Anda di halaman 1dari 11

Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

Contents lists available at ScienceDirect

Studies in History and Philosophy of Biological and


Biomedical Sciences
journal homepage: www.elsevier.com/locate/shpsc

Putting humanity back into the teaching of human biology


Brian M. Donovan
Stanford Graduate School of Education, Stanford University, 485 Lasuen Mall, Stanford, CA 94305, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Available online xxx

In this paper, I draw upon debates about race in biology and philosophy as well as the concepts of
ineliminable pluralism and psychological essentialism to outline the necessary subject matter knowledge
that teachers should possess if they desire to: (i) increase student understanding of scientic research on
genetic and behavioral variation in humans; and (ii) attenuate inegalitarian beliefs about race amongst
students.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Race
Genetics
Biology education
Essentialism

When citing this paper, please use the full journal title Studies in History and Philosophy of Biological and Biomedical Sciences

Scholars claim that racism should be challenged through


biology education. Within the eld of science education, Zeidler,
Sadler, Berson, and Fogelman (2002) advocate teaching about the
history of cultural prejudice in scientic research. Castra, Sarapuu,
and Clment (2013) and Puig and Jimnez-Aleixandre (2011) argue
that school science should challenge genetic determinism in order
to undermine racism or ethnocentrism. Donovan (2015) argues that
race should be directly discussed as a topic of biology textbooks in
order to increase understanding of human evolution and decrease
racial prejudice. Finally, Willinsky (1998) claims that it is irresponsible for the biology curriculum to ignore race in light of sciences contribution to racial prejudice. Instead, he argues school
biology should teach how racial ideas were socially constructed in
the history of science. Despite such arguments no one has outlined
the body of subject matter knowledge that science teachers ought
to know to accomplish these goals.
In the post-genomic era this oversight appears problematic. The
frequency of articles discussing genetic differences between races
has increased in the news media over the last twenty years (Phelan,
Link, & Feldman, 2013). Research also demonstrates that such articles
can strengthen racist attitudes amongst Americans (Condit, Parrott,
Bates, Bevan, & Achter, 2004; Phelan et al., 2013). Furthermore,
popular science books written for lay audiences, such as Wades
(2014) Our Troublesome Inheritance: Genes, Race, and Human History

have advanced the scientically specious thesis (see Coop, Eisen,


Nielsen, Przeworski, & Rosenberg, 2014) that complex human traits
and political systems differ between races because of recent human
evolution. If exposed to such media, then science teachers probably
possess biased beliefs about race, which makes them ill-equipped to
teach about scientic debates surrounding the nature of race.
Even the curriculum that science teachers inherit appears to
discuss race in an inappropriate manner by failing to address scientic controversies surrounding race (Donovan, 2015; Morning,
2008). Early in the 20th century biology textbooks directly taught
students that races were biological subdivisions of the human
species and some texts also taught students about a racial hierarchy
(Morning, 2008). Today, textbooks address race indirectly by
referring to race or racial categories in passing but not as an
explicit focus of learning (Morning, 2008). Usually these subtle
references occur in chapters discussing forensics, genetic diseases,
and human evolution (Morning, 2008). Furthermore, evidence
from a eld experiment carried out in eighth grade science classrooms demonstrated that when students encounter these subtle
references to race in the modern biology curriculum it can lead
students to agree more strongly that races differ in complex human
traits (e.g. academic ability & artistic ability) because of genetics
(Donovan, 2014). Put differently, there appears to be a hidden racial
curriculum in biology textbooks that is learned by students but
never purposefully taught by teachers. And arguably, this curriculum reinforces potentially prejudiced beliefs about race that are not
supported by biological theory or research (Donovan, 2015).

