Gould on Morton, Redux: What can the debate reveal about the limits of data?

Stephen Jay Gould’s Analysis of the Army Beta Test in The Mismeasure of Man: Distortions and Misconceptions Regarding a Pioneering Mental Test

In The Mismeasure of Man, Stephen Jay Gould argued that the preconceived beliefs and biases of scientists influence their methods and conclusions. To show the potential consequences of this, Gould used examples from the early days of psychometrics and allied fields, arguing that inappropriate assumptions and an elitist desire to rank individuals and/or groups produced incorrect results. In this article, we investigate a section of The Mismeasure of Man in which Gould evaluated the Army Beta intelligence test for illiterate American draftees in World War I. We evaluated Gould’s arguments that the Army Beta (a) had inappropriate content, (b) had unsuitable administration conditions, (c) suffered from short time limits, and (d) could not have measured intelligence. By consulting the historical record and conducting a pre-registered replication of Gould’s administration of the test to a sample of college students, we show that Gould mischaracterized the Army Beta in a number of ways. Instead, the Army Beta was a well-designed test by the standards of the time, and all evidence indicates that it measured intelligence a century ago and can, to some extent, do so today.

The fault in his seeds: Lost notes to the case of bias in Samuel George Morton's cranial race science

The discovery of nearly 180-year-old cranial measurements in the archives of 19th century American physician and naturalist Samuel George Morton can address a lingering debate, begun in the late 20th century by paleontologist and historian of science Stephen Jay Gould, about the unconscious bias alleged in Morton's comparative data of brain size in human racial groups. Analysis of Morton's lost data and the records of his studies does not support Gould's arguments about Morton's biased data collection. However, historical contextualization of Morton with his scientific peers, especially German anatomist Friedrich Tiedemann, suggests that, while Morton's data may have been unbiased, his cranial race science was not. Tiedemann and Morton independently produced similar data about human brain size in different racial groups but analyzed and interpreted their nearly equivalent results in dramatically different ways: Tiedemann using them to argue for equality and the abolition of slavery, and Morton using them to entrench racial divisions and hierarchy. These differences draw attention to the epistemic limitations of data and the pervasive role of bias within the broader historical, social, and cultural context of science.

Morton, Gould, and Bias: A Comment on "The Mismeasure of Science"

A closer look at Stephen Jay Gould’s criticisms of Samuel Morton vindicates Gould’s accusations of racial bias in Morton’s cranial measurements.

Thinking about populations and races in time

Biologists and philosophers have offered differing concepts of biological race. That is, they have offered different candidates for what a biological correlate of race might be; for example, races might be subspecies, clades, lineages, ecotypes, or genetic clusters. One thing that is striking about each of these proposals is that they all depend on a concept of population. Indeed, some authors have explicitly characterized races in terms of populations. However, including the concept of population into concepts of race raises three puzzles, all having to do with time. In this paper, I extend the causal interactionist population concept (CIPC) by introducing some simple assumptions about how to understand populations through time. These assumptions help to shed light on the three puzzles, and in the process show that if we want to understand races in terms of populations, we will need to revise our concept(s) of race.

The mind, the lab, and the field: Three kinds of populations in scientific practice

Scientists use models to understand the natural world, and it is important not to conflate model and nature. As an illustration, we distinguish three different kinds of populations in studies of ecology and evolution: theoretical, laboratory, and natural populations, exemplified by the work of R. A. Fisher, Thomas Park, and David Lack, respectively. Biologists are rightly concerned with all three types of populations. We examine the interplay between these different kinds of populations, and their pertinent models, in three examples: the notion of "effective" population size, the work of Thomas Park on Tribolium populations, and model-based clustering algorithms such as Structure. Finally, we discuss ways to move safely between three distinct population types while avoiding confusing models and reality.