The volume contains 17 articles, some in English and some in German. Ten of them are revised versions of papers originally read at the conference The Gothic Language: Origins, Structure, Development, held on 18–19 September 2008 at the University of Southern Denmark, Odense. The contributions cover a scope ranging from language affiliation and onomastics to the phonology and the writing system of Gothic. The volume is intended for readers interested in early Germanic languages and particularly the aficionados of Gothic

In a methodically interesting study, Robert Nedoma treats ‘Schrift und Sprache in den ostgermanischen Runeninschriften’, nine runic inscriptions that can be regarded as East Germanic with certainty or with reasonable possibility. The author discusses the inscriptions from six different points of view: (i) the milieu of the find (Fundmilieu), (ii) the context of the find (Fundkontext), (iii) the type of object, (iv) paleography, (v) language form, and (vi) the reference to a non-linguistic reality. Each inscription is finally summarized as to its East Germanic status. The objects treated are the spearheads from Kowel, Dahmsdorf, Rozwadów, and Mos, the spindle whirl of Leţcani, the gold ring of Pietroassa, the clasp from Ménföcsanac, the scabbard mount of Bergakker, and the clasp from Charnay-lès-Chalon.

Of the inscriptions, six are regarded as East Germanic on linguistic grounds: Kowel (Tilarīds), Dahmsdorf (Ran(n)ja), Leţcani (Raginō), Pietroassa (Gutanī ō(þal) wi(h) hailag), Bergakker (Huþuþiws ann kusjam loguns), and Charnay (un(þ)f(i)ndai Idolan Lianō). Of these, however, two inscriptions demand advanced readings. The runes raŋo on Leţcani are read raiŋo with a double mainstaff for | and a partly illegible , whose top part of the two bistaffs has been ruined. Two of the runes in the initial word on Charnay un[þ]f[i]ndai must be considered left out by the carver. Nedoma considers it questionable whether the inscription on Mos is East Germanic. Two of the inscriptions, Rozwadów and Ménföcsanac, can only be placed among the nine due to the milieus and contexts of the two finds, as the inscriptions cannot be interpreted.

Six, perhaps seven, of the inscriptions can be regarded as East Germanic because of the milieu of the find and five, perhaps seven, can belong to the group due to the inscriptions’ language forms, and four due to the type of the object, and one each for the paleography of the inscription and the reference to a non-linguistic reality. Nedoma concludes that, with great probability, Pietroassa is East Gothic, and that Charnay can be considered as late Burgundian. The ‘dialects’ of the rest of the inscriptions are not possible to determine. The article is of great methodological importance.

The onomastic evidence for the Heimat of the Goths is discussed by Jan Paul Strid in ‘The origin of the Goths from a topolinguistic perspective: A short proposal’. The author brings forward the tribal names Gautai, Goutai and Gutones used by Classical authors for the people, and arrives at the conclusion that they were Germanic names *Gautōz and *Gutaniz, closely related but with a variation caused by ablaut. These two are the equivalents of Modern Swedish Götar and Gutar, inhabitants of the provinces of Götaland and Gotland. Through the ɡutani of the Pietroassa ring, a word that corresponds to *gutaniz, we can translate the inscription ‘the sacred heritage of the Gutar’, which is claimed as a possibility that cannot be overlooked. Strid states that the etymologically closely related names Götar, Gutar and Goths can be traced to a people or a folk living in today's South-East Sweden. He then discusses place names with the initial elements Göt-, Göte- and Göta-, and a final element denoting sacral places. Strid claims that it is difficult to form a decided opinion, but he suggests an interdisciplinary approach that combines language history, onomastics, archaeology, geography and cultural history could compensate for the lack of literary sources for the societies of the Götar and Gutar in the early Iron Age.

Three of the articles treat various forms of language contact. Tette Hofstra discusses older views on Gothic loan words in the East Finnish languages in ‘Zur Frage der gotischen Lehnwörter in den ostseefinnischen Sprachen’. He poses three questions (201): (i) Ist es möglich oder gar wahrscheinlich, dass Goten Kontakte zu Ostseefinnen hatten? (ii) Hat vielleicht eine dritte Sprache dem Ostseefinnischen gotische Wörter vermittelt?, and (iii) Gibt es Merkmale, die gotische Originale von anderen germanischen unterschieden könnten? He discusses the questions in five paragraphs treating the possibilities for direct contacts, a third language, phonological evidence for consonants and vowels, and semantics. His conclusion is: ‘Das Vorhandensein gotischer Lehnwörter kann nicht ausgeschlossen werden, aber es ist bisher noch nicht gelungen, die got. Herkunft auch nur eines einzigen osfi. Wörter eindeutlich nachzuweisen’ (212), which is a position the author took already in his dissertation in 1985.

