A new earliest European record of Songhay

In 1713, the memorably named Jezreel Jones wrote a letter describing the Shilha Berber language of southern Morocco, giving an extensive vocabulary, which was published in John Chamberlayne's Oratio dominica in diversas omnium fere gentium linguas versa et propriis cujusque linguae characteribus expressa (1715). The Shilha vocabulary in this letter has already been analysed in detail by Stroomer (2001), and need not detain us here. However, one passage – not hitherto discussed – gives two phrases in quite a different language:

Nigri ex Regnô Tombotoo in Barbariam venientes intellegunt aliarum Nigritianae partium incolas ut Ni-mootii Gooma est quomodo vales frater : & say-borokoy, est in bonâ sâlute, gratias ago tibi: (source)

Stroomer renders this (with his suggestions in square brackets) as:

Black people coming to Barbaria from the kingdom of Tombutoo [Tomboctoo] can understand the inhabitants of other parts of Nigritia. E.g. nimotii gooma [T? nnɛmat i gʷma "well-being to my brother" (?)] means something like "How are you, brother?" and say-borokoy means "(I am) in good health, thank you".

Stroomer's Shilha suggestion for the former phrase is evidently unsatisfactory even to himself, and he ventures no suggestion for the second one. Given the context, we would expect this phrase to be not in Shilha, but in a language spoken at Timbuktu.

In Timbuktu Songhay (or Koyra Chiini, as its speakers call it), mote means "how?" specifically in greetings, while ni is "you (sg.)" (Heath 1998). The greeting ni mote, literally "you how?" is recorded verbatim in Hacquard & Dupuis (1897:92), and the appropriate reply is given as saabu Yerkoy, "thank God". Obviously this is the content of the phrases above. I'm not sure how to interpret "gooma" – possibly it was a switch into Shilha (gʷma); the Timbuktu Songhay word for "brother" is rather harme.

The earliest credible European vocabulary previously known for any Songhay variety is Denham & Clapperton (1826). Lyon (1821:153) gives, as the "Language of Tembuctoo", a menagerie of Tuareg words and unidentifiable forms with only a handful of Songhay terms scattered among them: the latter, oddly enough, often appear closer to Gao or Zarma than to Timbuktu Songhay. These include Meat – Taasoo (taasu "grain"), Small – Katch (Gao kačč-u), Flesh – Hamo (Gao ham-oo), Come – Ka (kaa), Nipples – Foffi (Gao faf-ey "breasts", Zarma fòfè), Go – Dodi (Timbuktu doodi "there"). The Narrative of Robert Adams (1816) gives a purported vocabulary of "the language of Tombuctoo" with at most one Songhay word (if "Gold – Or" is taken to reflect Songhay wuraa rather than French or); it seems to be a farrago of Arabic and half-understood Manding, understandably given the circumstances of his arrival. Jezreel Jones' two phrases predate these sources by more than a century, making them probably the first European record of Songhay.

However, while for many African languages that would be synonymous with "the first record of it", in Songhay's case that is far from true: the earliest attested words of Songhay are to be found in Arabic tomb inscriptions of the 13th century (Moraes Farias 2004), and occasional Songhay expressions are scattered through the Tārīkh al-Fattāsh and other pre-colonial local works. Nevertheless, particularly pending study of the Timbuktu manuscripts in Songhay, sources like these cast a welcome light on the language's history. Timbuktu Songhay is strictly SVO, whereas the mainstream Songhay varieties spoken downriver are all SOV. The expression saabu Yerkoy, with verb-object order, demonstrates that this divergence predates 1713.