2250 words

a predicted fact is a novel fact for a theory if it was not used to construct that theory  — where a fact is used to construct a theory if it figures in the premises from which that theory was deduced. (Musgrave, 1988; cf Mayo, 1991: 524)

Introduction

Previously I demonstrated that the HBD movement is a racist movement. I showed this by arguing that it perfectly tracks with John Lovchik’s definition of racism, which is where “racism is a system of ranking human beings for the purpose of gaining and justifying an unequal distribution of political and economic power.” There is, however, a different issue—an issue that comes from the philosophy of science. So a theory is scientific if and only if it is based on empirical evidence, subject to falsifiability and testability, open to modification or rejection based on further experimentation or observation and—perhaps most importantly—is capable of generating novel predictions, where a novel prediction goes beyond existing knowledge and expectation and can be verified through empirical testing.

Here I will show that HBD doesn’t make any novel predictions, and I will also discuss one old attempt at showing that it does and that it is an example of a degenerative research programme. Effectively, I will argue that contrary to what is claimed, HBD is a degenerating research programme.

On so-called novel predictions

HBD and evolutionary psychology falls prey to the same issues that invalidate both of them. They both rely on ad hoc and post hoc storytelling. In a previous article on novel predictions, I stated:

A risky, novel prediction refers to a prediction made by a scientific theory or hypothesis that goes beyond what is expected or already known within an existing framework (novelness). It involves making a specific claim about a future observation or empirical result that, if confirmed, would provide considerable evidence in support of the scientific theory or hypothesis.

So EP and HBD are cut from the same cloth. John Beerbower (2016) puts the issue succinctly:

At this point, it seems appropriate to address explicitly one debate in the philosophy of science—that is, whether science can, or should try to, do more than predict consequences. One view that held considerable influence during the first half of the twentieth century is called the predictivist thesis: that the purpose of science is to enable accurate predictions and that, in fact, science cannot actually achieve more than that. The test of an explanatory theory, therefore, is its success at prediction, at forecasting. This view need not be limited to actual predictions of future, yet to happen events; it can accommodate theories that are able to generate results that have already been observed or, if not observed, have already occurred. Of course, in such cases, care must be taken that the theory has not simply been retrofitted to the observations that have already been made—it must have some reach beyond the data used to construct the theory.

HBDers promote the tenets that intelligence (IQ), along with behavior and socioeconomic outcomes are strongly associated with genetic differences among individuals and groups. They also use the cold winter theory (CWT) to try to intersect these tenets and show how they evolved over time. According to the CWT, the challenges of surviving in colder climates such as the need to hunt, plan ahead, and cooperate exerted selective pressures which favored genes which fostered higher intelligence in populations that inhabited these regions. I have previously shown years back that the CWT lacks novel predictive power, and that there are devastating response to the CWT which show the invalidity of the theory. Rushton used it in his long-refuted r/K selection theory for human races. Further, for example Jablonski and Chaplin (2000) successfully predicted that “multiple convergences of light skin evolved in different modern human populations and separately in Neanderthals” (Chaplin and Jablonski, 2009: 457). This was a successfully predicted novel fact, something that HBD doesn’t do.

Urbach (1974) (see Deakin, 1976 for response) in criticizing “environmentalism” and contrasting it with “hereditarianism”, claimed that hereditarianism made novel predictions. He also claimed that the “hard core” of the hereditarian research programme was that (1) cognitive ability of all people is due to general intelligence and individual and (2) group differences are due to heredity. We know that (1) is false, since general intelligence is a myth and we know that (2) is false since group differences are due to environmental factors since Jensen’s default hypothesis is false (along with the fact that Asians are a selected population). Further Urbach (1974: 134-135) writes that 4 novel facts of hereditarianism are “(i) of the degree of family resemblances in IQ, (ii) of IQ-related social mobility, (iii) of the distribution of IQ’s, and (iv) of the differences in sibling regression for American Negroes and whites.”

But the above aren’t novel predictions.

(i) Hereditarianism predicts that intelligence has a significant hereditary component, leading to similarities in IQ scores among family members. (Nevermind the fact that environments are inherited by these family members as well.) The prediction appears specific, but it’s not novel in the framework of hereditarianism. The idea that IQ is heritable and that family members share similarities in IQ has been a main tenet of hereditarianism for decades, even in 1974 at the time of publication of Urbach’s paper,rather than offering a new or unexpected insight.

(ii) Hereditarianism also suggests that differences in IQ also have implications for social mobility, with people with higher IQs having a greater change for more upward social mobility. This, too, isn’t novel within the hereditarian framework since even in 1974 and the decades before then this was known.

(iii) Hereditarianism also predicts that IQ scores follow a normal distribution, with a majority of people clustering around the middle. This, too, isn’t a novel prediction, since even Binet unconsciously built his test to have a normal distribution (Nash, 1987: 71). (Also note that Binet knew that his scales weren’t measures but thought that for practical measures they were; Michell, 2012.) Terman constructed his test to also have it. Urbach (1974: 131) states that “even if researchers had set out to obtain a particular distribution of IQ’s, there was no divine guarantee that their efforts would have been successful.” But we know that the process of building a normal distribution is done by choosing only items that conform to the normal distribution are selected, since items most are likely to get right are kept while on both ends items are also kept. In their psychometrics textbook, Rusk and Golombok (2009: 85) state that “it is common practice to carry out item analysis in such a way that only items that contribute to normality are selected.” Jensen (1980: 71) even stated “It is claimed that the psychometrist can make up a test that will yield any kind of score distribution he pleases. This is roughly true, but some types of distributions are much easier to obtain than others.”

