spell

spell [-b] [-v] [-x] [-d hlist] [-s hstop] [-h hsuf] [-i] [-l] [+additions] [file ...]

-b
    Checks for British spelling variations instead of American English.

-v
    Prints all words not found in the dictionary, and also words that are 'close' to a dictionary word (e.g., derived forms).

-x
    Prints every plausible stem for each word, showing how spell's algorithms might interpret word structures.

-d hlist
    Specifies an alternative dictionary (history list) file to use for spell checking.

-s hstop
    Specifies an alternative stop list (words to ignore, even if they're not in the main dictionary).

-h hsuf
    Specifies an alternative suffix list, used for recognizing word variations.

-i
    Ignores troff/nroff constructs and directives within the input text, treating them as formatting and not part of the words to be checked.

-l
    Follows symbolic links in the input files (behavior may vary across implementations).

+additions
    Specifies a file containing a list of words, one per line, to be temporarily added to the dictionary for the current run. These words will not be flagged as misspelled.

The spell command is a traditional Unix utility designed to identify potential spelling mistakes in text. It reads input from standard input or specified files, processes each word, and outputs a list of words not found in its internal dictionary to standard output. Unlike more modern spell checkers, spell does not offer suggestions for corrections, nor does it provide interactive features. Its primary function is a simple dictionary lookup. It's often utilized in shell scripts or command pipelines due to its non-interactive nature and ability to filter out correctly spelled words.

Historically, spell implementations have often leveraged other standard Unix tools like sort and comm to achieve its functionality, demonstrating the powerful Unix philosophy of combining small, specialized programs. While largely superseded by tools like aspell or ispell for interactive or more sophisticated tasks, spell remains a foundational tool for quick, automated dictionary checks.

spell is a basic spell checker. It does not offer interactive correction, context-aware checking, or grammatical analysis. Its dictionary might be outdated or limited depending on the system configuration. Performance can be slower for very large inputs compared to modern, optimized tools, as it often relies on external commands like sort and comm.

spell is frequently cited as a prime example of the Unix philosophy: 'Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface.' Its common implementation as a shell script orchestrating tools like sort and comm perfectly illustrates this principle.

By default, spell typically defines a 'word' as a sequence of alphabetic characters (and sometimes apostrophes or hyphens) separated by non-alphabetic characters. It generally ignores numbers, punctuation, and whitespace unless specific options or dictionary rules modify this behavior.

The spell command is one of the oldest Unix utilities, originating from Bell Labs in the early 1970s. It was initially developed by Stephen C. Johnson and was a part of Version 6 Unix. Its design was innovative for its time, employing hashing techniques for efficient dictionary lookups. Later Unix and Linux implementations often re-engineered spell as a shell script that pipelines several standard commands (like tr, sort, uniq, and comm) to compare input words against a master dictionary. This modular approach exemplifies the Unix philosophy of building powerful tools by combining simpler, specialized ones. While its core function remains consistent, its internal implementation has varied significantly across different operating systems over the decades, leading to some differences in options and behavior.

aspell(1), ispell(1), comm(1), sort(1), look(1)