A central issue in understanding how language links the mental and the real world is the nature of the mental representations entertained during language processing. Are these mental representations closely linked to the perceptual experiences from which they were formed or are they somewhat removed from them? This review addresses this question by examining studies that have investigated motion verbs and sentences using functional magnetic resonance imaging. These studies tested whether language processing elicits modality-specific brain regions responsive to motion perception. Although the results of these studies are not definite due to the different tasks and analysis techniques utilized, they so far suggest that modality-specific brain regions processing visual motion are not automatically or habitually engaged in language processing. The occasional engagement of visual areas in language processing appears to result from tasks requiring integration of visual and linguistic information or attention to motion-specific features such as direction. The evidence reviewed therefore suggests that although perceptual representations may be flexibly engaged as a function of tasks and contexts, language comprehension in the absence of visual contexts habitually engages experience-based representations of motion events that are one-step removed from visual experiences, even in situations in which imagery is encouraged.

This paper is the first in a series of two papers presenting recent developments concerning the interaction between aspectual classes of predicates and the semantics of aspectual for-phrases. Aspectual for-phrases can felicitously modify stative and activity predicates, but not (basic) accomplishment and achievement predicates. Earlier literature proposed that this is because aspectual for-phrases must modify predicates which are homogeneous– meaning that the predicate spreads appropriately to subintervals – and it proposed a notion of homogeneity which is appropriate for stative predicates. We argue in this first paper that neither the earlier literature, nor later proposals have managed to come up with an adequate account of the felicity of aspectual for-phrases with eventive predicates, and that, in particular, accomplishment and achievement predicates with bare arguments and iterative constructions remain challenges that these accounts cannot properly meet. We show that the problem lies in the notion of homogeneity for eventive predicates: the semantic tradition has provided us with static notions of homogeneity – insensitive to the arrow of time – but what is needed is a dynamic notion.

This paper is the second in a series of two papers presenting recent developments concerning the interaction between aspectual classes of predicates and the semantics of aspectual for-phrases. Aspectual for-phrases must modify predicates which are homogeneous – meaning that the predicate spreads appropriately to subintervals. For eventive predicates the relevant notion of homogeneity should be a dynamic notion – sensitive to the arrow of time. We present a recently developed semantic framework in which eventive predicates are incrementally homogeneous, meaning that the predicate characteristics are preserved for each event from its onset through all incremental development stages. We show how incremental homogeneity deals with the felicity facts concerning aspectual for-phrases, and discuss complex-activity forming operations which turn heterogeneous predicates into homogeneous ones, in particular, as triggered by bare arguments and by iteration.

‘Syntactic microvariation’ and ‘microcomparative syntax’ are the terms for a fairly new research approach that applies the theoretical concepts and techniques of modern generative theory to dialectal and other small-scale variational data. Traditional studies in dialectology aim at a detailed and fine-grained description of language variants; the ultimate goal being a proper classification of the dialects, their exact areal distribution as well as their historical/diachronic development. For formal generative theory on the other hand, the foremost goal is to model the human language faculty with Universal Grammar as a theory about possible human languages (and how they can be acquired). Microcomparative syntax tries to reconcile these two research traditions by applying the formal theoretical concepts of generative grammar to those ‘minor’, ‘peripheral’, and sometimes kind of ‘squishy’ differences between closely related language variants as they are typically found in dialectal data. Research in microvariation tries to offer new concepts that can account for the range and (limits) of inter- and intra-speaker variation in a principled way while at the same time testing existing formal theories against these microvariational data and thus contributing to the theory of language variation. A profound understanding of microvariation will also open a way to a deeper understanding of the mechanisms of language change, given that language change necessarily preconditions variability in the data.