Concordance (KWIC) & Word Index

A concordance lists every occurrence of a word together with the words around it, lined up so the keyword sits in a column in the middle. It is one of the oldest tools for studying texts — scholars compiled concordances of the Bible by hand in the Middle Ages — and it remains central to corpus linguistics, lexicography, translation and editing. This generator produces keyword-in-context (KWIC) lines from any text, and can also list every word with the lines where it appears.

Searching

Type a word or phrase. With whole words on, “art” does not match “start”. An asterisk at the end matches any ending, so know* finds knows, known and knowledge. For anything more complex, switch on regular expressions: colou?r finds both spellings, and go|went|gone gathers the forms of an irregular verb. Searches ignore capitals unless you switch on case-sensitive matching.

Reading the lines

Each line shows the line number, the text before the match, the match in bold and the text after it, with the context width you choose. Line breaks inside the context are flattened to spaces so every match fits on one row. Scan down the keyword column to see at a glance how a word is used across the whole text.

Sorting to find patterns

Sorting by the word before the keyword groups modifiers and subjects together; sorting by the word after it groups objects and complements. With “truth” as the keyword, sorting by the following word immediately shows whether it is mostly “truth is”, “truth about” or “truth universally”. This is how lexicographers and language teachers find collocations and typical phrases.

The word index

Switch to “Word index” to list every word of at least four letters (adjustable), excluding common function words, with how often it occurs and the lines it appears on. It is a quick way to check how consistently a term is used and a useful starting point for a book index. A finished index groups related terms, adds subentries and uses page numbers; if your text has one page per line, the line numbers are page numbers.

Exporting

Both views download as CSV for a spreadsheet, a corpus tool or a translation memory workflow. The concordance CSV includes every match, even beyond the 1,000 shown on screen.

A worked example

Paste the opening of Pride and Prejudice and search for “truth”: two lines appear, “It is a truth universally acknowledged” and “this truth is so well fixed”, showing Austen returning to the word ironically within a sentence. Switching to the word index lists words such as “acknowledged”, “fortune” and “neighbourhood” with their line numbers.

Private

Texts stay in your browser, which matters for unpublished manuscripts and confidential corpora.

Tips for better searches

Start broad and narrow down: search the stem with a wildcard, sort by the word after it, then switch to a regular expression for the pattern you discover. For terminology checks, search each variant spelling in turn and compare the counts. Keep the context width small when scanning many lines, and widen it when you need full sentences.

Related tools

Count the most frequent words with the Word Frequency Counter, build a glossary of abbreviations with the Acronym Extractor, replace terms consistently with Find and Replace, or pull contact details out of a text with Extract Email Addresses.