A pair of fair dice is rolled, and the sum of the two numbers falling uppermost is observed. The probability of obtaining a sum of 2 is the same as that of obtaining a 5 since there is only one way of getting a 2—namely, by each die showing a 1; and there is only one way of obtaining a 5—namely, by one die showing a 2 and the other die showing a 3. What is wrong with this argument?