Edo and Portuguese Creole.

Uwagbale Edward-Ekpu writes for Quartz Africa about the influence of the Edo language on the creoles of the Gulf of Guinea:

Gulf of Guinea creoles are the main Portuguese creole languages still spoken today. There are a few other portuguese creoles spoken by a few thousand people in Malaysia, Sri Lanka, India, Singapore, and Indonesia. Several studies have shown though that the Edo language is the major African component that constitutes the foundation of the creoles of the Gulf of Guinea.

At least one variety of these creoles is spoken in Sao tome and Principe and Equatorial Guinea, with diaspora speakers mainly in Angola and Portugal, according to the Atlas of Pidgin and Creole Language Structures (APiCS), a linguistic atlas that provides expert-based information on 130 grammatical and lexical features of 76 pidgin and creole languages from around the world.

The creoles of the Gulf of Guinea were derived from the combination of the Portuguese language, Edo language (including closely related Edoid languages in the Niger delta), and Bantu languages (mainly Kikongo and Kimbundu), according to linguists. The creoles emerged from a first-contact language or pidgin resulting from the contact between the Portuguese colonizers and the slaves from the kingdom of Benin in Sao Tome. The Bantu languages came in contact with the newly formed Portuguese-Edo language in the island some decades later. […]

In a study, John Ladhams, a Linguistics expert in Portuguese-based pidgin and creole at the University of Westminster, explained that the grammatical process in which a sequence of meaningful word elements are composed has originated in an Edo noun prefix in the creole of the Gulf of Guinea and there is a higher proportion of Edo contribution to other word classes such as adjectives, verbs, and adverbs than nouns.

This, he said, suggests the Edo language played a more important or a more specialized role than the Bantu language in the formation of the creole since the borrowing of adjectives, verbs, and adverbs from one language into another occurs less frequently than the borrowing of nouns.

However, the number of speakers of the Gulf of Guinea creole is “rapidly” declining, especially among the younger generations, according to APiCS. The country of Sao Tome and Príncipe has a population of about 219,000 people. APiCS estimated that the Santome and Angolar varieties of the creole have about 60,000 speakers and 5,000 speakers respectively, on the larger island, Sao Tome.

Thanks, Trevor!


  1. David Eddyshaw says

    Uwagbale Edward-Ekpu links to this as his source:


    It seems a bit unconvincing to me. The argument seems to boil down to “Edo has a lot of vowel-initial nouns, and Portuguese-based creoles with agglutinated articles start with vowels, so this must be Edo influence.”

    (The fact that these Edo initial vowels are relics of Volta-Congo noun class prefixes seems to me to have no bearing on the argument at all. At any rate, they are in no way like articles.)

    [There was I, hoping that this would be about Portuguese creoles in Tokugawa Japan …]

  2. I figured this would lure you in! I also figured it was probably overstated, so thanks for clarifying that.

  3. We discussed the Japanese Edo (well, how to say the name in English) here. I remain of the opinion that “EE-doh” is silly.

  4. I pronounce “Edo” in a thoroughly Americanized way (with the FLEECE vowel in the first syllable), probably in large part because (as mentioned in that 2010 thread) it tends to come up for me in conversation not as a toponym remote in time and space but as the name of a teppanyaki restaurant less than half a mile from my house. By contrast I pronounce “Tokyo” and “Kyoto” in what might be thought an affectedly un-Americanized way.

    But to the point at hand, I take it the creoles are losing market share (in terms of who speaks them fluently) in Sao T & P to regular old standard Portuguese, or perhaps some regional variety thereof that is nonetheless too close to the metropolitan standard to be thought a creole. The expansion in L1 fluency in the former imperial languages under the post-imperial independent regimes seems a common phenomenon in Africa.. Maybe because the colonial gov’t cared less about formal education for the indigenes, maybe because other standardization-promoting phenomena like mass media have arrived on the scene post-independence, maybe something else.

  5. David Eddyshaw says

    There’s a soon-to-be-published grammar of Fa d’Ambô in the Mouton Grammar Library series, which looks interesting, although sadly not £136.50 worth of interesting.

  6. David Eddyshaw says

    To be fairer to Ladhams, I see that all Edo nouns begin with vowels, and that at least a few change the vowel in the plural (like Twi, and unlike the more closely related Yoruba) which does make his thesis a bit more plausible.

    Looking through Rebecca Agheysi’s dictionary, I found an old Volta-Congo friend which illustrates the point: ovbi [òʋì] “child (of)”, plural ivbi (= Kusaal biig etc etc.) Another example is ọmwan “person”, plural emwan. It does seem that only a few nouns have distinct plural forms, though.

