Continuing our series of interviews with OASIS 3 / SPE 12 invited speakers, today we're talking to Professor Isabelle Roy (pronounced /ʁwa/), a linguist working on syntax and the interfaces with semantics at Nantes Université. 

OASIS: Tell us a little about your interest in meaning.

IR: My interest in meaning goes back to when I was a child. I asked adults around me, "Who invented French?" French being my native language. The thing that puzzled me most was that everybody that spoke the same language knew which word to use to designate the same things. Who decided that a table would be called une table and a chair would be called une chaise and how did everyone know that so that they could understand each other? I was about six or seven. Nobody could answer that question. The best answer I got was that "Nobody invented French, it came from Latin." So I asked, "So who invented Latin?" And then nobody could answer the question. And I realized, you cannot trust adults, they are just ignorant! Now I am one of these ignorant adults, and I am trying to understand this.

At the same age, I once heard someone in the streets of Paris shouting at a dog in German. And I thought, "But this poor dog can't understand, because dogs speak French."

OASIS: Your PhD thesis in six words.

IR: How to predicate with nouns, adjectives.

OASIS: What was your most recent "aha" moment? 

I've been working on identity sentences, which are problematic. They seem at first to represent an identity relation, which would be symmetrical. But there's a mismatch between this and their asymmetric syntax. I'm interested in this because there's a mismatch between what we want to express and how we express it. Talking to philosophers recently, and in particular in the framework of mental file theory, we should really distinguish between two operations. One is just updating the file with new information, and this is predication. Another one is putting two files together - you have one file and another one, and then you suddenly realize that these two files relate to the same individual. 

Updating the file is the easy thing to do - you learn new information and add it to the file. It also corresponds to the simple structures in predication cross-linguistically. The other operation seems to be a more complicated file operation, and interestingly, it also corresponds to more complicated structures across languages. So my aha moment was that the conceptual operations really do relate more closely to the syntax than I had previously thought. [Professor Roy will be talking about these in her invited talk in the identity session at OASIS 3 / SPE 12.]

Another very nice moment recently was when I realized that Wolof (a Niger-Congo language spoken mainly in West Africa) exhibited the same three-way distinction among non-verbal predicates that I had identified in four other languages in my book - French, Spanish, Russian, and Irish. That was exciting, especially in relation to nouns and how they behave as predicates cross-linguistically.

OASIS: What's your favorite ontological entity and why?

IR: I want to say predicates or properties, although they are not necessarily ontological primitives; properties may be. ‍

By the way, one thing I'd like to say is that the grammatical ontology doesn't necessarily match what we understand as the conceptual ontology. Take for instance simple event nominals such as party or meeting. They behave grammatically like they're not picking out events - for instance, they can't take argument structure or aspectual modifiers, so you can't say *the meeting by the students for hours, even though you can say the students met for hours. But conceptually, a meeting can of course be described as an event, for instance it happens or occurs, and we can say the meeting lasted for hours.  

So predicates belong to the grammatical ontology, but it is at least in my view an open question whether properties are strictly conceptual, or whether they can have a grammatical primitive representation that is not itself a predicate.

OASIS: Thanks for joining us for this interview. We look forward to your talk at OASIS 3 / SPE 12.

IR: Thank you, and I really look forward to the interdisciplinary discussions that will undoubtedly emerge from this dual conference.