Theoretische Linguistik

Links und Funktionen



Current aspects of Preference Theory

Current aspects of Preference Theory
A symposium on the occasion of Theo Vennemann's 80th birthday

Saturday, December 9th 2017
Ludwig-Maximilians-Universität München
Amalienstr. 52, Raum K 001


10:00 Welcome
10:10 Guido Seiler
Is language change language improvement?
10:45 Dietmar Zaefferer
Survival of the most preferred.
A pilot study on the creation and selection of neologisms
11:20 coffee break
11:50 Stephen Laker & Robert Mailhammer
Allgemeine Prinzipien des Lautwandels und die Präferenzgesetze
12:25 Phil Hoole
Towards phonetic explanations for preferred sound patterns
13:00 lunch break
15:00 Manfred Krifka
Vowel Fronting in Daakie (Ambrym, Vanuatu)
15:35 Antoniy Dimitrov
Präferenztheorie und Strategien zur Optimierung subpräferabler Laute und Lautverbindungen
16:10 coffee break
16:40 Laura Catharine Smith
Moving from syllables to feet in the prosodic hierarchy: how foot-based templates reflect prosodic preferences
17:15 Patricia Noel
Preference theory and syntactic order
17:50 Closing remarks
18:00 End

Attendance is free of charge, but registration is requested by December 7 under


Patrizia Noel, Preference theory and syntactic order

According to the principle of natural serialization, „everything else being equal, having unidirectional serialization is preferred to not having unidirectional serialization“ (Vennemann 1983: 12). Ideally, all constituent operators are placed on the same side of their operands/heads, that is, [Operator [Operand]] in OV languages and [Operand [Operator]] in VO languages. In this paper, unidirectional serialization is extended to the ordering principles of elements in Wackernagel position and to the sentence-final particles in the „mirrored Wackernagel position“.

Vennemann, Theo (1983), Causality in language change: Theories of linguistic preferences as a basis for linguistic explanations. Folia Linguistica Historica 6, 5–26.

Survival of the most preferred
A pilot study on the creation and selection of neologisms

Dietmar Zaefferer

LMU Munich
Theoretical Linguistics and MCMP

Arguably, the most fundamental question of linguistic theorizing is: Why are human languages the way they are? This leads, in view of the obvious answer 'Because they became to be that way' to the next question: How did this happen? What are the forces that determined their prehistory and history up to their present stages? Whereas the evolution of full-fledged human languages from more primitive communication systems is bound to be highly speculative, the ongoing dynamics of languages is open to scientific investigation. Similarities with biological evolution have already pointed out by Darwin, but the helpfulness of this kind of analogies is still a hotly debated issue.
The pilot study to be presented here proves that at least in the area of neologisms the analogy with biological artificial and natural selection fosters theory building in that it leads to falsifiable predictions about the fate of names for new phenomena. It does so by interpreting fitness as preference: Given a set of competing labels for a new phenomenon, the one that survives, i.e. is accepted be the community, is the most preferred one.
This leads to the following research question: What are the factors that determine the preference ranking of competing labels in a community and hence the selection of the survivor? In order to answer this question a three step procedure has been devised:
A. Preparation: Nine subjects were asked to write down up to three labels for each one of five objects they had never seen before. Results showed different degrees of naming convergence: Morpheme sharing between subjects ranged from zero to nine.
B. Pretest: A different group of five subjects was presented with the same five objects, one at a time. Everybody (1) wrote down up to three labels, (2) presented them to the group, contributing thus to a pool and (3) selected from this common pool (a) the three personally preferred ones and (b) the three labels that presumably a larger community would prefer. From the results predictions were derived about the properties of the most preferred competitors.
C. Experiment: A third group of six subjects was presented with the same set of objects and went through the same steps as the previous group except that in step (3.b) the group had to agree on a common ranking of the first three items. Finally, the agreed-on rankings were compared with the predicted ones.
The results are presented and discussed, and prospects for integrating evolutionary thinking with preference theory in linguistic explanation are outlined.

