Blood thicker than water?
From ancient times, men have been bedeviled by doubts about whether they are truly the biological father of their own children. People who look very different from their siblings may develop their own doubts after constantly hearing friends and family remark about it. In fact, even a woman, if she has cheated on her spouse, may not be sure who the true father of her child is. Even after the mother has passed nine months of pregnancy and the agony of giving birth, there is no absolute assurance that after the infant was born it was not switched for another. Rumors that some prince was actually a peasant's child, or that Emperor Qianlong was of Han rather than Manchu descent, are grist to the mill of authors engrossed by royal intrigue.
The Song-dynasty forensics classic Witness to a Prosecution recorded a method for unraveling such mysteries: dripping the blood of children onto the skeletal remains for their parents, and using the ability of the blood to penetrate the bones as proof of a filial relationship.
Martial arts novels allude to another method for determining parentage--dripping the blood of two people into a bowl of water. If the blood blended together, this was taken to mean that the two had a familial relationship.
These tests using blood appear now to lack a scientific basis.
Consider the method of dripping blood into water. Taipei Veterans General Hospital (VGH) blood bank medical test technologist Liu Hsueh-mei points out that O-type blood contains both A and B type antigens, while B type blood contains both O and A type antigens. Dripping different types of blood into water will cause them to coalesce accordingly. But other than this, it is difficult to glean any further conclusions about the two people.
In fact, as early as 1935, doctors had begun using blood type to determine familial relationships. However, too many people have the same rheusus blood group for this method to be effective. For example, in Taiwan the most common blood type is O, possessed by 42% of the population--meaning more than one out of every three people--making it a poor tool for making the necessary distinctions. In addition, the discovery of the para-Bombay phenotype and other special blood types has created another blind spot for this method. In sum, using rhesus blood group to determine familial relationships can give erroneous results, and people have even been wrongly condemned on this account.
Tzeng Cheng-hwai, chief of the Section of Transfusion Medicine at Taipei VGH, points out that using blood type to determine parent/child relationships can only eliminate 60% of false claims (meaning that of every 100 "false" fathers, only 60 can be excluded).
In 1976, the human leukocyte antigens (HLA) test was introduced. Mutation rates in this polymorphous protein are quite high, allowing its accuracy to reach roughly 90%. However, for determining paternity, which demands certainty, HLA tests still leave something to be desired.
In the 1990s, DNA fingerprinting was adopted for use in paternity testing, resulting in the ability to correctly identify more than 99.99% of "false" fathers, making it virtually foolproof. Tzeng illustrates its accuracy by pointing out that if the fathers of everyone in Taiwan were not who they were supposed to be, the accuracy of DNA fingerprinting would mean only one person would not be ruled out.
Having endured nine months of pregnancy and the agony of giving birth, a mother has a greater sense of certainty that the infant is hers than does a father, who is largely an outsider for everything that happens between conception and birth.