Taalportaal (or Language Portal) is an interactive knowledge base about Dutch, Frisian and Afrikaans. It provides access to a comprehensive and authoritative scientific grammar for these three languages.

Up to now there has been no comprehensive scientifically-based description of the grammars of Dutch, Frisian and Afrikaans. This is a serious shortcoming, considering that

  • language is seen as an important part of cultural identity and cultural heritage
  • a large number of people learn these languages as a second language
  • educated speakers frequently lack grammatical knowledge of their native language
  • Dutch and Afrikaans an important object of study in linguistic theory and related fields of research

Taalportaal fills this gap by providing a thorough description of the phonology, morphology and syntax of the three languages.


The aim of Taalportaal is to collect existing information about the grammars of Dutch, Frisian and Afrikaans and to make it available through the internet in a way which is both accessible and scientifically viable. Such a language portal is unique in the world. Three core domains traditionally distinguished in linguistics – phonology, morphology and syntax – have been integrated into one portal using extensive cross referencing to ensure optimal linking. This offers interesting opportunities for linguists to discover relations and connections between linguistic phenomena that have remained hidden until now. Taalportaal is, however, equally important to people dealing with language and grammar in a more practical way, such as teachers, language advisors and authors developing teaching materials.

Taalportaal is a wide-ranging project involving many institutions in The Netherlands and South Africa. The Dutch part was funded by NWO (until 2015), the Netherlands Organisation for Scientific Research, while the Afrikaans part is sponsored by the Suid-Afrikaans Akademie vir Wetenskap en Kuns, Afrikaanse Taal- en Kultuurvereniging, Noordwes-Universiteit and Dagbreek Trust, until the end of 2019.


Taalportaal is a collaborative effort of the Meertens Instituut, Leiden University, the Instituut voor Nederlandse Taal, the Fryske Akademy, and the Virtuele Instituut Vir Afrikaans. In addition, researchers from a number of universities and institutes, both from the Netherlands and abroad, are involved.

The Taalunie (Dutch Language Union) and the Royal Dutch Academy of Sciences have agreed to guarantee the preservation and the curation of the Language Portal after its completion. Languages change continuously and new insights about the linguistic properties of Dutch, Frisian and Afrikaans will arise from ongoing linguistic research. Therefore Taalportaal will be extended and updated regularly.

Collaboration with CLARIN

Taalportaal has been developed in close collaboration with the CLARIN-NL project.

From May until October 2015, a CLARIN-NL funded pilot project (NL-15-001:TPC) was run with the aim of enriching parts of the Taalportal content with manually constructed xpath queries to treebanks and CELEX. This project was carried out by a team consisting of Jan Odijk (Utrecht University), Ton van der Wouden (Meertens Institute), Marjo van Koppen (Utrecht University), Gosse Bouma (Groningen University), Matje van de Kamp (Taalmonsters) and Frank Landsbergen (INT), together with students from the universities of Utrecht (Sylvia Wijsman, Jorik van Engeland & Erlinde Meertens) and Groningen (Martin Kroon, Mara van der Ploeg, Iris Scholten & Denise Zijlstra). The project was presented at the annual Clarin conference in Wrocław, Poland in 2015.

The main idea behind this pilot project is that the user, while reading a grammatical description or studying a linguistic example, is offered the possibility to find both potential examples and counterexamples of the pertinent constructions in a range of annotated corpora, as well as in a lexical database containing a wealth of morphophonological data on Dutch. This pilot has explicitly focused on resources with rich linguistic annotations, since the goal was to do more than just string searches: searching for construction types and linguistic annotations themselves is one way to reduce the problem of the massive ambiguity of natural language words.

In light of the restricted resources, in terms both of time and money, this CLARIN project is not aiming at exhaustivity, that is, not all grammatical descriptions and not all examples have query links. TPC is thus explicitly to be seen as a pilot project, aiming for a proof of concept by showing the feasibility of efficient coupling of grammatical information with queries in a number of corpora.

Manually added xpath queries can be recognized by the clickable book icon.

About this release

Current release: version 2.1 (May 2020)

Previous releases:
  • version 1.8 (March 2020)
  • version 1.7 (May 2019)
  • version 1.6 (November 2018)
  • version 1.5 (March 2018)
  • version 1.4 (September 2017)
  • version 1.3 (March 2017)
  • version 1.2 (June 2016)
  • version 1.1 (January 2016)
  • version 1.0 (December 2015)