Overcoming "mutating table" error in a trigger

I have a column in a table where I want to place a UNIQUE "constraint". However, I want to allow "privileged" users to override this "constraint". That is, users will not be able insert a row if there is already a row in the table with a duplicate value for that column (unless that user is privileged).

My first implementation of this feature was to have a before-row trigger to check the row being inserted to see if it the column value already exists in the table. However, this creates a "mutating table" error (see page 8-9 of Application Developer's Guide) because I'm SELECTing from a table I'm trying to INSERT into. The documentation suggests using a temporary table, a PL/SQL table, or a package variable (page 8-10). The temporary table might work but it seems like a lot of resources would be needed to copy the rows into the temp table. And I don't see how using a PL/SQL table or package variable would help in my case because I would need to "protect" the PL/SQL table or package variable from "unprivileged" users.

Am I missing something obvious? Are there other ways to do what I want? Any suggestions would be appreciated.

-- 
Cheong Yu             |                        |        (513) 865-7048
Mead Data Central     |                        |   Fax: (513) 865-1655
P.O. Box 933          |                        |      kcy_at_meaddata.com
Dayton, Ohio  45401   |                        |    uunet!meaddata!kcy