How many dialects were spoken in nineteenth-century Amsterdam? In his Algemeen Nederduitsch en Friesch Dialecticon (1874) Johan Winkler stated, after consultation with Jan ter Gouw, that in 19th-century Amsterdam 19 different dialects could be distinguished. This article investigates whether it is possible to find evidence for this assertion in the surviving language material. For this purpose all language phenomena mentioned in 57 sources up till the mid-twentieth century have been put into a database, with information on the neighbourhood where they were used, and other metadata. The resulting database contains 9000 language phenomena of which around 4000 could be linked to a specific neighbourhood. From this it appeared that the number of 19 dialects mentioned by Winkler and Ter Gouw is an exaggeration: on the basis of the available linguistic information, we can only distinguish 5 of the 19 dialects mentioned by them. Next to these, however, we can distinguish a dialect not mentioned by Winkler and Ter Gouw, that of the higher classes (spoken along the Herengracht and Keizersgracht), and 5 sociolects or technical jargons: the Bargoens of thieves and tramps, the jargons of diamond workers, dock-workers, street musicians and players of bingo. Around 1900 the variation is reduced and the dialects gradually merged into a more or less uniform Amsterdam city dialect, due to mobility of labour.

No abstract

The distribution of linking elements in Dutch nominal compounds seems erratic at first glance. The literature attributes this to the origin of the linking elements as case markers.When the decay of the case system set in, these markers were retained in compounds but gradually lost their meaning. This allowed the case markers to spread to compounds with a paradigmatically incongruous first element and to acquire their new functions as plural ending and prosodic element. This sketch suggests that the decline of the case system was the main factor in the development of the linking elements. We argue against this view.The loss of case does not overtly correlate with the increase of linking elements. Furthermore, already in Middle-Dutch, the linking element had a prosodic function. The loss of case can only partly account for the spread of linking elements. We hypothesize that an equally important factor is the presence of head-initial stress, which also set in motion the decay of the case system. The erratic distribution of the linking elements is the result of an interplay between phonology, morphology and semantics.• 1 inleiding SamenstellingenDe schijnbaar onbegrensde vrijheid om in het Nederlands nominale samenstellingen te vormen spreekt al eeuwen tot de verbeelding. Zo merkt de zeventiende-eeuwse taalliefhebber en predikant Petrus Montanus (1635: 135-136) op dat het Nederlands "tot verwondering toe" ruimte biedt aan samenstellingen en daarmee "booven de Griexe, ende alle andere Taelen des Werlts te verheffen" is.1 Het recept is simpel: neem twee zelfstandige 121De historische ontwikkeling van de tussenklank in Nederlandse nominale samenstellingen naamwoorden, plak ze aan elkaar, en je hebt een nieuw woord. Dit procedé is beschikbaar voor alle woorden in het lexicon. Nieuwvormingen als gordijnbrief en schilderijspreker zijn -gegeven een passende context -gewoon te interpreteren. Bovendien zijn er geen grenzen aan het aantal aaneengeschakelde nomina. Denk maar aan het welbekende hottentottententententoonstellingsterrein. Dit voorbeeld laat tevens zien dat er soms een extra klank tussen de samenstellende delen staat. We spreken van een tent-en-tentoonstelling en een tentoonstelling-s-terrein. Het gebruik van tussenklanken 2 is aan bepaalde regels gebonden. Hoewel er wel degelijk enige vrijheid is (1), zijn de samenstellingen onder (2) onwelgevormd.(1)Spelling-gids, spellings-gids; vis-kom, vissen-kom. (2) *Bomen-stam; *leeuws-deel; *zon-straal.Iedere spreker van het Nederlands voelt wel aan welke tussenklank een samenstelling kan of moet krijgen, maar het blijkt bijzonder problematisch om die intuïties in regels te vangen (Van den Toorn 1982a, 1982bMattens 1984Mattens , 1990. Dit heeft ongetwijfeld te maken met de ontwikkeling die de tussenklank gedurende de eeuwen heeft doorgemaakt. De ontwikkeling van de tussenklankDe hedendaagse tussenklanken, zo verwoordt Booij (1996: 128) het, "zijn historisch gezien geen verbindingsklanken, maar iets anders, in veel gevallen oude naamvalssuffixen". In het Middelned...

The diffusion and cause of procope and prothesis of h in Middle DutchIn this article we examine the geographical distribution of procope and prothesis of h in Dutch charters from the 13th and 14th centuries, and the frequency with which these phenomena occur, while trying to find patterns in the data. We then venture an explanation of why these phenomena originate in the Flemish dialects. Our hypothesis is that procope of h was induced by language contact between Dutch (Flemish) and French, due to the many bilinguals in Flanders. The contact situation triggered an already existing internal tendency by which h was lost or weakened before sonorants and glides to also affect prevocalic h.

Many languages use verbs of perception to express evidentiality. This paper studies the evidential use of the object-oriented perception verb eruitzien 'look' in Dutch. The results of a Twitter corpus study show that, whereas ziet eruit alsof 'looks as if' occasionally comes with an evidential interpretation, the construction ziet ernaar uit 'looks like' predominantly expresses inferential evidentiality. A diachronic investigation shows how the evidential reading of this construction developed from the object-oriented use of the verb, through a stage in which the construction is used to mark a prediction. This predictive reading, which is still available in present-day Dutch, is not evidential. It does not indicate a speaker has supporting evidence for a factual claim, but rather they have evidence for something they expect to become a fact in the near future. Our diachronic study also reveals how another construction, with a subject-oriented verb (eruit) zien