将order by子句置于子查询之外-

SELECT *
FROM ( SELECT DISTINCT tb.primarytitle AS Title, tb.genres AS Genres, tr.averagerating AS Rating, tb.endyear AS Year
                FROM title_basics tb, title_ratings tr
                WHERE tb.genres LIKE '%Comedy%' AND (tb.endyear = '2001' OR tb.endyear = '2002' OR tb.endyear = '2003' OR tb.endyear = '2004' OR tb.endyear = '2005')

)
WHERE ROWNUM <= 5
ORDER BY Year DESC

你可以用 FETCH 子句指定要返回的行数。它比rownum（docs）灵活得多

SELECT DISTINCT tb.primarytitle     AS title,
                tb.genres           AS genres,
                tr.averagerating    AS rating,
                tb.endyear          AS year
  FROM title_basics   tb,
       title_ratings  tr
 WHERE tb.genres LIKE '%Comedy%' AND
       ( tb.endyear = '2001' OR
         tb.endyear = '2002' OR
         tb.endyear = '2003' OR
         tb.endyear = '2004' OR
         tb.endyear = '2005' )
 ORDER BY tb.averagerating DESC
FETCH FIRST 5 ROWS ONLY;

你的问题真没道理。平均分级表中没有联接列。可能有一列将其与另一个表联系起来：

SELECT *
FROM (SELECT tb.primarytitle AS Title, tb.genres AS Genres, tr.averagerating AS Rating, tb.endyear AS Year
      FROM title_basics tb JOIN
           title_ratings tr
           ON tb.? = tr.?   -- presumably some JOIN condition here
      WHERE tb.genres LIKE '%Comedy%' AND 
            tb.endyear IN ('2001', '2002', '2003', '2004', '2005')
      ORDER BY tr.averagerating DESC
     ) br
WHERE ROWNUM <= 5;

这个 ? 用于指定 JOIN 条件。
如果正确地联接表，则可能不需要 SELECT DISTINCT .