The puzzle of the extra pronouns

Like most (all?) Songhay languages, Kwarandzie has two sets of 3rd person pronouns: in this case, they are a (sg.) / i (pl.) vs. ana / ini. In southern Songhay (eg Koyra Chiini, the longer set are used as logophors - that is, used to refer back to the speaker in reported speech. This is not the case in Kwarandzie, though.

ana/ini are obligatory in pre-sentential topic and focus position (including when followed by a preposition), while a/i are obligatory for possessors:

ana (*a) a e-kka. ghi "it's him that hit me."
ini (*i) i-bbey ibbagen "them, they know tales."
an (*anan) kembi "his hand"

But in normal object position, either set can occur:

e-ggwa / e-ggwana "he saw him"

After much checking, I still have no idea what factors drive the selection of one or the other in this position. These are not used, Algonquian-style, for tracking two distinct referents: ²e-gga.r.ana ²e-kka.r.a "I found him and hit him" can as easily refer to hitting the same person as to two different objects.

So it's not logophoricity (much less reflexives), it's not an obviative or a switch-reference system, it's not related to politeness or gender... can anyone think of another possibility for me to check before I leave?

Songhay words in El Jadida, Morocco

Bulbul sent me a link I just had to post about: the article describes, among other things, a secret language used by the Gnaoua, descendants of West Africans brought to Morocco as slaves in precolonial times, in El Jadida, Morocco (on the Atlantic coast.) The author makes no attempt to seek an etymology for the words recorded, but a lot of them are immediately obvious to me - as Songhay. Thus:

* sindi "sommeil": Songhay cindi "rest"
* kuy barkuy "on s'en va": Songhay koy "person", koy "go"
* katihari "...apporter de l'eau": Songhay kati hari "bring water!"
* noro "money": Songhay nooru
* dangi bamatcin "tais-toi": Songhay dangey "be silent", ciine "speak"

Significantly, these words do not display any characteristics that would link them with Kwarandzie. To the contrary - noro and hari are unambiguously Southern, not Northern, Songhay in form, and most of the other words haven't survived in Kwarandzie.

A few words are clearly non-Songhay, and as such harder for me to identify, but these include some Bambara words:
* sgho "viande" - Bambara sogo
* dominika "nourriture" - Bambara dumuni ke

Elsewhere, I've read of Hausa words showing up in Moroccan Gnaoua music (I don't have the reference handy here in Tabelbala). The various sources of the vocabulary attest to the wide geographical range from which slaves were brought, and it's interesting that the words were preserved at all. I look forward to finding out where the other words come from... any ideas?

Metathesis everywhere

When two sounds exchange their positions (for example, clip > plik) we call it metathesis. In most languages, this doesn't seem particularly common, neither in historical changes nor in the grammar. Kwarandzie has no grammatically caused metathesis, but nonetheless is absolutely full of historically metathesised words, sometimes even coexisting with non-metathesised variants. Thus for palm spines, some speakers say taqaneft and others tanaqeft; "forget" is dnagh for some speakers, dghan for others; "irrigation channel" is variously qentret or qetrent... I've found tens of examples where either synchronic variation or transparent external comparison demonstrates metathesis (usually of non-adjacent consonants, though there are one or two cases with vowels, not counting standard North African schwa alternations), and hear new ones every couple of days. Does this remind anyone of anything they've seen, or is it just odd?