The question of the possible impact language contact had on Gothic word formation is treated by Antje Cassaretto in ‘Evidence for language contact in Gothic’. The author considers whether 44 Gothic adjectives, corresponding to Greek or Latin adjectives, consist of endocentric or exocentric compounds. Summing up his results, Cassaretto finds three: ‘we find three levels of adjectival calques in Gothic’ (233). These are (i) morphologically and semantically unremarkable formations, the main body, (ii) lexicalizations, only some, and (iii) irregularities, only a few (233–234). Cassaretto's final words are that these adjectives give the impression that they were made up ad hoc by the translator and that they were probably never used outside the specific context of the religious Christian language of the Goths.

Valentine Pakis treats the historical present and its possible dependency of Old Latin in ‘Praesens Historicum and the question of Old Latin influence on the Gothic Bible’. Pakis takes his point of departure in the claim that the Old Latin text of the Vetus Latina tradition influenced the text of the Gothic Bible. He states that this influence worked on the lexical level, and he raises the question ‘whether it might have influenced the grammatical choices’ (239). The author asks why the historical present was used so infrequently – if it was to some extent natural in Gothic, and if this tense was unidiomatic, why the translators [sic!] deviated from their exemplar in some cases, but not in others. Pakis presents two tables, one of 10 cases where the Gothic text agrees with the Greek (present tense), the other of 138 cases where it agrees with the Latin text (past tense). The Vetus Latina can thus give an explanation for the regular Gothic deviations from the Greek historical present. This kind of irregularity would then be the result of Translationese.

Carl-Erik Lindberg treats the subjunction þatei in ‘Die Subjunktion þatei nach verba dicendi und sentiendi im Codex Argenteus’. He sets out with stating the limitations of studying a language with only texts translated from Greek, and the difficulties of studying the syntax of the translation and the preserved text in Codex Argenteus. The subjunction often occurs in the four evangelists in subordinate clauses, but they are unevenly attested. The author sets the origin of þatei in connection with the transition from paratax to hypotax as a development of the demonstrative pronoun sa. He studies direct speech both without and with subjunction and sees it as a sign of quotation, corresponding to Greek ὅτι. The second part of the study concerns þatei-clauses as direct objects by verba dicendi and verba sentiendi. The third part treats the tense sequence Preterit + þatei + Present, and Lindberg's conclusion is that the Gothic version mirrors the Greek.

In ‘Got. gards οκοσ und garda αὐλή: semantische Konsistenz und Raumkonzept’, Ludwig Rübekeil treats Gothic gards and garda from a semantic view. The noun gards has different meanings ‘palace; entrance-hall; house, home’, wheras garda has a narrow meaning ‘[sheep]fold’. The author furthermore discusses the compounds ingardja ‘house members’, weinagards ‘vineyard’, aurtigards ‘garden’, midjungards ‘earth’, miþgardawaddjus ‘partition; dividing wall’, gardawaldands ‘master of the household’, and þiudangardi ‘kingdom’. His conclusion is that the denotation of gards is ‘house’, to which also ingardja belong. The denotation of garda is ‘sacred area’. The compounds weinagards and aurtigards do not denote houses but are considered as younger and transparent formations. For þiudangardi there is no plural attested, and the author hesitates whether the meaning adheres to gards, i.e. ‘house of the king’, or to the stem-formation þiudangardi.

