.th SPELL I 5/31/77 .sh NAME spell \*- find spelling errors .sh SYNOPSIS .bd spell [ .bd \*-v ] [ .bd \-1 ] file \f3...\fP .sh DESCRIPTION .it Spell collects words from the named files, and looks them up in a spelling list. Words that neither occur among nor are derivable (by applying certain inflections, prefixes, or suffixes) from words in the spelling list are printed on the standard output. If no files are named, words are collected from the standard input. .s3 .it Spell omits .it nroff\^\c (I), .it troff\^\c (I), .it neqn\^\c (I), and .it eqn\^\c (I) constructions from the input. .s3 The process may take several minutes. .s3 Under the .bd \*-v flag, all words not literally in the spelling list are printed, and plausible derivations from spelling list words are indicated. .s3 The .bd \*-1 option causes the final output to appear in a single column instead of three columns. The normal header and pagination is also suppressed. .s3 The spelling list is based primarily on Kucera and Francis, .it "Computational Analysis of Present-Day English" and the Merriam Webster .it "New International Dictionary, 2nd edition." Other sources include lists of chemical elements, states, countries, provinces, capital cities, major cities; given names from Kucera and Francis; the most common surnames from a large telephone book; common names from the index of .it "Fieldbook of Natural History" by E. L. Palmer and H. S. Fowler; selected names from .it "Bulfinch's Mythology;" Bell System Practices; Bell Laboratories technical papers and manuals; the .it "Federalist" papers; random literary fragments; etc. .s3 If the file ``/usr/dict/spellhist'' is writable, .it spell accumulates copies of its output there. .sh FILES /bin/deroff, /usr/lib/spell[0123]: programs .br /usr/lib/w2006: list of common words for primary filtering .br /usr/dict/spellinglist .br /usr/dict/stoplist: likely misspellings (e.g. thier=thy\*-y+ier) that would otherwise pass .br /usr/dict/spellhist .sh "SEE ALSO" typo(I) .sh BUGS The coverage of the spelling list is uneven; new installations will probably wish to monitor the output for a few months to gather local additions.