Beyond Agreement: Theoretical and Experimental Approaches to Syntactic Feature Manipulation in Real Time

Syntax can be viewed as a computational space where a multitude of syntactically relevant features are manipulated in the context of a well-defined structure-building procedure. A prototypical case of featural manipulation is agreement, a pervasive property of natural language at the core of which is a match between morphosyntactic features like person, number, and gender across two syntactic constituents. Agreement served and continues to serve as a primary source of theoretical insights regarding the architecture of syntactic components in the speaker's mind and its interaction with morphology, among other things. In the last couple of decades or so we have also learned about the manner agreement is processed in real time. We know, in particular, that people are prone to making 'agreement attraction' errors while they tend to miss such errors in comprehension ('grammatical illusions'). Studying patterns of these errors in detail gave rise to a number of influential processing models highlighting grammatical factors such as structural distance between the elements involved in the agreement process, on the one hand, and processing factors such as memory encoding and interference when establishing an agreement dependency in real time, and their interplay, on the other.

But agreement is not the only known manifestation of featural manipulation. When element A determines the value of a syntactic feature on B, then we are dealing with the closely related yet not identical phenomenon known as feature assignment. A prototypical instance of feature assignment is a transitive verb like 'hit' assigning Accusative case to its object in a phrase hit the ball where, even the verb itself does not bear a case feature but determines its appearance on the noun. Feature assignment has been treated differently from agreement in the earlier syntactic frameworks but it was unified with agreement in more recent ones including the highly influential Minimalist program. Can we then expect, by analogy with the syntactic thinking, that people will process feature assignment in a manner similar or identical to agreement? Just as in the agreement attraction research, speech and comprehension errors in feature assignment may provide us with a clue. Psycholinguistic investigations of patterns of such errors that are currently underway constitute the new line of research sprouting from the agreement attraction literature. If the answer to the above questions turns out to be positive, we will, among other things, have converging psycholinguistic evidence for their theoretical unification.

In a broader perspective, identifying common processing patterns in different types of featural dependencies will bring us closer to understanding the underlying mechanisms and parsing routines geared specifically at manipulating syntactic features and their interaction with the morphological richness of respective languages. This line of exploration invites further interesting questions regarding real-time processing of other types of syntactic featural interaction known in the syntactic literature, e.g. long-distance agreement, anti-agreement, wh-agreement, and so on.

In light of the above, this Research Topic invites contributions on the topic of featural manipulation in different types of syntactic constructions from theoretical and experimental
perspectives as well as reflections on their processing aspects.