(iv) Lastly, hereditarianism predicts that differences in sibling regression or the extent to which sibling IQ scores deviate from the population mean could vary between racial and ethnic groups. The prediction seems specific, but it reflects assumptions of genetic influences on psychological trait—which already were assumptions of hereditarian thought at that time and even today. Thus, it’s not a new or unexpected insight.

Therefore, the so-called novel predictions referenced by Urbach are anything but and reflect existing assumptions and concepts in the field at the time of publication, or he’s outright wrong (as is the case with the normal distribution).

Modern day hereditarians may claim that the correlation between genetics and IQ/educational attainment validates their theories and therefore counts as novel. However, the claim that genes would correlate with IQ has been a central tenet in this field for literally 100 years. Thus, a prediction that there would be a relationship between genes and IQ isn’t new. Nevermind the fact that correlations are spurious and meaningless (Richardson, 2017; Richardson and Jones, 2019) along with the missing heritability problem. Also note that as sample size increase, so to does the chance for spurious correlations, (Calude and Longo, 2016). The hereditarian may also claim that predicting group differences in IQ based on genetic and environmental factors is a novel prediction. Yet again, the idea that these contribute to IQ has been known for decades. The general prediction isn’t novel at all.

So quite obviously, using the above definition of “novel fact” from Musgrave, HBD doesn’t make any novel predictions of previously unknown facts not used in the construction of the theory. The same, then, would hold true for an HBDer who may say something along the lines of “I predict that a West African descendant will win the 100m dash at the next Olympics.” This doesn’t qualify as a novel prediction of a novel fact, either. This is because it relies on existing knowledge related to athletics and racial/ethnic demographics. It’s based in historical data and trends of West African descendants having been successful at previous 100m dash events at the Olympics. Therefore, since it’s not a novel insight that goes beyond the bounds of the theory, it doesn’t qualify as “novel” for the theory.

Why novel predictions matter

Science thrives on progress, so without theories/hypotheses that make novel predictions, a scientific program would stagnate. The inability of hereditarianism to generate risky, novel predictions severely limits it’s ability in explaining human behavior. Novel predictions also provide opportunities for empirical testing, so without novel predictions, hereditarianism lacks the opportunity for rigorous empirical testing. But a proponent could say that whether or not the predictions are novel, there are still predictions that come to pass based on hereditarian ideas.

Without novel prediction, hereditarianism is confined to testing hypotheses that are well-known or widely accepted in the framework or the field itself. This then results in a narrow focus, where researchers merely confirm their pre-existing beliefs instead of challenging them. Further, constantly testing beliefs that aren’t novel leads to confirmation bias where researchers selectively seek out what agrees with them while ignoring what doesn’t (Rushton was guilty of this with his r/K selection theory). Without the generation of novel predictions, hereditarianism lacks innovation. Lastly, the non-existence of novel predictions raises questions about the progressiveness of the framework. True scientific progress is predicated on the formulation of testing novel hypotheses which challenge existing paradigms. Merely claiming that a field generates testable and successful novel predictions and therefore that field is a progressive one is unfounded.

Thus, all hereditarianism does is accommodate, there is no true novel predictive power from it. So instead of generating risky, novel predictions that could potentially falsity the framework, hereditarians merely resort to post-hoc explanations, better known as just-so stories to fit their preconceived notions about human behavior and diversity. HBD claims are also vague and lack the detail needed for rigorous testing—the neck isn’t stuck out far enough for where if the prediction fails that the framework would be refuted. That’s because the predictions are based on assumptions they already know. Thus, HBD is merely narrative construction, and we can construct narratives about any kind of trait we observe today have the story conform with the fact that the trait still exists today. Therefore hereditarianism is in the same bad way as evolutionary psychology.

I have previously compared and contrasted hereditarian explanations of crime with the Unnever-Gabbidon theory of African American offending (TAAO) (Unnever and Gabbidon, 2011). I showed how hereditarian explanations of crime not only fail, but that hereditarian explanations lack novel predictive power. On the other hand, Unnever and Gabbidon explicitly state hypotheses and predictions which would follow from. The TAAO, and when they were tested they were found to hold validating the TAAO.

Conclusion

In this discussion I have tried to show that hereditarian/HBD theories make no novel predictions. They are merely narrative construction. The proposed evolutionary explanation for racial differences in IQ relying on the CWT is ad hoc, meaning it’s a just-so story. Lynn even had to add in something about population size and mutation rates since Arctic people, who have the biggest brain size, don’t have the highest IQ which is nothing more than special pleading.

Urbach’s (1974) four so-called novel predictions of hereditarianism are anything but, since they are based on assumptions already held by hereditarianism. They represent extensions or reformulation of existing assumptions, while also relying on retrospective storytelling.

I have provided a theory (the TAAO) which does make novel predictions. If the predictions wouldn’t have held, then the theory would have been falsified. However, tests of the theory found that they hold (Burt, Simons, and Gibbons, 2013; Unnever, 2014; Unnever, Cullen, and Barnes, 2016; Herda, 2016, 2018; Burt, Lei, and Simons, 2017; Gaston and Doherty, 2018; Scott and Seal, 2019). The hereditarian dream of having the predictive and explanatory power that the TAAO does quite obviously fails.

Therefore, the failure of hereditarianism to produce successful, risky novel predictions should rightly raise concerns about its scientific validity and the scientific credibility of the program. So the only rational view is to reject hereditarianism as a scientific enterprise, since it doesn’t make novel predictions and it’s merely, quite obviously, a way to make prejudices scientific. Clearly, based on what a novel prediction of a novel fact entails, HBD/hereditarian theory doesn’t make any such predictions of novel facts.