Nederlandse Taalkunde - Volume 25, Issue 2-3, 2020

Laboratory sentences in the era of quantitative linguistics

This paper briefly compares an early (1996) and a later (2010) article that I published in Nederlandse Taalkunde, concentrating on the different methodologies underlying the two papers: A formal theoretical in-depth analysis of one idiolect in the first paper and a large-scale quantitative and theoretical approach in the second. It is argued that the two methods make different contributions and are both necessary to develop a theory of (im-)possible syntactic structures in natural language.

This article discusses the intuitionist approach to the study of syntax: the study of the internal structure of phrases/sentences with the help of data obtained by introspection. Some critics claim that this approach is inadequate (if not obsolete) on the assumption that introspection data are not empirical data and are therefore inherently inferior to corpus data based on ‘real language’. Fortunately, not all linguists who promote the use of corpora are of this opinion: Odijk (2020), for instance, stresses that data from corpora and data collected in artificial experimental settings (including introspection) should all be considered empirical data: all relevant evidence should be taken into account, and no form or source of evidence has a privileged status. Although I agree with this statement in principle, this article will argue that there are reasons for assuming that introspection research is a better method for collecting synchronic syntactic data than corpus research.

In this paper, I focus on strengths and weaknesses of corpus research in linguistics, and how it relates to intuition.

Frequency and arm-chair linguistics. A Frisian linguist looks back on Dutch Linguistics and on himself

In this contribution I focus on a paper written by Maarten Lemmens on progressive constructions, which appeared in Dutch Linguistics (2015). This paper illustrates the usefulness of frequency in linguistic research convincingly. Additionally I discuss the tension within generative grammar between vision on the one hand and a methodologically adequate treatment of data on the other hand. Finally, I look back on my own activities as a linguist, presenting some personal thoughts about the question what the usefulness is of mediocrity.

Expect the unexpected. About the different ways in which the Dutch and the Flemish use er

In this paper I report experimental data I have ignored for almost ten years because I did not know what to do with them. I still don’t know exactly, but the 25^th anniversary of this journal is a key occasion to share them, if only because they fittingly illustrate that the syntax of Netherlandic and Belgian Dutch are diverging. The new data demonstrate that contrary to categorical Netherlandic intuitions that er is ungrammatical in sentences with a fronted locative (such as Op de envelop zat een postzegel), er is not only grammatical, but also beneficial to the Dutch language user. Compared to Belgian Dutch, however, the principal beneficiary of er’s processing advantage no longer is the subject noun phrase, but anything that follows.

The acquisition of [PART+AUX] and [AUX+PART] word order in Flanders. A descriptive, methodological and theoretical addition toMeyer & Weerman (2016)

This paper presents new data on the acquisition of verb clusters in Flemish children. The data were collected by means of a sentence repetition task and the results are in line with the development path for verb clusters in Dutch children as proposed by Meyer & Weerman (2016). While Flemish children also show a development from more 2-1 orders in the youngest group to more 1-2 orders in the older group, this development seems to happen more slowly in Flemish children than in Dutch children. In spite of the fact that the results of both the Flemish and the Dutch study refute an analysis that takes the input adult language as the main factor in verb cluster formation in children, the Flemish data suggest that the higher frequency of 2-1 orders in the Flemish context could help to explain why 1-2 orders are acquired more slowly in Flemish children than Dutch children. In addition, this paper also discusses the results of a production test in Flanders that shows a high preference for 2-1 orders until the age of 7, thus questioning the type of linguistic skills that are assessed in a sentence repetition task.

First-constituent-deletion in Dutch. What topic drop is and what it is not

This paper discusses the phenomenon of pronoun deletion in Dutch. In the position before the finite verb a 3^rd person pronoun may be deleted. The deletion of the pronoun is constrained by the recoverability condition, which requires that its referential features can be reconstructed from the context. It will be argued that only the deletion of a d(emonstrative)-pronoun is ‘topic drop’, which is typical for spoken Dutch. Deleted topic d-pronouns are subject to the same syntactic conditions as overt topic d-pronouns (Van Kampen 2010). Like the overt topic d-pronoun, the deleted d-pronoun refers to the focus constituent of the preceding sentence. A deleted p(ersonal)-pronoun, by contrast, does not have a uniquely determined antecedent and therefore it cannot be analyzed as discourse topic drop. In written texts, it solely maintains the preceding subject referent. I will further discuss the deletion of 1^st person pronouns and the deletion of d-pronouns in imperatives.