In ‘Die Flexion der gotischen Verbalabstrakta vom typus Laiseins’, Luzius Thöny treats the morphology of verbal abstracts to the weak verbs of the first declension, like laiseins ‘doctrine’ to laisjan ‘to teach’, naiteins ‘blasphemy’ to naitjan ‘to verbally abuse’. They show a mixed declension between an i-stem and an ō-stem. In the plural, they are declined as ō-stems in the nominative and the genitive, but as i-stems in the dative and the accusative. The author poses two questions: (i) From where do the unexpected ō-stem declension emanate?, and (ii) Why in just these two cases? (286). After a discussion of the destiny of the suffix -īni- in North and West Germanic, Thöny discusses five different explanations: (i) analogical change from the Genitive Plural of the īn-stems, (ii) a suggestion of old í/jó-stems, (iii) a general confusion of the feminine ī-declension and the ō-declension, (iv) a confusion with the suffix *-inō-, and (v) an analogical formation based on the possessive pronouns or the adjectives in -eins. Thöny concludes that none of the explanations can completely explain the phenomenon, but that the old ī/jō-stems and the confusion of the feminine ī-declension and the ō-declension, must be ruled out, and he suggests that an analogical formation based on the īn-stems in the Genitive but on the ō-stems in the Nominative or a transformation built on the possessive pronouns serves the best explanation, but that both have the weakness that they do not explain why this happened in only two of the cases.

Magnus Snædal discusses the variation between unsis and the shortened form uns for the accusative and the dative of the first person plural weis in ‘Uns/unsis and Colloquial Gothic’. Rhythmical and rhetorical causes have been suggested for this variation by Dickhoff (Reference Dickhoff1913). Through statistical presentations, Snædal shows that there are many exceptions in the Gospels, and he suggests, after a close reading, that the longer form unsis can be assumed to be the more formal of the two variants and the shorter form uns the more colloquial and intimate form, used in the Lord's Prayer and by the evil spirits and the swine heard. For the same variation in the Epistles, Dickhoff (Reference Dickhoff1913) has suggested stylistic and syntactic causes. The author states that the short form is the rule and the long form the exception (38 vs. 5, p. 317). He concludes also here that uns has been used to make the Epistles more intimate or colloquial, a conclusion that is not supported by the rather complex syntax in these texts. Snædal thus gives a stylistic or sociolinguistic explanation to this variation.

In ‘Gothic tweihna- Old English twegen, and some further formations with Gmc. *twi’, Alfred Bammesberger discusses the Germanic root *twi-, which occurs in the Gothic numeral *tweihnai ‘two in number, numbering two, two each; dual’, attested only twice in the Gothic texts. The author is not kind to his possible readers in not translating his examples from Greek, Gothic or other languages.

Patrick V. Stiles discusses the mismatch between Gothic and the other Germanic languages for preterite endings in ‘The Gothic extended forms of the dental preterit endings’, and he questions if the extended forms are an archaism or an innovation (343). The standard explanation for forms like Gothic first person plural tawi-dē-d-um ‘we did; we made’ compared to first person singular tawi-d-a is that they originate from the preterit plural of do in West Germanic. Stiles consideres whether this can be explained by (i) chance, (ii) etymological identity, or (iii) a borrowing of some kind. He states, after a convincing survey of the material, that the Gothic extended forms are isolated within Germanic (356) and concludes, having applied the comparative method, that the extended forms were not original to the dental preterit, but an internal haplology, *(dĕ)dē, *(dĕ)dun, in Gothic.

Three of the articles treat phonological matters. Guus Kroonen's ‘Gothic iup and the Germanic directionals’ takes its starting point in the semantic variation between ‘above’ and ‘under’. The author states that the most important difficulties concerning iup are not of a semantic nature, but of a formal one. Gothic iup contains a full grade of the root whereas the other Germanic languages do not, and, moreover, it has a root-final *p. The question becomes how to explain iup from its base. Kroonen discusses the directional system of Germanic adverbs from a locative, allative and ablative compared to the preposition, i.e. Gothic in ‘in’, us ‘from’ and uf ‘up’. He concludes that the Proto-Germanic forms *inn-, *upp- and *utt- must have developed through a stressed n-suffix, and that the geminates arouse under Kluge's law. He states that ‘the adverbials were created at an extremely early stage, i.e. before the paradigmatic leveling of the full-grade of *h ₂pó, and the subsequent evolution of this case into a preposition’ (376).

In five sections of his interesting article ‘Verner's Law’, Anatoly Liberman discusses various forms of stress (word, sentence, tonal) and stress from Proto-Indo-European to Proto-Germanic before he discusses the nature and cause of Verner's Law or the exceptions to it and their implication for Gothic. Liberman concludes that his views are odd on the lack of word stress in Proto-Germanic, the importance of sentence stress, and the role of ‘double power’ in early Germanic, but that stress cannot be discussed without defining it in functional terms and that this is a task for psychologists and acousticians rather than for linguists; he also concludes that the reification of position and relative chronology produces fiction, that voicing (weakening) by Verner's Law began a process that still continues, and that Verner's Law did not produce allophones, as it operated when voiceless and voiced fricatives were independent phonemes.

