Henry James

Concordance to
etexts on this web-site

Introduction

by Adrian Dover

This concordance indexes all the etexts currently available on my web-site. It shows, for each word, the total number of occurrences in these texts and then the texts involved (by letter codes) with the separate number of occurrences in each. The codes are arranged in chronological order of publication by James, to help track words appearing and disappearing over time.

So that you don’t have to wait for ever on a slow Internet connection, the concordance is divided into pages by the words’ initial letters (as shown on the menu, left) and all except the most infrequent letters are further subdivided into pages for chunks of about 100-150 words. Even so some of the pages are up to 300K in size so you may have to be patient on a slow link or when the server is busy!

Some of the most frequent English words are excluded from the concordance through use of a stop-word list, as even their numerical distribution is deemed to be of little practical interest.

There is a link to each text from its code letters. Please note however that, because it would make the text files themselves so huge, it has not been possible to link directly to an occurrence of a word in the text. So, when you have followed a link to a text page, you should use your browser’s text search facility to find the actual word (probably ‘Find’ or ‘Find in frame’ on the ‘Edit’ menu or try ‘Ctrl-F’). Make sure your cursor is in the text frame before you search and remember that this search will probably not be able to match special characters such as accented letters and curved apostrophes. I have also to point out now (November 2003) that two of the texts follow their source New York edition’s quirk of inserting a space into many contractions, such as “would n’t”. For indexing in the concordance these spaces are stripped out so that for example the text’s “would n’t” will be counted under the concordance heading “wouldn’t”.

To get back to the concordance after you have checked out an occurrence, click the ‘back’ button on your browser: some browsers require you to do so twice (once for the text and once for the menu) but sensible ones, like Opera, realise that you loaded three frames with one click and ‘back’ them all with one click. Use the ‘concordance’ option on the text’s menu to re-load the concordance menu and this introduction.

As an alternative to checking the contexts with the links here, you could, having identified the text(s) you are interested in, download the relevant ASCII file(s) and work with it/them (one file per title). Another benefit of that method will be that you can search for the special characters by using their ASCII character representations explained on my editorial page covering downloadable texts. It will also enable you to compile a concordance to a single text or a small set of texts, and to do more complex text analysis, such as proximity searching. A suitable program for these tasks has been written by a colleague of mine here at the University of Birmingham, Alan Reed, who also has an excellent set of links about text processing on his site. Other software is available – try a web search for ‘concordance software’.

Within my concordance, there are separate pages for numbers (in figures) appearing in the text and also for James’s fictional names. If you are interested in the latter particularly, don’t forget that you can find further information in published encyclopedias of James’s fiction or characters, such as A Henry James encyclopedia / by Robert L. Gale. - Westport, Conn.; London : Greenwood, 1989. - ISBN 0-313-25846-5; or, Who’s who in Henry James / Glenda Leeming. - London : Elm Tree Books, 1976. - ISBN 0-241-89425-5. Please note that real proper names, for example ‘Byron’ or ‘Paris’, and fictional names not invented by James, for example ‘Hamlet’, are included in the main sequence of the concordance.

Note also that for the time being, words in the stage-directions and character directs in the play-texts are included in the concordance. Another problem you may enounter is that foreign words are not identified separately. I have projects in hand to implement improvements in both these areas once all the texts have been upgraded to a common standard which is amenable to automatic processing.

If you want to look up a lot of words and are not worried about being able to link to the texts from them, you can download a plain-text file, either with or without the text-codes, from the separate page of links, then browse that on your own PC (or even, whisper it not, print it out on reams of paper!)

The pages which make up this concordance are generated by another of my natty, home-grown, Perl scripts and are updated each time an addition is made to the set of texts available here. I’d like to thank Casey Abell and Richard Hathaway for their valuable comments on draft versions of these pages and Alan Reed (software consultant for the Concordance of Medieval Occitan) for useful discussions about online concordances. Anyone requiring details of the processing to support academic use of this concordance is welcome to email me with specific questions or for the algorithm used.

November 2003

this introduction © 2003

this page is written, designed and maintained by
Adrian Dover – latest update 2003-11-14