Cuckoldry rates in Germany are ~1 percent

By Razib Khan | February 4, 2013 1:50 am

One of the quasi-facts which I often stumble upon is the idea that in 10 percent of cases paternity is misattributed. That is, the presumed father is cuckolded. I often encounter this “fact” in a biological context, where someone with an advanced degree in biology will relate how it turns out that there is a great deal of delicacy in situations of transplant matching because of this fact. When pressed on the provenance of this fact most demur. The reason people demur is that the factual basis of this assertion is very thin. In particular, very high estimates of cuckoldry come from databases of disputed paternity, which are obviously going to be a biased sample. A more thorough survey suggests that there is a wide variation in misattributed paternity across populations.

In the interests of disabusing the public of this myth, I point to a paper from Germany, Estimating the Prevalence of Nonpaternity in Germany. The sample consists of the families of children who require bone marrow transplants. The authors note two important conditions: 1) the details of the results as they might relate to paternity are not divulged, 2) none of the parents refused to be typed. Since susceptibility to childhood cancers are evenly distributed across the population the biases introduced in other surveys presumably do not apply to this situation.

Why does any of this matter? Because models of paternity uncertainty are important priors in shaping our view of the course of human evolutionary history. Sexual jealousy and mate guarding loom large in evolutionary psychology. I don’t particularly know how high paternity certainty impacts these arguments, but it needs to be brought to the fore, rather than relying on an old chestnut of wisdom based on nothing.

And yet as suggested by the title, I grant that this result may not be generalized. But it’s a place to start. Perhaps in some societies paternity certainty really is low. That’s something to consider. Though at least in the developed West that does not seem to be the case (OK, at least in Germany!).

MORE ABOUT: paternity

Comments (16)

  1. TheBrett

    Even 0.94% seems quite high in absolute numbers, although it implies a much lower adultery rate than the one you see in popular writings (unless adulterous people are better about using birth control).

    • razibkhan

      varies across SES. 1% seems reasonable given the difference in fam structures up and down the SES ladder

    • ohwilleke

      The adultery rate is almost surely much higher. The average woman in the developed world has 2 children or less in a lifetime, out of about twenty-five years of fertility. Everybody in the modern world is good about controlling their fertility, and the cuckold figure excludes men having affairs with unmarried women and almost all of the people whose affairs led to divorces.

  2. RogerB

    There’s a series of an English surname found in a small area that shows an
    average nonpaternity rate of 1.3% over the period since the adoption of the name, 700 years.

    I’ve heard an anecdotal story on the early studies of the human blood groups (red cell antigens). These were mostly carried out by British blood bankers in the 1930-60 period. To show mendelian inheritance they assembled pedigrees of multiple families. To check paternity they also typed the individuals using all the known blood groups to that date. When discrepancies showed up they discarded all or part of the affected pedigree.

    Genealogists use the term “nonpaternity event”, since nonconformity of paternal name and DNA can reflect adoption as well as marital infidelity. Since recording of adoptions wasn’t routine before the 19th Cent, at least in the US, it’s often impossible to distinguish the causes of discrepancies before that time.

    • stevesailer

      Something we can’t know is the fidelity rate of surnames that have gone extinct. (By one estimate, 75% of English surnames have gone extinct.) Based on Gregory Clark’s Farewell to Alms research on wills in 1200-1800, I suspect that names that prospered, such as Sykes, lived more orderly lives than those that died out.

  3. RogerB

    I should have added that the British blood group data suggested a nonpaternity rate of 1-3%, IIRC. There was no study designed to come up with a number, but the number of cases was huge.

  4. Douglas Galbi

    Part of the problem is that different statistics get confused. The share of men who wrongly believe a child is their biological child (among men with children) is different from the share of children (within their birth cohort) who wrongly believe a man to be their biological father.

    This study uses a child-based measure across age cohorts from a bone marrow transplant database at a German university hospital from 1993 to 2008, where each child has a mother and presumed father. That selects for families with a mother and a presumed father present. A man who is not biological the father plausibly has weaker relation to the child’s mother is thus is plausibly less likely to be around as the child ages. The paper doesn’t report the average age of the children. It’s not even clear whether “children” include an age specification or just a relational specification, e.g. mom and dad and 25-yo daughter/child. That seems to me a serious weakness of an apparently technically sophisticated study.

