Editing Generalized phrase structure grammar (section)

== Counterarguments ==
Evidence soon emerged, however, that CFGs could not describe all of natural language (with examples in particular from Dutch and Swiss German [[cross-serial dependencies]]<ref>{{cite book |editor1-first=Ger J. |editor1-last=de Haan |editor2-first=Mieke |editor2-last=Trommelen |editor3-first=Wim |editor3-last=Zonneveld |last=Huybregts |first=Riny |chapter=The weak inadequacy of context-free phrase structure grammars |title=Van periferie naar kern |publisher=Foris |date=1984 |pages=81–99 |isbn=978-9067650199 }}</ref><ref>{{cite journal |last1=Shieber |first1=S. M. |title=Evidence against the context-freeness of natural language |journal=Linguistics and Philosophy |date=1985 |volume=8 |issue=3 |pages=333–343|doi=10.1007/BF00630917 |s2cid=222277837 |url=http://nrs.harvard.edu/urn-3:HUL.InstRepos:2026618 |url-access=subscription }}</ref>), and Gazdar, along with most other syntacticians, accepted that natural languages cannot in fact be adequately described by CFGs.<ref>{{cite book |last1=Gazdar |first1=Gerald |title=Natural Language Parsing and Linguistic Theories |volume=35 |pages=69–94 |chapter=Applicability of Indexed Grammars to Natural Languages |date=1988 |doi=10.1007/978-94-009-1337-0_3|series=Studies in Linguistics and Philosophy |isbn=978-1-55608-056-2 }}</ref> As a result, Generalized Phrase Structure Grammar was soon abandoned as a framework for describing natural languages, although CFGs are still used in computing languages. Most of the syntactic innovations of GPSG were subsequently incorporated into [[head-driven phrase structure grammar]].