    Vowels get added at the front of loanwords too, e.g. etọnni “ton”, etolotolo “turkey.” Yoruba doesn’t do this, e.g. titi “street” (from English), bọrọkinni “gentleman” (from Songhay.)

    So the evidence is better than I thought on the assumption that Edo nouns were just like Yoruba nouns. (I’m still not overwhelmed, though,)

  7. Not sure if this is gratuitous to point out, but the very first sentence of the block quote is completely wrong. Best as I can tell, all sources agree that the “Gulf of Guinea” creoles spoken on Sao Tome and elsewhere have in the aggregate significantly fewer speakers than the “Upper Guinea” creoles spoken on the Cape Verde Islands and in Guinea-Bissau (plus various overspills and diasporas etc.). I think an Edo substrate would be a harder sell in terms of historical-demographic plausibility for those creoles. (FWIW it sounds like in Guinea-Bissau the authorities are happy for the creole to be the de facto national language and everyone’s L2 so there seems to be less shift to fluency in “regular” Portuguese.)

  8. all sources agree that the “Gulf of Guinea” creoles spoken on Sao Tome and elsewhere have in the aggregate significantly fewer speakers than the “Upper Guinea” creoles spoken on the Cape Verde Islands and in Guinea-Bissau

    Thank you for pointing that out! Not gratuitous at all. That was the first thing I wondered when reading this post, but I didn’t try and ferret out the figures and do the calculation.

    I was wondering how great the similarities are between “Upper Guinea” creoles and “Gulf of Guinea” creoles, and how speakers of these two different groups might have interacted, because of the Cape Verdean immigration to São Tomé and Príncipe, as described in this song:

    Who showed you…
    this distant way?
    Who showed you…
    this distant way?
    This way…
    to São Tomé

    Longing, longing,…
    For this land of mine, São Nicolau

    If you write me letter,…
    I will write you back
    If you forget me,…
    I will forget you…
    Until the day…
    You come back

  9. @Xerîb: “…how speakers of these two different groups might have interacted, because of the Cape Verdean immigration to São Tomé and Príncipe…”

    I’ve read somewhere that Cape Verdean Creole (CVC) is now the second most spoken Creole in São Tomé and Príncipe, having surpassed Angolar. Meanwhile, the Principense variety of Forro, Lung’ie, is dead or moribund.

    In “Estudo do léxico do são-tomense com dicionário” by Ferreira Fontes (2007), Edo and Kishikongo are both named as substrate languages. But I couldn’t find a single Edo word in that vocabulary, in contrast to a hundred or so Kongo loans.

    It seems that Forro itself is in decline on the island, while young Santomeans living in Portugal are keen on speaking it among themselves. Neither Forro nor Angolar has any official status in São Tomé, while CVC is due to become official one day, as per the Cabo Verdean constitution.

    CVC has an impressive body of literature, including the lyrics of this song (“Sodade”), made famous by Cesária Évora. There’s an early recording from the 1950s by Amândio Cabral somewhere on YouTube, well worth a listen too.

  10. Where else in the world is there a situation of stable multilingualism in two or more creoles? Suriname (Sranan/Saramaccan)? Where else?

  11. Papua New Guinea, definitely. Coastal West Africa, likely. CVC is also a collection of creoles – each island has its own – but they are close enough to be considered variants of the same.

    On the other hand, is it really multilingualism when different groups living in in the same country speak different creoles?

  12. Papua New Guinea? What creoles other than Tok Pisin?

    From what I understand, Sranan is a lingua franca and Saramaccan is spoken by one ethnic group, so I imagine Saramaccan speakers are at least bilingual in the two.

  13. You’re right, at least some of them are bilingual.

    In Papua New Guinea, on the island of New Britain, Rabaul Creole German (Unserdeutsch) was spoken until recently by people of mixed Papua-German ancestry. Apparently, it emerged when Tok Pisin speakers started re-lexifying their language with German words.

    I’m not sure if Hiri Motu qualifies as a creole – apparently, it’s merely a pidgin and is dying out anyway.

  14. David Marjanović says

    Oh, my comment has disappeared. The impression I get of Hiri Motu is that in its heyday (the mid-20th century) it was something comparable to L2 Hausa – and the police’s own argot, which is too cool for words (ironically).

  15. Oh, my comment has disappeared.

    Weird. There’s no comment of yours in either moderation or the spam folder.

  16. David Eddyshaw says

    the police’s own argot

    He do the police in different languages.

Speak Your Mind