There is still no dictionary of particles, although particle research is on-going

This paper seeks to explain why there is (still) no comprehensive descriptive work of the particles of Dutch. It does so by focusing on a number of particles that appear to go against common wisdom concerning Dutch sentence structure, in particular V2-violations and wide scope particles that are deeply embedded in complex phrases.

This paper discusses a pattern in which a finite verb is preceded by a locative PP which combines with an initial or final occurrence of the adverb dan (‘then’) and which may look like a violation of the V2 constraint. The paper is relevant in the context of how to handle what look like exceptions to the V2 pattern, a point that is sometimes neglected in the formal literature, though Zwart (2008a) draws attention to the importance of also capturing such recalcitrant data. Based on standard arguments for constituency, it will be shown that the adverb, which has a non-temporal reading, and the associated PP form one constituent. Thus, a pattern that might at first appear to be a violation of V2 in fact is shown to be in line with the constraint.

Reconciling with the prepositional object. On the influence ofVandeweghe (2011)

In this paper I argue for the positive aspects of the claim in Vandeweghe (2011) that there are two types of prepositional object: a primary and a secondary type, which have a different relationship to the predicate. The most discussed consequence of this theory is that it allows a simple sentence to contain two prepositional objects provided that they have a different hierarchical status. Here I focus on sentences with only one possible secondary prepositional object and no other objects. These sentences show that a fixed preposition is not a sufficient condition for a PP to be considered a secondary prepositional object; the preposition should also be used in a non-literal, abstract way. When it is difficult to decide whether this is the case, other semantic properties have to be taken into account, such as the reciprocal character of the predicate.

New insights on the prepositional object

An influential article by Hans Broekhuis in Nederlandse Taalkunde 9 (2004) treated PP complements (type He thought of her all day) in Dutch. It contained sharp observations, intriguing hypotheses and a somewhat puzzling classification of the verbs allowing these complements. It provoked a lively discussion, on matters such as the number of possible PP complements in one sentence, the value of distinctive criteria such as an isolation test with en doet dat, the functional status of the preposition. Broekhuis’ generative-style syntactic reasoning at times clashed with approaches inspired by form/content analysis, valency theory and constructional grammar. In all, the original contribution to this journal and the subsequent discussion brought new and interesting insights in this traditionally hard-to-deal-with syntactic category.

The acquisition of word stress in perspective

This paper reflects on the acquisition of Dutch word stress reported in Nederlandse Taalkunde 1 (1996), where I argued that children systematically build up a grammar for word stress that fits a parameter framework without assuming innate knowledge. In the past 25 years this work has been praised and criticized because (a) the theoretical framework changed to Optimality Theory, (b) the proposed stages did not always adequately fit the data, and (c) new evidence from infant speech perception suggested that children know the word stress system before they start speaking. To fully understand how children acquire word stress, the next 25 year requires researchers from various disciplines to join forces to study representations and perception-production processes in tandem, the mechanisms that cause learning, and the interaction of word stress with other linguistic subdomains.

Light and right

The adjectives licht and recht can be used as adverbs of degree; they can also be used to express a negative-polar meaning in combination with a restricted number of verbs. In this squib, we will focus on the negative-polar and collocational properties of licht and recht by means of corpus data. Licht is shown to be associated with futural clauses, often in combination with the verb vergeten ‘to forget’, while recht is associated with cognitive verbs, especially weten ‘to know’, begrijpen ‘to understand’ and verstaan ‘to understand’.

