Some Observations about Zulu (and French!)

In a previous post, I promised a tinfoil-hat crazy theory about French and Zulu. I decided to tone that down a bit, since I don't actually hold crazy beliefs about a relationship between the two. Instead I have a few observations about similar patterns of linguistic behavior, and speculation as to why linguists sometimes treat these similar patterns quite differently. The first half of this post will be a few observations about unexpected (to me) similarities between Zulu and French, and will be slightly but hopefully not too technical. The second half is some musings on why these similar patterns are treated as different things, the history of linguistics, and triumph of modern linguistics.

I've been studying Zulu because (1) Penn has amazing language resources that I would be hard pressed to find elsewhere, including a native speaker of Zulu from kwaZulu Natal who also happens to have a PhD in linguistics, and (2) It's Zulu. why wouldn't I?!

For those unfamiliar with Zulu, it is a Bantu language spoken in South Africa. It's got everything: ejectives, ingressives, clicks, tone -- you name it. It's also got some very interesting syntax, with a lot of stuff all directly attached to the verb. So a given verb might have a root (say, -fund- 'learn') but it will also have subject and object markers; tense, aspect, and (/or) mood markers; applicatives which may change the meaning (e.g., the causative -is- which changes the verb to -fundis- 'teach', there are also benefactives for when you do something for someone's benefit, among others),  passivizing morphology (e.g., 'the ball was hit' as opposed to 'I hit the ball'), reciprocal markers ('we will see each other' the 'each other' is realized on the verb) I said, a lot of stuff.

This means that some things that require sentences in English can be expressed entirely with one verb. For instance:

  • babhaliselwani? 
  • "why are they being made to enroll?"


In general, for simple things, the sentence constituent order is Subject-Object-Verb ("I it eat"), or Subject-Auxiliary-Object-Verb ("I it will eat.").

I was having a very hard time with this, for the first few weeks of class, until I had a realization: most verbal complexes in Zulu have the exact same constituent order as modern colloquial French (that is, not classroom French), which I don't find daunting. Below are some examples (top is Zulu, then French, then a word-for-word gloss for both Zulu and French, and finally a translation --  Zulu is not written with the dashes; those are to clearly separate the constituents for easy comparison. Any mistakes are my own!):

  • ba-phuza-ni aba-ntwana
  • ils boivent quoi, les enfants
  • they drink what the kids
  • "what are the kids drinking?"


  • u-si-letha-ni?
  • il nous apport quoi?
  • he us brings what?


  • i-bhubesi li-dla ezi-nye izi-lwane
  • le lion, il mange les autres animaux
  • the lion, he eats other animals
  • "Lions eat other animals."

This even works for more involved examples:

  • um-fundisi u-ba-bona abantwana
  • le prof, il les voit, les enfants
  • the prof he them sees the children
  • "the professor sees the children"

For the above, French requires special prosody, with pauses. It is also considered a "lower class" and stigmatized way of speaking. That said, it's completely grammatical in the spoken language, and the kind of thing I hear frequently. In fact, I recently heard my uncle, a professor and notorious language stickler say il faut y penser á ça ("one must think about that"), where both y and á ça refer to the same thing. When I asked later about it (after hearing him and others do the same thing multiple times), he flatly denied saying it...before going on to use the same structure later without noticing.

I find the parallel between French and Zulu here striking. So why then, despite having very similar constituent order (in the handful of carefully curated examples I presented here), are these considered completely different things in Zulu and French?

Well, for starters, there's the fact that in French the auxiliaries (vais 'going' in je vais le voir 'I'm going to see him.') behave like their own words. While the order of constituents is the same, in Zulu it behaves like one prosodic word, in French you need at least two "words" as soon as there's an auxiliary. Then there's the fact that Zulu has its own pronouns separate from the markers (mina, ngizokuthi... me, I'll tell you... )  although you could make that argument about French (moije vais te dire...).  There's the fact that when you start adding more and more stuff to the sentence, the parallels kind of break down, especially around negation or where the subjunctive is marked, etc. 

That said, there's also the fact that French and Zulu have been analyzed, historically, by very different kinds of linguists following very different traditions, with very different terminology. And on some level, the people who first described a lot of African languages (in the academic literature) had a vested interest in them being different.  It's a lot harder to justify genocide, extraction of resources, and apartheid when you recognize that the language of the people you're oppressing is remarkably similar to the language you think of as the international language of diplomacy, philosophy, and thought -- precisely how French was imagined in the popular European consciousness in the 1700s and 1800s. To some extent, there's still historical baggage. And to some extent there must be a founder effect, where we still talk about similar things with very different terminology just because the terminology is inherited. Many of the academic books I have checked out to do research on Zulu end up being about Bantu, and they divide up the languages by region and number, so Zulu is in the S "zone" (for Shona, another language, spoken in Zimbabwe), and is numbered 42. So I've read texts that don't even bother treating quite different languages as different, or even worthy of their own names. Having to do research by scouring texts on the language family for mentions of S42 is like if you were interested in French, but all you could get is books on European, and had to just find Romance language number 14 or something.

This points, then, toward one of the great strengths of modern approaches to linguistics. We've made huge leaps in dispassionately describing the facts, and the starting point for most modern theories is that languages are using the same structural tools -- with slightly different parameter settings -- to do the same kinds of things. An analysis that can describe French and Zulu with the same conceptual toolkit is now privileged over one that posits some fundamental, irreconcilable difference. Gone are the days of describing Bantu languages as the cursed languages of the descendants of Ham (I'm not kidding).

To be fair, though, I'm also not claiming a real connection here, at least in terms of historical contact or anything 'external.' To the extent that there's a connection, it's the shared conceptual architecture that's available to all humans, and that French and Zulu make use of in similar ways. Seeing such similar manifestations across such obviously unrelated languages is one of the payoffs of studying linguistics. Studying (modern) linguistics is like learning to see the Matrix. Suddenly, things that seemed very complex and wildly different reduce to minor variations on the same highly structured conceptual architecture, and that, for me, is one of the real payoffs of studying linguistics.



©Taylor Jones 2016

Have a question or comment? Share your thoughts below!