Answer 1

This will need to be done in steps. First find the duplicate records in your table having a count > 1. Next insert those duplicates into a temp table. Next delete all of the duplicate records from your table, finally insert the rows from the temp table back into your table. Run the below to see this in action using temp tables (run in steps and select * from each table at any time).

CREATE TABLE #NAMES(LastName varchar(25), FirstName varchar(25), Record_Description varchar(100))
INSERT INTO #NAMES VALUES ('Smith', 'John', '5 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'John', '5 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'John', '5 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'John', '5 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'John', '5 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'Jane', '3 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'Jane', '3 Original Duplicates')
INSERT INTO #NAMES VALUES ('Smith', 'Jane', '3 Original Duplicates')
INSERT INTO #NAMES VALUES ('Doe', 'John', '2 Original Duplicates')
INSERT INTO #NAMES VALUES ('Doe', 'John', '2 Original Duplicates')
INSERT INTO #NAMES VALUES ('Jones', 'Bill', '1 Original No Duplicates')

--FIND RECORDS HAVING A COUNT > 1 AND INSERT INTO TEMP TABLE
SELECT LastName, FirstName, Record_Description, COUNT(*) AS [RECORD_COUNT]
INTO #TEMP_DUPLICATES
FROM #NAMES
GROUP BY LastName, FirstName, Record_Description
HAVING count(*) > 1

--DELETE RECORDS FROM SAMPLE TABLE THAT ARE IN TEMP TABLE
DELETE #NAMES
FROM #NAMES, #TEMP_DUPLICATES
WHERE #NAMES.FirstName = #TEMP_DUPLICATES.FirstName
AND #NAMES.LastName = #TEMP_DUPLICATES.LastName
AND #NAMES.Record_Description = #TEMP_DUPLICATES.Record_Description

--INSERT UNIQUE RECORDS FROM TEMP TABLE BACK INTO SAMPLE TABLE
INSERT INTO #NAMES
SELECT LastName, FirstName, Record_Description
FROM #TEMP_DUPLICATES
DROP TABLE #TEMP_DUPLICATES

--VIEW FINAL RESULTS
SELECT * FROM #NAMES
DROP TABLE #NAMES