Vowel Fronting in Daakie (Ambrm, Vanuatu)

Manfred Krifka
Leibniz-Zentrum Allgemeine Sprachwissenschaft & Humboldt-Universität zu Berlin

The oceanic languages of Vanuatu are known for their rich vowel systems; only few of them have continued the five vowel system of Proto Oceanic (cf. François 2005 on Northern Vanuatu languages). In my talk I will report on the sound system of Daakie, a language spoken in South Ambrym, Vanuatu, based on extensive linguistic field work. The language features probably eight phonemic vowels and a phonemic length contrast. In addition, short back vowels can occur fronted in certain phonological contexts, leading to distinct realizations of three vowels. In my talk I describe the system, compare it to neighbouring languages and motivate the fronting, which appears to be triggered by apical and, to a lesser extent, labial consonants, in a similar way as has been discussed by Harrington e.a. (2011) for German.

François, & Alexandre. 2005. Unraveling the History of the Vowels of Seventeen Northern Vanuatu Languages. Oceanic Linguistics 44: 443-504.
Harrington, Jonathan et al. 2011. The physiological, acoustic, and perceptual basis of high back vowel fronting: Evidence from German tense and lax vowels. Journal of Phonetics 39: 121-131.

Towards phonetic explanations for preferred sound patterns

Philip Hoole, IPS Munich

As part of our overall interest in the phonetic forces shaping the structure of sound systems we have in recent year carried out extensive investigations of consonant clusters, examining in particular the contention of Chitoran et al. (2002) that preferred clusters may represent a good compromise between parallel transmission of segmental information, i.e large overlap (efficient for the speaker) and clear modulation of the acoustic signal (efficient for the listener). We will focus here in particular on clusters involving obstruent plus nasal, lateral, or rhotic, since these are well known to differ widely in their diachronic stability (Vennemann, 2000). Articulatory measurements of German and French speakers were supplemented by articulatory synthesis to simulate the aerodynamic conditions in these clusters. In addition, we will present an articulatory analysis of syllabic consonants in Slovak, since the patterns of gestural coordination provide insight both into why these sounds are overall dispreferred but also why they may nonetheless become well-established in specific languages.

Bombien et al. (2013). Articulatory coordination in word-initial clusters of German. J. Phonetics 41

Chitoran et al., (2002), Gestural overlap and recoverability: Articulatory evidence from Georgian. In Gussenhoven et al. (eds.) Laboratory Phonology 7

Hoole et al. (2013). Articulatory coordination in obstruent-sonorant clusters and syllabic consonants: data and modelling. In Spreafico & Vietti (eds.), Proceedings of Ratics3, Bolzano University Press

Pouplier & BeÁuš (2011). On the phonetic status of syllabic consonants: Evidence from Slovak. J. Laboratory Phonology 2

Vennemann (2000). Triple-cluster reduction in Germanic: Etymology without sound laws? Historische Sprachforschung (Historical Linguistics) 113

Is language change language improvement?

Guido Seiler (LMU)

The idea that language change is meliorative (i.e. local structural optimization, cf. Vennemann 1988, 1993) has become a standard assumption of theoretical historical linguistics. We will try to challenge this assumption, asking: Are there convincing examples of language change that is non-meliorative or even dysfunctional in the light of generally accepted crosslinguistic preferences? And if so, how can they be explained? I will first propose how preferences can be integrated into current models of language change (based on cultural evolution). I will then briefly discuss examples of phonological, morphological, syntactic and semantic change that (seem to) run counter to the intuition of local structural optimization. The examples have in common that they are representative of a type of language change that is indeed difficult to account for in terms of optimization: loss of functional motivation.

Präferenztheorie und Strategien zur Optimierung subpräferabler Laute und Lautverbindungen