Against linguistic prudishness. About -gate and other libfixes

This study aims to discuss libfixing as a non-morphemic process of word formation. Libfixes are ‘liberated’ elements that originate from the reanalysis of existing words, usually opaque forms or blends. A well-known example of a libfix is -gate from Watergate, whose borrowing and spreading in Dutch has been discussed by Hüning (2000). Among the other examples that are discussed are English -cation as in mancation, Franken- as in Frankenfood and Dutch -naise as in yogonaise and -talië as in Kapitalië. This contribution shows how widespread the process of libfixing is. Moreover, it is claimed that libfixing operates systematically and can therefore be a subject of morphological analysis and theory. In addition, it is shown in this analysis that it is irrelevant whether a new formation is consciously formed or that it is the result of an unconscious productive process. What counts is whether the neologism is acceptable as a word in the language in question. Examples that are discussed in this article come from English and Dutch.

Two early construction-grammatical studies in Dutch Linguistics. Looking back onSchermer-Vermeer (2001) andVerhagen (2003a)

This article looks back on the articles by Schermer-Vermeer (2001) and Verhagen (2003a), on the double object construction and a Dutch equivalent of the English ‘way’-construction, respectively, which can be considered the first two articles to have appeared in Nederlandse Taalkunde that address problems of Dutch grammar from an explicitly constructionist perspective. I illustrate how, on the basis of data from a large web corpus such as NLCOW14, the formal and semantic analyses offered in the two articles can be refined in a number of ways, but I also show that they include hypotheses and ideas that are still most relevant and relate to topical discussions in construction grammar.

A new look at the imperative participle

This article presents an update of Coussé & Oosterhof (2012), which studies a small group of past participles in Dutch used with an imperative meaning. We revisit in particular the need for abstraction in the constructional schemas developed in the original study. Our analysis explores the role of prototypes and exemplars as an alternative to traditional abstractions in the open slots of constructional schemas.

The loss of a personal pronoun. Why they will not be saying hun anymore

The personal pronoun hun ‘them’ meets a lot of criticism in Dutch society, not just from language purists, but from language users in general. This can be attributed to a strong mistrust of the pronoun, given that it is well-known for violating no less than two prescriptive rules, one of which prohibits its use as a subject, and the other its use as a direct object or complement of a preposition. This has resulted in a tendency to avoid the use of this personal pronoun across the board. Despite the fact that hun ‘them’ as a personal pronoun has the advantage of exclusively referring to animate or even human individuals, I argue that it is fighting a losing battle with the other personal pronouns that are used to express third person plural. I conclude that it will withdraw from the competition in order to commit itself entirely to its function as a possessive pronoun ‘their’, in which capacity it is unique.

A subject what needs research. Revisiting DeSchutter & Kloots 2000

De Schutter & Kloots (2000) present an explorative analysis of the use of relative pronouns and relative adverbs in seventeenth-century literary Dutch, when w-forms are gradually replacing d-forms. They argue that the incoming w-forms were informal compared to the older d-forms. In the present paper, however, it is argued that recent historical-sociolinguistic research suggests that the new w-forms represented a change from above, and thus did not index informality. More research into changes in relativization is needed, and some suggestions for future research are given at the end of the paper.

The spine of language change

In his contribution to the 2005 anniversary issue of the journal Nederlandse Taalkunde, Fred Weerman remarked on the famous S-curve underlying language change, and claimed that a good explanation for this pattern is still lacking. We pick up the thread and assess what 15 years of research have clarified about the nature of the curve. We look at two aspects: the onset of the curve (also known as the ‘actuation problem’), and the sigmoid trajectory (known as ‘propagation’). For the actuation problem, we highlight the role of external variables, notably the role of cities in what kind of changes are more likely to occur. Higher urbanization leads to morphological simplification. For the propagation, we investigate the underlying mathematics of the curve, and its conceptual motivation. We argue that the lesser-known probit function is conceptually more insightful than the commonly used logit function, and marginally outperforms the latter as well, when tested on real data. The difference is so small, however, that in actual practice, the logit function, which is mathematically simpler, may continue to be preferred