Trösklar och knutar inom grammatiken: teori och praktik

The aim of this dissertation is to map out and illustrate particularly troublesome and central areas in grammar learning. The point of departure is Josefsson and Lundin’s three suggested threshold concepts within grammar: to be able to distinguish between meaning, form and function, to be able to distinguish between a linear and a hierarchical dimension and to be able to identify and segment clauses and phrases.

I draw on Josefsson and Lundin’s three suggested threshold concepts to design, implement, and evaluate three linguistic inquiries in a grammar course within the teacher education programme specialising in Swedish. The analysis is based on the participants’ written analyses and notes produced during these inquiries.

In the study, I deconstruct Josefsson and Lundin’s threshold concepts into smaller components, which I term knots. In total, 19 knots are identified. I argue that resolving these knots reveals a higher degree of systematicity in grammar and may reduce working memory load, thereby facilitating continued learning. In the subsequent stage of the analysis, I situate the knots within four broader areas of understanding: constituency, recursion, function, and semantic scope. I derive these areas of understanding through an analysis of the types of difficulties represented by the knots. I conclude that two of these areas – constituency and recursion – can be considered threshold concepts. The threshold concepts that I identify are not the same as those proposed by Josefsson and Lundin. In the discussion, I explain the similarities and differences between the threshold concepts I identify and those proposed by Josefsson and Lundin and justify my own.

My proposed threshold concept constituency includes both operational and structural understanding, inspired by Sfard’s theory, originally applied to mathematics. Operational understanding is process-oriented and relies on rules of thumb and partial analyses, whereas structural understanding reflects a more integrated grasp of linguistic structure. I argue that my second proposed threshold concept, recursion, is a prerequisite for developing structural understanding of constituency and thus acts as a bridge between operational and structural insights into constituency.