There seems to be no end to illogical and even comical “findings” from MZ-DZ twin method comparisons, where the original twin researchers argue that the greater behavioral resemblance of reared-together MZ (monozygotic, identical) versus same-sex DZ (dizygotic, fraternal) twin pairs demonstrates the “heritability” of the behavioral characteristic in question. Among these we find a twin study whose authors concluded in favor of a genetic basis for being a “born again Christian” (65% heritability), another that found important genetic influences on tea and coffee drinking preferences, and still another that found that the heritability of “loneliness in adults” is 48%.
In a similar manner, in an April 13th, 2015 article published in The Independent, British twin researcher Tim Spector argued that his TwinsUK study findings show that “research with twins suggests picking who to vote for in an election might have more to do with your genes than the policies of the parties.”1 Twin research in political science goes back to 2005 and earlier,2 and has not gone unchallenged in the field.3 Spector and colleagues asked 612 reared-together pairs of twins born in the United Kingdom “whether they intended to vote, what their political party of choice was and…their personal rating of the main party leaders.” Predictably, like most studied behaviors and behavioral disorders, MZ pairs correlated higher than DZ pairs for several characteristics, a result that twin researchers attribute to genetics on the basis of their assumption that MZ and DZ pairs grow up experiencing similar environments. This is known as the MZ-DZ “equal environment assumption” (EEA). As I show in my recent book The Trouble with Twin Studies, and elsewhere, the evidence clearly shows that MZ twin pairs grow up experiencing much more similar environments, and experience psychological closeness and attachment to a far greater degree, than experienced by DZ twin pairs. Although most twin researchers now recognize that MZ childhood environments are more similar, they uphold the validity of the EEA on the basis of circular arguments and other types of illogical reasoning.4
In a statement that even most of his twin researcher colleagues would disagree with, Spector upheld the EEA by claiming, without qualification, that “both identical [MZ] and non-identical [DZ] twins normally share the same environment while growing up.” This false claim allowed him to conclude that the greater behavioral resemblance of MZ pairs is caused by genetic factors. According to Spector:
“We found that voting Conservative [Tory] (or not) is strongly influenced by genetics. When it came to voting Tory, we found that 57% of the variability (differences or similarity) between people’s voting preferences were due to genetic effects. This percentage is called heritability. That means the identical twins were more likely to vote the same way than the non-identical twins–suggesting [that] an underlying genetic influence was stronger than environmental or random factors. For UKIP [UK Independence Party] voting preferences, there was also a moderately strong heritability of 51%. This was closely followed by Labour and the Green Party both with 48%. The exception seemed to be voting for the Liberal Democrats, which was affected entirely by environment, with no genetic influence.”
Based on accepting both the validity of the EEA and the controversial “heritability” concept, Spector concluded that the heritability of voting Conservative (Tory), UKIP, Labour, or Green was “moderately strong,” whereas there was no genetic influence on voting for the Liberal Democrats. In addition, Spector concluded that “geography also played a possible role–as voting for the SNP [Scottish National Party] in Scotland was also completely environmental.” Apparently, there are genes predisposing people to vote for some (British) political parties, but not for others!
Suppose that Spector had found that MZ pairs correlate at 75% for supporting the Tories, and that same-sex DZ pairs correlate at 25%. Because simple heritability estimates are based on doubling the MZ-DZ correlation difference, in this case the heritability of voting Tory would be 100%, which would lead to the absurd conclusion that “the policies of the parties” had virtually no influence on whether or not someone voted Tory.
Turning to the candidates running for office, Spector wrote,
“The question of whether a leader would make a good prime minister produced mixed responses. David Cameron [Tory] had the stronger genetic influence on opinions, with 50% heritability, followed by Nick Clegg [Liberal Democrat] at 37%. Views on all the other party leaders were purely environmental.”
UK residents carry genes predisposing them to vote for Cameron or Clegg, according to the logic of Spector and the twin method, but carry none for the other candidates.
In addition to defending the twin method’s untenable equal environment assumption, twin researchers of political behavior, and of behavior in general, focus narrowly on the twins they study, and usually overlook obvious real-world natural experiments contradicting their claims.5 For example, the people living in “socialist” North Korea and “capitalist” South Korea are members of the same ethnic population, but are divided for political and military reasons. Koreans on either side of the demilitarized zone that separates the two countries hold differing political views and support different political leaders, for reasons having nothing to do with genetics. For the same reason, we would expect attitudes toward “government provided health care” to show marked differences between people of British ancestry living on either side of the U.S.-Canada border. Genetic predispositions also do not explain why Germany was heavily anti-Nazi in the late 1920s, and heavily pro-Nazi just a few years later. The list of examples is endless.
