As Martin Smith commented, "It depends." The factors that control the processing of Full Text indexes depends on the configuration you use in your CREATE FULLTEXT INDEX
command.
- Language - Which controls the parsing of words and phrases according to that language's rules.
- Stop Word List - Words to be left out of the search, though their position in the string is recognized.
- Accent Sensitivity - Determines whether accents are processed or ignored.
There is a tool to help you test your queries: https://msdn.microsoft.com/en-us/library/cc280463.aspx
You can use the sys.dm_fts_parser command
to test your full text results. For example, three similar searches using this function.
select * from sys.dm_fts_parser('"of" and "-" and "Department"',
1033, NULL, 0) --American English, No stop words, Accent Insensitive;
Returns 2 tokens: of, Department
select * from sys.dm_fts_parser('"of""-""Department"',
1033, NULL, 0) --American English, No stop words, Accent Insensitive;
Returns 2 tokens: of, Department
select * from sys.dm_fts_parser('"of-Department"',
1033, NULL, 0) --American English, No stop words, Accent Insensitive
Returns 3 tokens: of-Department, of, Department
In general a "-" is ignored when standing alone. However, depending on the Language choice in your Full Text Index, the "-" serves as a hyphen in text and may be reported as shown in the third example which includes "of-Department" as one of the tokens returned.