In January 2020, I begin a three-year position as a British Academy Postdoctoral Fellow at SOAS University of London. In my research project, I will be studying complex motion predicates in languages around the world. You can see the full description of the project at this link. For a brief introduction, just keep reading!
Let’s go eat! means “go and then eat”. It cannot mean “go while eating”. Why should that be the case? For English, the answer might seem obvious: it’s the order of the verbs. First go and then eat. The complication is that not all languages look like English. Consider this sentence from an Australian language called Kayardild:
(The dugong) is coming behind from the south, coming up to graze.
Word by word, the speaker says graze and then go up. However, the meaning is that first the animal goes up, and then it grazes. So the word order and the order of the actions are reversed. Reversed word order for this kind of sentence is very rare. But there is another type of complication that is much more common.
From the south (they) run down now.
In this sentence, word by word, the speaker says go down and then run. However, the meaning is not that they first go down and then run. Neither does it mean that they ran and then went down. Here, the running and going down are simultaneous. They are running in a downwards direction. They are going down by running.
The above are just a few examples of complex motion predicates. Let’s go eat is an example of a complex motion predicate. It is a predicate—its meaning describes the action or state of the subject. It is a complex predicate, because it is composed of at least two different words: go and eat. And it is a complex motion predicate because it requires a motion verb like go, and it describes movement from one place to another. (Notice that you’ll never say Let’s sit eat, for example.)
There are many different kinds of complex motion predicates. One aspect of my project will be to summarize what kinds of complex motion predicates have been documented by linguists. After studying what kinds of complex motion predicates exist, I will attempt to devise a semantic system that explains both how and why we find certain kinds of complex motion predicates all over the world, but not other logical possibilities.
In a sense, this project is a new perspective on an old question: What do all languages share in common? Specifically, do all languages have the same kind of semantic system for describing motion? Or are languages fundamentally different in the way they divide up and process motion meaning?
If we can create one semantic model that explains every language with no exceptions, that allows us to assume that the mechanism for semantic composition is something that humans intuitively know. It does not have to be learned. It is either hardwired into the brain, or is a natural consequence of the way our minds work.
If, instead, it appears that speakers of different languages use fundamentally different ways of composing motion meaning in complex predicates, that suggests that when we learn a language, we also have to learn the system for semantic composition. This would be a significant challenge to a basic assumption in linguistics: that the mechanisms for composing meaning do not differ significantly from one language to another.
The Kayardild examples are from: