Brains "versus" software: new possibilities and limitations of computer assisted historical studies of English syntax

Abstract

Although the article was written fourteen years ago, its main idea is still valid: despite great progress in computer technology, also applied in linguistic studies, successful interpretation of the corpus material can only be performed by the human brain. The problem of identifying multiple meanings has been partially solved by better and better automatic disambiguation techniques. Upgraded search engines can now retrieve words in their original spellings. Many more diachronic English databases have become available since the end of the 20th century. Libraries all over the world have been supplying digitalized facsimiles of more and more medieval manuscripts, which enables scholars to have their own interpretations of ancient texts without referring to the printed ‘emended’ editions. This has already resulted in the publication of several important studies which changed generally accepted views. However, we were too optimistic concerning the Dictionary of Old English project: while the complete DOE Corpus is ready, the compilers of the dictionary itself are currently working on the entries beginning with the letter H. The Third Edition of the Oxford English Dictionary is still rather far from completion. Not only in this respect we have a long way to go…