Problem description

I try to join or select unique rows from the same table from different columns and rows. The unique ID exists in both Column cPlayerA and cPlayerB. The table can grow dynamically and I cannot hard code the SQL.

Table Structure and data

CREATE TABLE tJoinTwoColumn(
cId int(11) NOT NULL AUTO_INCREMENT,
cPlayerA int(10) DEFAULT NULL,
cPlayerB int(10) DEFAULT NULL,
PRIMARY KEY (cId)
) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8 COLLATE=utf8_bin;

#insert data
INSERT INTO tJoinTwoColumn (cId,cPlayerA,cPlayerB) VALUES (1,2001,1001);
INSERT INTO tJoinTwoColumn (cId,cPlayerA,cPlayerB) VALUES (2,1001,2001);
INSERT INTO tJoinTwoColumn (cId,cPlayerA,cPlayerB) VALUES (3,4001,3001);
INSERT INTO tJoinTwoColumn (cId,cPlayerA,cPlayerB) VALUES (4,3001,4001);

select * from tJoinTwoColumn;

Data returned

Shows what is selected and what I need to filter away

Code I tried

select distinct t1.cPlayerA , t1.cPlayerB  from tJoinTwoColumn t1
join  tJoinTwoColumn t2 on t1.cPlayerA = t2.cPlayerB
group by t1.cPlayerA , t1.cPlayerB;

But it returns 4 rows when I want to return only 2 rows in above example I would like row 2 and 4 to be filtered out in the select.

Is this possible?