    Reviewing a lot of data, I estimated roughly 5% of children having false paternity beliefs in high-income countries. Details here:

    I would reduce that estimate by half if the average age of children in this study were 2 years. I wouldn’t reduce it all if the average age of the children were 20 years.

    • razibkhan

      with widespread genotyping we’ll know. how much money would you bet on your estimate? (given some reasonable intervals)

      • Gregory Cochran

        You are such a plunger, Razib.

      • Douglas Galbi

        Is knowing the average/median age of the children important for evaluating sample selection in this study? Do you think you can get those figures from the authors? I emailed the lead author last night asking for the figures. Haven’t heard anything back.

        While I’m aware that prediction markets are a thriving field, I don’t bet on scholarly questions. Since I haven’t looked at the data in more than three years, I can’t truly give you more data on higher-order moments of what my belief distribution was after closely looking at the relevant studies.

        You say “with widespread genotyping we’ll know.” I disagree. I believe that in a decade, despite the importance of genetic data for medical treatment and cheap genotyping, less than 10% of 21 year-olds will know with genetic testing who their biological fathers are. Here’s some justification for that belief:

        • Douglas Galbi

          In the U.S. in 2005, the median age of a bone marrow donor recipient
          was about 40 years of age. See

          The lead author of the German study responded to my email. He said that they have no information on the ages of the focal persons in their study. Hence the study uses the term “children” in only a relational sense. A median age of 40 is probably a good rough estimate of the median age of the focal persons in the German study. For each of these persons, the hospital obtained a genetic sample from the mother and the putative biological father. Being able to do that for patients of median age roughly 40 seems to me to be strong selection on family structure. A man not biologically related to a woman’s child is likely to have a lower probably of remaining accessible to the woman over decades while remaining falsely identified as the biological father. This downward selection bias is probably much greater than the estimated sampling error bounds.

    • ohwilleke

      The observation that the percentages of parents who are deceived and children who are deceived is not symmetrical is a worthwhile point.

      While some fathers are cuckolded without their knowledge or suspicion, there are plenty of Josephs out there who consciously know that a child for whom they choose to assume responsiblity is not their genetic child, or who deliberately choose not to investigate the possibility that a spouse’s child is their genetic child despite knowing that it is likely. In those cases, the line between true cuckoldry, and mere informal adoption or tacitly accepted de facto sperm donor arrangements, can be pretty thin.

      Presumably, the percentage of mothers who are unaware that the purported fathers of their children are not the genetic fathers of their children when this is the case is even lower, although probably not zero. Mothers generally know all of the potential fathers and are generally more accurately aware of when they are potentially fertile than their partners.

      Children are far less equipped to have accurate suspicions in the absence of an obvious difference in physical appearance at odds with that of a purported father (a situation in which the purported father almost surely also knows). So the percentage of children who are deceived about the identity of their genetic father is almost surely materially higher than the percentage of purported fathers who are deceived about whether their children are genetically related to them.

      On another point, given the very strong empirical link between suspected female infidelity and severe domestic violence in a recent study, I suspect that severe domestic violence is probably a particular strong predictor of non-paternity.

      • stevesailer

        Excellent points. That framework really makes things clearer.

        There are also outsiders beyond the father-mother-child trio, who can misattribute paternity as well. I was surprised to learn from David Maraniss’s biography “Barack Obama: The Story” that many teachers and children at Barry Soetoro’s school in Indonesia initially assumed that, while his mother was white, his father must be Indonesian. They assumed his father must have Papuan ancestry from the Indonesian island of Ambon.

    • stevesailer

      Good point. “Misattributed by whom?” is a question that should be asked a lot when thinking about these issues.

  5. stevesailer

    I can only read through the first two pages of the article, but let’s try to think through a scenario:

    A couple bring a sick child to the medical center. A nurse asks, “Are you the parents?”

    The man says, “Yes.”

    The nurse says, “Here, each of you fill out a family medical history form so we can see if any inherited conditions might be affecting your child.”

    Is the man the genetic father 99% of the time?

    I doubt it. I suspect that 5% or 10% or whatever of the time the man then says, “Oh, actually, I’m the stepfather, not the biological father.”

    The nurse says, “Oh, never mind.” Then she turns to the mother and asks her to fill out forms for herself and the biological father if she can remember.

    Later, the doctor tells the mother to contact the biological father to give a blood sample. She calls the man she assumes is the biological father. And, 99% of the time the mother is right!


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar