find spelling errors
spell collects words from the named
documents 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
If no files are named, words are collected from the standard
spell ignores most troff, tbl, eqn, and pic
constructions. Copies of all output may be accumulated in the history file,
if one is specified.
follows chains of included files (“.so” and
The default spelling list is based on Webster's Second
International dictionary and should be fairly complete. Words that appear in
the “stop list” are immediately flagged as misspellings,
regardless of whether or not they exist in one of the word lists. This helps
filter out misspellings (e.g. thier=thy-y+ier) that would otherwise pass.
Additionally, the british file is also used as a
stop list unless the
-b option is specified.
Site administrators may add words to the local word list, /usr/local/share/dict/words or the local stop list, /usr/local/share/dict/stop.
All word (and stop) lists must be sorted in lexicographical order
with case folded. The simplest way to achieve this is to use “sort
-df”. If the word files are incorrectly sorted,
spell will not be able to operate correctly.
The options are as follows:
- Check British spelling. Besides preferring
etc., this option insists upon
-isein words like standardise, Fowler and the OED to the contrary notwithstanding. In this mode, American variants of words are added to the stop list.
- Use the specified word list instead of the default system word list. The word list must be sorted as specified above.
- Store misspelled words in the specified history file. The output of
who -mis appended to the history file after the list of misspelled words.
- Instruct deroff(1) to ignore “.so” and “.nx” commands.
- Use delatex instead of deroff(1) if it is present on the system.
- Enable support for common troff macro packages; this option is passed
verbatim to deroff(1). Refer to the
-mdescription in deroff(1) for details.
- Use the specified stop list instead of the default system stop list. The stop list must be sorted as specified above.
- Use detex instead of deroff(1) if it is present on the system.
- Print all words not literally in the spelling list in addition to plausible derivations from spelling list words.
- Print every plausible stem, prefixed with ‘=’.
- Use extra_list in addition to the default word list. The extra word list must be sorted as specified above.
- Default spelling list.
- American spelling of certain words.
- British spelling of certain words.
- Default stop list.
- Local spelling list (optional).
- Local stop list (optional).
- Binary executed by the shell script /usr/bin/spell.
deroff(1), look(1), sed(1), sort(1), tee(1)
spell command first appeared in
Version 5 AT&T UNIX. It did not appear in
freely redistributable BSD releases for licensing
reasons. After Caldera relicensed early UNIX releases the
4.4BSD version was added to OpenBSD
Unlike historic versions, the OpenBSD
spell command does not use hashed word files.
Instead, it uses lexicographically sorted files and the same technique as
The spelling list lacks many technical terms; new installations will probably wish to monitor the output for several months to gather local additions.
British spelling was done by an American.
-x mode it would be nicer if the stems
were grouped with the appropriate word.