This project studied morphological borrowing, i.e. the transfer of grammatical morphemes (inflection, derivation, and function words) from one language to another through language contact. The aim was to test and refine existing claims about borrowability (e.g. vocabulary before grammar; derivational before inflectional morphology) and to reveal further regularities and universal tendencies in this hitherto understudied aspect of contact-induced change. The problem of morphological borrowing was tackled from various angles. One was an in-depth case study of Resígaro (Arawakan), which has extensively borrowed morphological material from the unrelated Bora language (Witotoan), including entire sets of classifiers, number markers, and personal pronouns, but no verbal morphology and only very little vocabulary. This case constitutes an intermediate case between fully mixed languages and casual borrowing. To account for these descriptive facts, we proposed the Principle of Morphosyntactic Subsystem Integrity (PMSI) in language contact, which describes the compartmentalization of native and borrowed morphological material according to tightly integrated morphosyntactic subsystems in language contact. This principle not only applies to Resígaro, but can also explain a wide range of data from cases of massive morphological borrowing that are reported in the literature. Another focus of this project was the study of the distribution of borrowed bound inflectional and derivational markers in corpora. This allows for an assessment of the productivity of borrowed grammatical markers on native stems. If borrowed markers are highly productive with native stems and hardly used on borrowed stems (as in Resígaro), this can be used as evidence for "direct" borrowing of grammatical morphemes, as against a scenario in which borrowed markers enter a language attached to borrowed stems, as part of unanalyzed complex forms, and only spread to native stems at a later stage.