Several kinds of systematic deviations from the Greek original, including simple insertions and omissions of subject pronouns and transformations of nonfinite or impersonal Greek constructions into personal finite clauses, provide evidence concerning the distribution of null and overt referential subject pronouns in Gothic. While the evidence leaves no doubt that Gothic was a null-subject language, it also reveals a tendency, not found in Ancient Greek, toward the use of overt subjects for nontopic antecedents. This Gothic pattern is reminiscent of what a number of researchers have found recently in some other null-subject languages such as Italian, but Gothic appears to occupy an intermediate position between Ancient Greek and Italian.*

In the mid 1870s, Friedrich Kluge spent four semesters in Leipzig studying comparative linguistics and classical philology under the Neogrammarians Georg Curtius, Friedrich Zarncke, Hermann Osthoff, and Wilhelm Braune. Although Kluge applied and refined their methods for many years to come, he was even more deeply impressed by another teacher in Leipzig, the non-Neogrammarian Rudolf Hildebrand. Within the context of Kluge's famous Etymologisches Wörterbuch, of which Kluge oversaw the first (1881–1883) through the tenth (1924) editions, this article asks: In what way did Hildebrand's research come to bear on Kluge's own work? And to what extent did Kluge acknowledge Hildebrand's influence? The answer to the first question entails a fundamental shift in Kluge's view on the goal of etymology, while the response to the second one, which must also be pursued in other writings by Kluge, can be interpreted further by examining certain statements made by Kluge in the dictionary.

During the Northwest Germanic period, *ɣaβiɣs ‘rich’ (Go. gabigs / gabeigs ‘rich’ < giban ‘to give’) and related forms (Go. gabigjan ‘to make rich’, gabignan ‘to be rich’) were reanalyzed as consisting of the prefix *ɣa- + root. This reanalysis was triggered by the prevailing Germanic stress pattern (indicated where necessary by a raised tick '), since *ɣaβiɣs was stressed on the first syllable of the root (thus *'ɣaβiɣs). while nominal and adjectival compounds that consisted of *ɣa- + root ('ga-qumps ‘assembly’, 'ga-fulgins ‘hidden’, 'ga-hails ‘hale’) were stressed on the prefix. Thus, an extrapolated root form *ɣiγ- > *biγ- was created that then existed parallel to *γaβiγs in Northwest Germanic. The newly created form *biγs survived and developed into ME big(ge) ‘strong, sturdy’ (> Eng. big), while OE gifig, ON g˛fugr ‘noble’, and related forms in Old High German and Gothic are reflexes of *ɣaβiɣs.

A survey of the literature on Old Saxon derivational suffixes and suffixoids does not yield a neat and clear-cut picture of forms. The reason is that some researchers take a diachronic view of the data, while others take a synchronic view. In this article I recommend a panchronic view of the data in hopes of clarifying the situation. As part of the discussion I argue that affixoids are a good example of grammaticalization. I thus also discuss and—as far as research on German is concerned—aim to resurrect the notion of AFFIXOID as a linguistic category. If successful, the results of this paper may be applied to other languages.

Hopper and Thompson's (1980) seminal article on transitivity has brought forth renewed interest in the passive and other correlates of transitivity. Langacker (1982) and others working within the Cognitive Grammar framework argue that the passive voice is an independent construction and that it is not a reorganization of the active voice. This paper builds on this view of the passive and presents an analysis of the processual passive in Old Saxon, using the theory of Cognitive Grammar. This semantic analysis focuses on the retention of the dative case in the passive and the similarity in event structure in events encoded by the active or passive voice.

One of the enduring cruxes in Afrikaans historical linguistics has been the origin of the so-called “double” or more properly “brace” negation, specifically with respect to the negative particle nie in sentence-final position. Though bipartite negation is well represented in the Germanic languages, the Afrikaans pattern stands alone. The brace negation is an innovation that came about through the reanalysis of a discourse-dependent (pragmatically conditioned) structure in metropolitan Dutch. The agents of the change were Khoikhoi and enslaved peoples at the Cape in the context of wholesale language shift and basilectalization under the pressure of a socioeconomic order based on caste. Given the intensive mixing between mesolectal and basilectal varieties as part of a shared repertoire, the innovation was accepted by rural, lower-class Europeans living in closest proximity to indigenes and slaves, with stylistic and social variation.