Spector concluded that “the findings of this study suggest that our choices at the polling booth may not be as free or rational as we would like to believe.” Very true, but the reduction of free will and rationality is not “in the genes,” but is instead a product of manipulation and propaganda by powerful political and economic (corporate) forces through education, the media, the Internet, religion, speeches, and so on. Voting patterns are also influenced by the common practice of parties and politicians campaigning on the basis of one program, and then carrying out a different program once in office (usually referred to as “broken campaign promises”). The Independent is not a satirical publication, and Spector’s article seems more appropriate for the science/technology section of The Onion.
Twin researchers in political science continue to uphold the validity of the EEA in twin studies of behavior,6 at times relying on convoluted and illogical arguments.7 But the fact remains that MZ pairs’ more similar political views, party affiliation, candidate choices, and voting patterns can be completely explained by the more similar treatment they receive growing up, in addition to their greater tendency to model their behavior on each other, their much greater levels of closeness, loyalty, and attachment, and their greater tendency to experience identity confusion than DZ pairs.8 This conclusion is supported by the ongoing failure to identify genes for political characteristics at the molecular genetic level.9
Twin researchers sometimes claim that the arguments of their critics should be questioned because these critics do not perform twin studies, and therefore are not experts on twins. This occurred in a recently reopened debate on twin research in American criminology, where criminologists Callie Burt’s and Ronald Simons’ conclusion that twin studies are methodologically and conceptually flawed was criticized by twin researchers in the field. These twin researchers attempted to counter Burt and Simons’ critique, in part, by claiming that they “have never worked with twin data, and they show no signs of being familiar with the large and substantial body of work that supports the use of these behavioral genetic methods.”10 Aside from the fact that Burt and Simons possess great knowledge of the criminology twin and adoption study literature,11 expert status is not always necessary to debunk a research technique or scientific theory, if the technique or theory is based on obviously false assumptions.
In fact, people outside of twin research and behavioral genetics are better equipped to see the glaring problems and false assumptions underlying studies of twins reared together and of twins reared apart—problems and false assumptions that twin researchers do not allow themselves to see, or dismiss with illogical arguments. As the pioneering twin research critic Leon J. Kamin once wrote,
“The [twin] investigators, after all, tend to analyze their data in the same ways, reflecting the same theoretical preconceptions. The problem is nothing so simple as the suppression of embarrassing data. Theoretical commitment makes it unlikely that embarrassing patterns within the data will even be noticed.”12
Indeed, theoretical commitment to the twin method and its equal environment assumption often compels twin researchers to overlook obvious embarrassing patterns both within their data and in society, and at times leads them to arrive at absurd conclusions such as the claim that preferences for some political parties and candidates have a “strong” or “moderately strong” genetic basis.13
* * * * *
1. Spector, T., (2015, April 13th), General Election 2015: Do Your Genes Determine How You Vote? The Independent.
2. Alford et al., 2005, Are Political Orientations Genetically Transmitted?, American Political Science Review, 99, 153-167; Joseph, J., (2010), The Genetics of Political Attitudes and Behavior: Claims and Refutations, Ethical Human Psychology and Psychiatry, 12, 200-217.
3. Charney, E., (2008a), Genes and Ideologies, Perspectives on Politics, 6, 292-319; Charney, E., (2008b), Politics, Genetics, and “Greedy Reductionism,” Perspectives on Politics, 6, 337-343; Charney, E., (2012), Behavior Genetics and Postgenomics, Behavioral and Brain Sciences, 35, 331-358; Charney, E., (2013), Nature and Nurture, Perspectives on Politics, 11, 558-561; Charney, E., & English, W., (2012), Candidate Genes and Political Behavior, American Political Science Review, 106, 1-34; Shultziner, D., (2013), Genes and Politics: A New Explanation and Evaluation of Twin Study Results and Association Studies in Political Science, Political Analysis, 21, 350-367.
4. Joseph, J., (2004), The Gene Illusion: Genetic Research in Psychiatry and Psychology under the Microscope, New York: Algora; Joseph, J., (2015), The Trouble with Twin Studies: A Reassessment of Twin Research in the Social and Behavioral Sciences, New York: Routledge.
5. Like most twin researchers, Spector believes that “twins provide a unique natural experiment for research.”
6. Alford et al., 2005; Hatemi et al., (2014), Genetic Influences on Political Ideologies: Twin Analyses of 19 Measures of Political Ideologies from Five Democracies and Genome-wide Findings from Three Populations, Behavior Genetics, 44, 282-294.
