Laptot, Signare.

I’m editing a book on language in Africa, and a chapter on Senegal has introduced me to two obscure French terms:

Signare “was the name for the Mulatto French-African women of the island of Gorée in French Senegal during the 18th and 19th centuries”; according to Wikipédia, it’s from Portuguese senhora, which makes sense but which I wouldn’t have guessed.

Laptots “were African colonial troops in the service of France between 1750 and the early 1900s. The term laptot probably derives from the word lappato bi in the Wolof language, referring to interpreters, intermediaries or brokers.” If anyone can explain how lappato bi works in Wolof, I will of course be grateful; I don’t have a Wolof dictionary.


  1. The Dictionnaire wolof-français et français-wolof by Jean-Léopold DIOUF says (via GB) that lapto and its variant làppatoo mean ‘interpreter’ (as a verb, ‘to serve as an interpreter’). Bi is the most common of Wolof’s ten or so definite articles; they are postpositive and canonically joined to their nouns with a hyphen. The grave on a is a purely conventional and normally redundant marker that a geminated consonant follows.

    Wolof is unusual for putting its TAM markers on personal pronouns, not on the verb.

  2. Interesting, thanks!

  3. David Eddyshaw says

    “Wolof is unusual for putting its TAM markers on personal pronouns, not on the verb.”

    Hausa is the same.

  4. In Russian, лопота́ть (lopotat’) is to mumble, which also applies to clear speech, but in unknown for the listener language. Of course, it is onomatopoetic and informal.

  5. Phil Jennings says

    But Hausa is Afroasiatic and Wolof is Niger-Congo. Somebody wandered from a sprachbund? Not that I even know what a TAM marker is. Ask me where I put my TAM markers and I’m likely to say the middle room in the basement.

  6. @Phil Jennings: Tense-aspect-mood.

  7. David Eddyshaw says

    Certainly much of West Africa is a Sprachbund, but the core Hausa-speaking areas are pretty remote from Senegal.

    However, it’s very common in West Africa for tense/aspect/mood to be marked primarily by clitic particles intervening between subject and verb, from where it’s not too difficult to get the particles cliticising to pronoun subjects rather than the verb; subsequent sound changes can then give rise to portmanteau forms of pronouns which mark TAM. This seems to be the accepted hypothesis for Hausa, anyhow, in which a good many of the pronoun forms remain fairly easily segmentable, particularly in some dialects.

  8. marie-lucie says

    David, Exactly! And this does not have to be restricted to Africa.

  9. Trond Engen says

    I was thinking that this would develop more readly in an isolating language, with little or no pre-existing semantic load on variation of the verb. Wikipedia kindly confirms this for Wolof. But what about inherited Niger-Congo traits that might have prevented it? Did the branch that would eventually produce Wolof lose all PNC noun class morphology on the verb, or replace it with post-positive determiners, ot did it split off before the prefigating Niger-Congo system was fully developed?

    What about other isolating languages? English has some verbal inflection left, but clitics ranging from well-established “-ll” through the accepted colloquialism “-guna” to deprecated “-ma” do show the development, How is Chinese?

    The Irish inflected prepositions must be the same thing but with prepositions and not pronouns happening to be in front of the clitics. Another parallel is that Wolof has Initial mutations, and I think that helps explain some of the pronominal forms. (The pronominal paradigm looks like a fun exercise in internal reconstruction.)

  10. David Eddyshaw says

    Noun class agreement on verbs isn’t at all common in West Africa. I think it’s pretty much a feature of the subfamily that Bantu is a twig of.

    Much the most typical pattern in West Africa is for verbs to show no agreement even in number, let alone noun class, and not even with subjects, let alone objects. This is true for example of Fulfulde, despite a wonderfully exuberant system of noun classes with full-fledged agreement of personal pronouns. I don’t know of any Gur language with agreement on verbs; in a lot of other West African Niger-Congo languages the matter doesn’t arise because the classes have been lost as an agreement system, or probably never existed at all in the first place, as with Mande and Dogon.

    I know little about Bantu (and a lot is there to be known) but it seems overwhelmingly more likely that that group has secondarily agglutinated pronouns to verbs than that agreement was a feature of the protolanguage, lost without trace everywhere else.

    Or as Fela Kuti puts it so memorably: “No agreement today, no agreement tomorrow …”

  11. An informative comment with excellent use of musical reference! (Fela Kuti, “No Agreement.”)

  12. Trond Engen says

    Thanks. I knew that the full-fledged system of Bantu was a local development, but I thought that the cognacy of a rudimentary system was more or less a diagnostic feature of NC. But maybe the indeclineable verb, the noun classsz, and the modifying particles prone to cliticism are those rudiments.

  13. David Eddyshaw says

    Having incautiously invoked Dogon, I must forthwith admit that Dogon verbs do agree with subjects in person and number …

    There’s a fair bit of Niger-Congo verb morphology which is presumed to be inherited, in the form of suffixes conveying things like causative and applicative senses; the literature uses the somewhat nonstandard term “verbal extensions” for these. A lot of the work seems to be very Bantu-centric, as such extensions are not only very prominent in many Bantu languages but formed so productively as to be quasi-flexional. The supposed “Atlantic” group to which Wolof and Fulfulde both belong tends to star in such accounts too for similar reasons. However, there are a lot of difficulties correlating form and function across subfamilies of Niger-Congo. A number of ‘extensions’ may, in addition, be areal phenomena, turning up even in Chadic and Nilo-Saharan. I read a recent review paper which only quoted one source for all of Gur, a grammar of Moore which in the relevant sections has pretty certainly incorrectly segmented many of the verb stems in question. Still at least it seems that proto-Niger-Congo (so far as one can talk of such an animal) had a fair bit of derivational suffixal morphology in verbs.

