The Even Yearbook 14 (2020)

ELTE SEAS Working Papers in Linguistics, ISSN 2061–490X

authors: den Dikken & O’Neill, Jánosy, Starčević, Őri, G. Kiss & Szigetvári,
editors: Mark Newson & Péter Szigetvári (editorial note)


Marcel den Dikken and Teresa O’Neill
Clausal parataxis — and how!
The Even Yearbook 14: 1–16.   download

This paper presents an outlook on the syntax of Luhya sentences in which the equivalents of ‘how’ and ‘thus’ show subject agreement marking on the adverbial element. While prima facie unusual, we argue that this agreement pattern requires no modification of any standard assumptions regarding the Agree relation. The adverbial is treated as a predicate of a separate clause, linked paratactically to the main clause, with a silent pronoun coindexed with the subject of the main clause. This analysis accounts for the Luhya data, makes novel and accurate predictions, and carries over to a rarely discussed construction in Germanic.

keywords: ‘how’-questions, agreement, parataxis, asyndeton, locative inversion

Márton Jánosy
Stress patterns before the -able suffix
The Even Yearbook 14: 17–25.   download

This paper examines the factors influencing stress-placement before the -able suffix. These include the the status of the stem (free or bound), a full (tertiary-stressed) vowel at the end of a stem, syncope, gliding, and trisyllabic shortening. Based on corpus evidence, I argue that the treatment of -able either as a categorically stress-neutral suffix or as a categorically stress-placing suffix is untenable.

keywords: morphophonology, variable stress, stress-placement, suffixation

Attila Starčević
A very special word of Old English.
The Even Yearbook 14: 27–89.   download

Old English exemplifies a number of rounds of deletion of the high vowels, some of which have been traditionally described as the result of both syncope and apocope, or a combination of these, dependent on syllable weight. In this paper, high vowel deletion (or syncope) and high vowel apocope are both analysed as the result of templatic deletion of *i/u of pre-Old English: the two vowels were deleted if they failed to attach to a CVCV template (word < *wordu ‘words’ vs scipu < *skipu ‘ships’, firen < firenu ‘crimes’ vs nītenu < nītinu ‘animals’). The analysis is couched in CV phonology. One of the words that opens up a window onto this period of Old English is the neuter plural nom/acc hēafudu. The sequence of two high vowels separated by a C (*iCi, *iCu, *uCu, *uCi) seem to have constituted a special phonological environment in which syncopation was suspended. It also seems the dative plural -um could not have contained a high back vowel at the time when syncopation was banned in high V + C + high vowel sequences.

keywords: Old English, high vowel deletion, high vowel apocope, high vowel syncopation, template, dative plural -um, syllable weight

Péter Őri
An alternative laryngeal analysis of languages with two obstruent series.
The Even Yearbook 14: 91–133.   download

Most current laryngeal analyses use two melodic elements (or distinctive features) to create two-way laryngeal contrasts in obstruents: |L| (or [voice]) in voicing languages like Hungarian and |H| (or [spread glottis]) in aspirating languages such as English. In this paper, I argue that |H| is enough to distinguish the two sets of obstruents in both language types and that languages only differ in the phonological processes operating on |H|, with a potential cross-linguistic variation in the phonetic realization of the obstruent series. I also show how binary-contrast systems can be categorized in this alternative analysis.

keywords: laryngeal properties, phonological contrast, phonetic variation, laryngeal assimilation, language typology

Zoltán G. Kiss and Péter Szigetvári
Telling fortis and lenis apart in English obstruent clusters.
The Even Yearbook 14: 135–156.   download

There is an almost unanimous agreement among those concerned that English has clusters like [sp], [ft], [kt] (in, for example, sport, after, packed/act). Based on theoretical considerations, we contend that this is not the case, these words contain [sb], [fd], [kd], and [ɡt], respectively. In this paper we report on an experiment showing that the acoustic properties of the plosive clusters in the strings acting and packed in are different in a statistically significant way. The relative length of the vowel in acting is greater than in packed in, the voicing duration is longer in the cluster in acting, and VOT is larger before the /i/ in acting. This suggests that the former cluster is /ɡt/, while the latter is /kd/.

keywords: English, obstruent clusters, laryngeal specification, fortis/lenis