Antoniy Dimitrov

Durch die Unterschiede in den Phoneminventaren und Silbenstrukturen zwischen Sprachen (aber auch innerhalb einer Sprache, also etwa zwischen den Dialekten oder zwischen neueren und älteren Sprachstufen) entsteht bei Entlehnungsprozessen ein Adaptionsdilemma: Zum einen sollte die Lautung der Ausgangssprache möglichst genau wiedergegeben werden, zum anderen sollten aber die Regeln der Zielsprache nicht zu stark verletzt werden. In einigen Fällen bedient man sich zur Überwindung dieses Dilemmas gewisser im Rahmen des Systems der Zielsprache wenig beliebter (subpräferierter / subpräferabler) Strukturen (Lautkombinationen, Silbenschemata). Dieses Verfahren ist notwendig, wenn ein Laut der Ausgangssprache in der Zielsprache gar nicht vorkommt und somit kein Laut als einfacher Ersatz dienen kann. Insbesondere gilt das bei einer Asymmetrie der Phoneminventare, also wenn einer vokalarmen Sprache eine vokalreiche gegenübersteht, wie beispielsweise Arabisch-Ungarisch oder Bulgarisch-Deutsch. In diesem Falle wäre eine die Wiedergabe verschiedener Phoneme durch ein einziges eine suboptimale Lösung, weil dies zu Unklarheiten und (bei einer großen Anzahl von Entlehnungen) zu unerwünschten Homonymen führt.
Anhand von Beispielen deutscher Lehnwörter im Bulgarischen, ergänzt durch innersprachliche (interdialektale) Entlehnungsbeispiele, werden die Anpassungsvorgänge dargestellt. Dieser Mechanismus kann man sowohl „lautgesetzlich“ als auch „präferenztheoretisch“ beschreiben, also entweder durch ein System mit festen Regeln oder durch eines mit abgestuften Präferenzen. Letzteres bietet ein schlankeres Erklärungsschema, also eines mit zahlenmäßig weniger Zusatzannahmen.
Die so gewonnenen Erkenntnisse lassen sich auch für die Untersuchung von Entlehnungen in anderen Sprachen sowie bei lautgeschichtlichen Untersuchungen innerhalb einer Sprache anwenden.

Moving from syllables to feet in the prosodic hierarchy: How foot-based templates reflect prosodic preferences

Laura Catharine Smith, Brigham Young University

Although syllables had long figured into explanations of sound changes and phonological patterns in terms of identifying the phonetic environment of those changes, it was not until Murray and Vennemann’s (1983) seminal article followed by Vennemann’s (1988) treatise on the Preference Laws for Syllables that scholars began to see the syllable as a motivating factor guiding sound change. This shift in thinking stemmed from the ability of the Preference Laws to show how phenomena which had been viewed up to that point as divergent and unrelated were remarkably more similar across the world’s languages than previously thought. By drawing attention to the role of prosody via syllables, the Preference Laws invited scholars to explore the role of prosody in sound change, not only in terms of the phonetic environment as had been the case, but indeed as a motivating force behind the changes themselves.
In this paper, I shift my view up the prosodic hierarchy from syllables to feet formed by sequences of syllables to demonstrate how the foot has also been shown in more recent times to drive sound change and shape lexical and phonological patterns. In particular, I draw on data from the history of the West Germanic languages in which the trochaic foot played a role in the loss of the high vowel i following heavy stems in Old High German jan-verbs (e.g., hoorita → hoorta ‘heard’, 1st and 3rd sg.) and Old Saxon i-stem nouns (e.g., gastigast ‘guest’) in contrast to the maintenance of i after light stems (e.g, OHG denita ‘lengthened, 1st and 3rd sg; OS wini ‘friend’). In terms of foot structure, when the high vowel was footed such as with the light stems, then it was retained; when unfooted as it was following the heavy stems, then it was lost. A similar analysis will be shown to account for the reason why high vowels remained full (footed) or were reduced (unfooted) in Old Frisian vowel harmony. In all three of these historical cases, the trochaic foot is the common factor.
Data from Modern German including plural formation is also presented to demonstrate a continued role for the trochee in modern times. As sound changes impacted the endings of the plurals in Old and Middle High German, the plurals came to be reinterpreted in terms of the bisyllabic trochee in more modern times. The ability of the foot to shape these patterns is shown to be a psychological reality and not just a linguistic representation. Indeed, as will be shown, an overwhelming preference for the trochee as an organizing unit for plural formation is evidenced by a recent study (Champenois, Smith and Schuhmann, 2016) examining how native speakers as well as advanced learners of German form the plurals of nonsense words.
Since one of the bedrocks of the Preference Laws is that these laws need to be able to account for not just the data in one language or language family, the paper highlights before concluding how the foot has been shown to shape the development of non-Germanic languages as well, including e.g., Mixtec. The paper is intended to invite other scholars to consider what role prosody may play in terms of not only the syllable but also sequences of these syllables in terms of the foot in helping account for sound changes and lexical developments in a broader range of languages and language families.