibility of Spanish and Portuguese. Hispania 72(4), 848–852. jstor.org; the summary quoted is the abstract at ERIC. We have not seen the paper's exact score for each direction quoted anywhere we could check, so it is not given here.
- Wikipedia: Comparison of Portuguese and Spanish — "mutual intelligibility is greater when written rather than spoken"; Portuguese phonology; Spanish phonology.
- Lison, P. & Tiedemann, J. (2016). OpenSubtitles2016: Extracting Large Parallel Corpora from Movie and TV Subtitles. Proceedings of LREC 2016. aclanthology.org
- Dave, H. FrequencyWords: frequency lists from OpenSubtitles 2018, content licensed CC BY-SA 4.0. github.com. The shared-spelling shares are computed by
scripts/frequency_extras.py(portuguese_spanish), with the same list of non-words and false friends as the Portuguese vs Spanish post.