7. For example, in 2012 Kevin Smith and a group of leading political science twin researchers concluded that even if the critics are “wholly correct” that the causes of MZ-DZ differences are “exclusively environmental,” this finding would “provide reasons for political science to pay more rather than less attention to the biological basis of attitudes and behaviors.” It is illogical, however, to state that political scientists should “pay more attention” to biological influences on political attitudes and behaviors if explanations for MZ-DZ differences are “exclusively environmental.” See Smith et al., (2012), Biology, Ideology, and Epistemology: How do We Know Political Attitudes are Inherited and Why Should We Care?, American Journal of Political Science, 56, 17-33, p. 17. For a critical analysis of this study, see Joseph, 2015; Joseph, J., (2013), The Use of the Classical Twin Method in the Social and Behavioral Sciences: The Fallacy Continues, Journal of Mind and Behavior, 34, 1-39.
8. See Joseph, 2015, Chapters 7 and 8.
9. Hatemi et al., 2014.
10. Wright et al., (2015), Mathematical Proof Is not Minutiae and Irreducible Complexity Is Not a Theory: A Final Response to Burt and Simons and a Call to Criminologists, Criminology, 53, 113-120, p. 117.
11. Burt, C. H., & Simons, R. L., (2014), Pulling Back the Curtain on Heritability Studies: Biosocial Criminology in the Postgenomic Era, Criminology, 52, 223-262. See also Burt, C. H., & Simons, R. L., (2015), Heritability Studies in the Postgenomic Era: The Fatal Flaw is Conceptual, Criminology, 53, 103-112.
12. Kamin, L. J., (1981), Commentary, in S. Scarr (Ed.), Race, Social Class, and Individual Differences in I. Q., (pp. 467-482), Hillsdale, NJ: Lawrence Erlbaum Associates, p. 480.
13. As the author of an eight-word response to the same Spector article reprinted in the April 15th, 2015 edition of The Guardian put it, “Science has gone flipping mad these days, unbelievable.” See the comment by “Thompson1001.”
I remember watching some tv program, I think this is it …
Even if this is true I find it hard to believe why people are trying to figure this out, feels like it kind of makes a mockery of democracy.
Thanks for this excellent article Jay. We should take such careful, scholarly research very seriously.
Have you heard of the following studies? It has been determined that:
– Children’s preference for Mars bars (as opposed to Snickers) has been determined to be 74.82% heritable. Snickers executives are upset about this finding.
– Genes contribute to 54.53% of a person’s chance of getting pulled over for speeding. This is what you can tell the policeman the next time you get pulled over – “My genes made me do it.”
– Whether young people identify more with Thor or Loki (from the Avengers movies) is 67.04% attributable to heredity, and over 90% heredtiary among those people of Nordic descent, whose genes presumably would have been more heavily influenced by exposure to the Norse gods.
– Genes contribute to 84.31% of the variance in whether young boys chooses football, basketball, or baseball for their first team sport. European researchers are complaining that these limited choices demonstrate an overfocus on American sports, and that the genetic influence becomes less when soccer is included.
– Genes are responsible for 28.21% of the variance in what people’s favorite colors are. This would explain why people’s favorite colors often change as they grow up, because it is one of the few things in life not primarily determined before people are even born.
I’m sorry, but I’m currently unable to locate the links to these studies. But you will surely be hearing about them soon.
We should give thanks that we live in an enlightened age, one where such wisdom of geneticists can be appreciated and shared, and where taxpayer money is devoted to the study of this groundbreaking, empowering subject.
Behavioral genetics is at the foundation of the pseudoscientific support for the medical model; your debunking of this pseudoscience is historically important.
Thank you for your community service, Steve
Jay – Thanks for letting us in on the ground floor of the debate internal to the field–I especially appreciate the clarity in that sense.
Jay, you fail to appreciate the irony that your MIA blog critiquing heritability estimates is, itself, 68% heritable.
Just kidding, thanks for another informative post and keep up the great work!
“For UKIP [UK Independence Party] voting preferences, there was also a moderately strong heritability of 51%. This was closely followed by Labour and the Green Party both with 48%. The exception seemed to be voting for the Liberal Democrats, which was affected entirely by environment, with no genetic influence.””
Oh that one is great… Makes perfect sense. I suggest a follow up study – checking how the political preferences of people changed after disaster in Fukushima – I’m sure the rise in opposition to nuclear power can be best explain by radiation-induced genetic mutations, right? A clear candidate for IgNobel prize.