Relativisation in Swahili and Safwa: A Contrastive Analysis of Subject and Object Relative Clauses

Imagine the world of language as a giant, bustling construction site. In this city, every noun (like "girl," "food," or "boys") has a specific "uniform" or badge that tells you what kind of thing it is. In many African languages, these uniforms are called "noun classes." Now, imagine you want to add a little extra detail to a noun, like saying "the girl who is running" instead of just "the girl." This extra bit of information is called a "relative clause."

The big question linguists ask is: How do these languages stick that extra detail onto the noun? Do they glue it right onto the verb (the action word) until it becomes part of the verb's own body? Or do they keep the detail separate, like a distinct tag hanging off the side? This study dives into that question by looking at two languages: Swahili, which is like a famous, well-documented city with blueprints for everything, and Safwa, a smaller, less-explored village nearby. The researchers are using a concept called "grammaticalisation," which is basically watching how loose, separate words slowly get glued together over time to become tight, unbreakable parts of a sentence. They are also using a "cycle" idea, which suggests languages might move through different stages of how they glue these pieces together. Understanding this helps us see how human brains organize complex thoughts and how different cultures solve the same puzzle in unique ways.

The Great Glue-Up: A Tale of Two Languages

This paper is a linguistic detective story comparing how two neighbors, Swahili and Safwa, handle the tricky job of connecting a noun to a description. Think of it like two different ways of building a sandwich. You have your main ingredient (the noun, like "the girl") and you want to add a topping (the description, like "who loves Jesus").

The Swahili Style: The Fusion Sandwich
In Swahili, the researchers found that the language loves to fuse everything together. When a Swahili speaker says "the girl who loves Jesus," the word for "who" doesn't sit out on its own. Instead, it gets baked right into the verb "loves." It's like the verb and the "who" are married; they are inseparable. The "who" becomes a tiny, mandatory piece of the verb's uniform. If you tried to pull it out, the verb would fall apart. The study shows that in Swahili, this "gluing" is so strong that the relative marker (the "who") is locked inside the verb's internal structure, right next to the tense and the subject. It's a high-tech, integrated system where everything is one tight unit.

The Safwa Style: The Modular Sandwich
Now, look at Safwa, the language from the Mbeya region of Tanzania that hasn't been studied as much. Here, the builders do things differently. They still have the same goal: connect the girl to the love. But in Safwa, the "who" marker stays separate. It stands at the very front of the sentence like a distinct flag or a separate tag. It agrees with the noun (so it changes form if the noun is plural or singular), but it doesn't get glued inside the verb. The verb and the "who" remain two distinct blocks that sit next to each other. It's like a modular Lego set where the pieces are clearly visible and can be taken apart, rather than being melted into a single plastic blob.

The Big Discovery
The paper's main finding is that even though Swahili and Safwa are related languages and they both use these noun-class uniforms to make sure the grammar matches up, they have chosen different paths to get there. Swahili has taken the "highly integrated" route, where the relative marker is fully absorbed into the verb. Safwa has taken the "transparent" route, where the marker is still visible and separate.

The authors suggest that these differences might represent different stages in a "cycle" of how languages evolve. Imagine a timeline where languages start with separate words, and over time, they might get glued together tighter and tighter. Swahili seems to be further along in this "gluing" process, while Safwa is still keeping its pieces distinct. However, the researchers are careful to say they aren't proving that Safwa used to be like Swahili or that it will become like Swahili. They are just saying that right now, these two languages show us two different levels of how "glued together" a language can be.

What They Ruled Out
The study explicitly rules out the idea that all Eastern Bantu languages are identical in how they build these sentences. For a long time, linguists might have assumed that because Swahili is so famous, everyone else in the family probably builds sentences the same way. This paper says, "Not so fast!" It proves that even closely related languages can have totally different structural strategies. It also clarifies that while the function is the same (linking a noun to a description), the mechanism is not universal.

How Sure Are They?
The researchers are very confident about the patterns they found in the data they collected. They interviewed five native Safwa speakers and compared 30 pairs of sentences. They are sure that Safwa uses separate markers and Swahili uses integrated ones. However, they are not claiming to have solved the history of these languages. They cannot say for sure when or how Safwa got this way, or if it will change in the future. They are suggesting that their findings fit well with a theory called the "Bantu Relative Agreement Cycle," but they are using that theory as a map to understand the present, not as a time machine to prove the past.

In short, this paper is a celebration of diversity. It shows that even in a family of languages that shares a common ancestor, there is no single "correct" way to build a sentence. Some languages prefer a fused, high-tech approach, while others prefer a clear, modular one. And that's a beautiful thing for anyone curious about how human language works.