    The morphological features usually prayed in aid of the existence of Niger-Congo as an actual genetic group (as opposed to a huge Sprachbund) are the noun class system, which is very much a feature of nominals rather than verbs, along with these verbal extensions, which are a quite separate thing. There is a bit of fairly convincing correlation between form and function across subgroups with the noun classes, like the human-plural ba identifiable in Kusaal as a noun suffix (and pronoun) and in Xhosa as a prefix.

  14. English has some verbal inflection left

    Yes, and the curious thing is that it has developed directly contrary to Watkins’s Law, a typological process by which the 3SG inflection comes to be treated as zero and other person/number suffixes are added to it. For example, the PIE present tense singular forms of ‘be’ are *h₁és-mi, *h₁és-i, *h₁és-ti but in Polish and Persian the 3SG has become the base to which the other endings are added, leading to hast-am, hast-i, hast-Ø and jest-em, jest-eś, jest-Ø respectively. (Watkins identified it first in Old Irish.) In English, of course, 3SG is the only surviving person/number ending, and all the others have become Ø instead.

    The morphological features usually prayed in aid of the existence of Niger-Congo as an actual genetic group

    Well put. Greenberg had a lot of ideas, of which Niger-Congo was the only good one (well, he disposed of “Hamitic” as a group, which is something).

  15. David Marjanović says


    (Well, *h₁és-si on the morphological level, but long consonants weren’t allowed. The ending *-si is ancestral to the Polish .)

  16. Trond Engen says

    Derek Nurse, Sarah Rose and John Hewson: Verbal Categories in Niger-Congo.

    Not exhaustive but very broad in perspective.

    It wasn’t an easy task, though. This paragraph from the introduction starts very well but turns funny in a sad way:

    The second step is that we have taken some liberties with the content of sources. Most were written in the late twentieth century in different conceptual frameworks – generative, functional, “traditional”, descriptive – and they use a range of terminology. We ourselves found it difficult in some cases to unravel it, and so we have presented everything in a more or less unified framework. This makes for easier reading, even though it might upset some of the source authors. We point out where we depart from our sources. We have been in contact with many authors to get their reaction to our procedure and approach. We had hoped to use a more or less unified terminology set out in one chapter but in the end this proved beyond us. While all three authors shared many theoretical assumptions, we disagreed strongly on others, and we could not arrive at a unified approach nor a unified terminology. The chapters by Hewson use one set of terminology (explained in the material after Chapter 22), those by Rose use another set (explained in her Brief Note on Terminology), and those by Nurse use a third set (more or less explained in Chapter 1).

  17. That is hilarious!

  18. marie-lucie says

    The second half of 20C linguistics.

  19. Trond, Hat, Marie-Lucie: What’s even funnier is that Sarah Rose did her M.A. and PhD. at Memorial University, i.e. at the very University where both John Hewson and Derek Nurse work (Both are now Emeriti professors), so that her using her own terminology cannot be blamed on her background/academic training.

    It makes sense, in a twisted sort of way…if doctors make the worst patients, then it stands to reason that linguists have more trouble communicating than most, and it thus elegantly, indeed almost inevitably, follows that communication involving multiple linguists must be even more difficult.

    (Insert the “I don’t understand” interruption from a linguist somewhere here, followed by the laugh track).

    I think they do deserve brownie points, however, for being explicit about their different terminological choices: all too often these differences are NOT stated explicitly, yielding a great deal of frustration for readers.

    For instance, a scholar working on pidgin and creole languages observed in print once that scholars studying pidgins and creoles had real difficulty understanding one another because of different terminological choices and scholarly philosophies, and that from this point of view there was an amusing contrast between the creators/users of pidgin languages, who created and used a new language to successfully communicate across a language barrier, and the scholars who study these languages and their speakers, who cannot even make their meaning clear to other scholars working on pidgins, including some they actually share an L1 with.

    To a degree, of course, this happens to any field of study over time: Stanislaw Lem, in his wonderful novel SOLARIS (About a vast intelligent planet-spawning Ocean-like being which humans seek to communicate with), made his (imaginary) history of Solaristics very believable by making it splinter into various squabbling academic sects over time, leading to the protagonist’s mentor making the very same observation to a group of Solarists: how can you hope to communicate with the Ocean when you cannot communicate among yourselves?

    Hmm, if anybody out there wants an original research project, here’s my suggestion: a comparative study of A) The time and conditions required for a single language to turn into mutually unintelligible languages, and of B) The time and conditions required for a single academic field of study to turn into multiple subfields, with mutually unintelligible specialized vocabulary and incompatible goals and presuppositions…There might be an interesting parallel to make between the acceleration of linguistic differentiation as a result of a desire to highlight a separate social identity, on the one hand, and the acceleration of the differentiation of a field into separate subfields as a result of grant money being showered upon anything which can market itself as “new, bold and paradigm-shattering”…

  20. Trond Engen says

    I think they do deserve brownie points, however, for being explicit about their different terminological choices: all too often these differences are NOT stated explicitly, yielding a great deal of frustration for readers.

    Very much agreed. And the attempt to reorganize all these disparate descriptions after a common formula earns them a good deal of credit in my book. But it is hilarious that they set out to unite everything and had to leave their own differences for the readers to solve.

  21. There’s an academic ‘principle of plenitude’, that between any two specialties one may interpose a third. Paradigms may break, but they more commonly interfoliate.

  22. David Marjanović says

    and those by Nurse use a third set (more or less explained in Chapter 1).



  1. […] Hat notes two obscure words of Senegalese French, “laptot” and “signare”. What do […]

Speak Your Mind