In ‘Gothic and Early Runic: Two sound systems compared’, Hans Frede Nielsen compares the sound systems of Gothic and Runic. Regarding the accented vowels, he concludes that the fact that the reflex of late Proto-Germanic ē ² coalesced with ē ¹ in Gothic was of fundamental importance, but that ē ² was retained in North and West Germanic. The monophthongization of /ai/ and /au/ in Gothic occurred, therefore, relatively late, as suggested by the etymological consistency of the orthography of Biblical Gothic. The development of the unaccented vowels is more difficult to account for. The consonant systems of Gothic and Early Runic were basically in agreement. Nielsen concludes that Gothic differs from Early Runic and the later North and East Germanic languages, but that the syllable reduction in Gothic was more advanced than in Early Runic, but similar to the reduction of the syllables in the later North and West Germanic. Gothic had, however, a relatively small number of accented vowels, but a relatively large number of unaccented ones, as in Early Runic, but different from the state of affairs in Old Norse and Old English.

The writing system is treated by Elmar Seebold and Antoaneta Granberg. Seebold discusses the names of the Gothic letters in his 90-page article, ‘Die gotischen Buchstabennamen’, containing a lengthy excursus, 57 pages, on the English manuscript runes. The names used are given in the so-called Wiener-Alquin manuscript (Wien ÖNB 795). In section F, ‘Der Vergleich der Namen’, he compares and treats the names in three different groups: comparisons without problems (d, e, l, n, o, r, t), comparisons that need an explanation (f, g, j, h, m, s, w, u), and the problematic names (a, b, k, þ, z, p/q). For the comparison he uses the Greek letters and the runes, and the Old English and the Old Norse names.

Seebold's problem lies in his trust in the Gothic names supplied by the manuscript. He states that in aza < (*ansu-z) a nasal sign is missing and so is the final s in the expected azas. He raises the question of whether it is a form from another regional language or another word.

In my opinion, the author puts too much trust in the names of the Wiener manuscript. Wolfgang Krause suggests e.g. a reconstruction *ahsa ‘axle’ in his Handbuch des Gotischen (Section 46). A comparison between the names that only have their counterpart in the English names yields daaz ~ dæg, eyz ~ eh, geuua ~ geofu, uuinne ~ wyn, utal ~ oedil, and pertra/quertra ~ peorð/cweorð. If the scribe did not know a Gothic name for a letter, he may easily have supplied it from the Old English name of a rune or from some other source, ezec ~ ilcs /z/, enguz /ʃ/, or uuaer //.

Antoaneta Granberg treats the Gothic writing system in ‘Wulfila's alphabet in the light of neighbouring scripts’. She compares Greek, Coptic, Ethiopic, Gothic, Armenian Georgian, Glagolitic and Cyrillic alphabets. The first three of these were created before Christianization, but ‘are important in order to understand the mechanisms and the results of borrowing alphabet structures and for an understanding of the process of inventing new alphabets, especially in an earlier written tradition’ (169). The author claims that she studies the order and the numerical value of the letters, and the shape of the letters (the layout of the alphabet). Granberg's conclusion is that ‘the Coptic, the Gothic and the Cyrillic alphabets [sic!], which came into existence in the context of an earlier written tradition, do have a layout and numerical value for the letters similar to the Greek alphabet’ (187). It is a bold statement to talk of a written tradition for Gothic, a tradition evidenced by only one inscription. She further claims that ‘[i]t was [on] many occasions the high status of the Greek alphabet that was important for establishing a new alphabet in a culture that was already literate’ (ibid.). Why this should have been so, is never explained. One could, in fact, argue the opposite, that a society without a literary tradition would slavishly follow a system copied.

It is with great satisfaction that I notice that the interest in dead languages is still alive and thriving. A few words of reproach, though, are called for. The editors should have executed a more diligent editorship. There are several mistakes in the bibliographies, authors and their works are missing in quite a few of the articles, and the years given in the texts are sometimes wrong when randomly compared to the list. A few of the articles should have been rewritten or excluded.