E-mail address: bdonovan79@gmail.com.


http://dx.doi.org/10.1016/j.shpsc.2015.01.011
1369-8486/ 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

The purpose of this paper is to argue for the necessary subject


matter knowledge (SMK) about human races that science educators should possess if they want to teach about genetic and
behavioral variation in human populations without strengthening
inegalitarian beliefs about race. Importantly, this argument does
not outline what science teachers should teach about race or what
students ought to know about race and biology. Nor, does this paper
explain the way in which teachers should represent and formulate
their SMK about race so that it is comprehensible to students. In
other words, it does not outline the pedagogical content knowledge
(PCK) (Shulman, 1986) that teachers need to possess to teach about
race effectively. By outlining a body of requisite SMK about race,
however, this paper provides a foundation for determining the PCK
required for teaching about race through school science. The paper
establishes what might be called the necessary but not sufcient
subject matter knowledge for teaching about race. What follows is
not the nal word on this subject, but it is the rst argument in
what could be a long debate. In brief, the paper makes an argument
that the following four components be included in the necessary
SMK for teaching about race:
1. Psychological essentialism is associated with a misunderstanding of human genetic diversity and the belief that racial
inequality is immutable.
2. Biologists can apply different methods to the same human genetic data and infer different things about the reality of race.
3. Philosophers who disagree about the reality of race can each be
concerned with eliminating racial inequality.
4. Scientic research on human behavior cannot be used to
conclusively support the essentialist claim that racial inequality
is immutable for biological reasons.
Component one argues that a cognitive bias called psychological
essentialism leads people to believe that races are biological subdivisions of the human species that possess different genetic predispositions for behavior. Consequently, psychological essentialism
leads people to believe that social inequalities between races are
natural and immutable. Given this problem, educators might be
tempted to claim that race is not biological in order to challenge
essentialist beliefs about race. Yet, in component two it is argued
that biologists can look at the same genetic data and infer different
things about the reality of race. Therefore, educators should be
cautious about appealing to biological data to claim that race is not
real. The reality of race is a question best answered through philosophy. Component three explains that philosophers who argue
that race is biologically real and those who argue that it is socially
real can disagree about the nature of race but still agree that racial
inequality should be eliminated. Thus, there are alternatives to
essentialist thinking that exist in the philosophy of race, but it is not
the case that all of the alternatives concerned with eliminating
racial inequality are also anti-biological. Finally, in component four
it is argued that science teachers need to know that there are
ontological inconsistences and epistemological limitations in human behavioral research that make it extremely difcult, if not
impossible, to denitively claim that racial inequality is immutable
on the basis of human behavioral research. In summary, knowledge
of these four components makes it difcult for the science educator
concerned with the elimination of racial inequality and the scientic racist concerned with the justication of inequality to appeal to
the authority of science to support a racial ideology. To know how
the issue of racial equality can be divorced from the scientic study
of human difference is the necessary SMK knowledge required to
teach about race.
Before beginning, however, a denitional point should be made.
The term racial inequality refers to a state of social affairs in which

members of different races do not enjoy similar social standing (e.g.


economically, educationally, etc.). The goal of teaching about race
through science education is to improve students understanding of
genetic and behavioral variation in humans without increasing
racial inequality. And, if this is the goal to which teachers aspire
then they ought to know about psychological essentialism, because
it is associated with misunderstanding of intraspecic genetic
variation and it leads individuals to believe that racial inequality is
immutable, and therefore, not worthy of redress.
1. Component one
The etiology of American racial thinking is not perfectly understood. Descriptive studies demonstrate that American children
demarcate humans on the basis of their spoken language rather
than race in early childhood (Kinzler & Dautel, 2012). But during
late childhood this propensity appears to transform into the tendency to distinguish human groups using racial phenotypes
(Kinzler & Dautel, 2012). Experimental studies demonstrate that
beliefs about social groups are perpetuated culturally through the
use of language (Rhodes, Leslie, & Tworek, 2012). Thus, ideas about
race communicated through writing and speaking appear to be
important factors in the formation of racial beliefsdmore important than the observation that human skin color varies. In other
words, children do not learn about race by passively observing
variation in human skin color rather they construct theories about
race as they encounter representations of race in culture
(Hirschfeld, 2012). A theory of race common in American culture
that is implicated in inegalitarian racial thinking is essentialist
thinking about race (Morning, 2011).
Studies document that children and adults in cultures around
the world tend to think about social and biological categories as if
they are committed to metaphysical essentialism (Gelman, 2004;
Henrich, Heine, & Norenzayan, 2010; Hirschfeld, 1998; Prentice &
Miller, 2007). In other words, people act as if biological and social
categories possess an underlying essence that causally determines
the properties of the organisms in that category (Medin & Ortony,
1989). Racially speaking, this essence can be biological, cultural,
or merely the belief that scientists will one day nd the underlying
essences that determine the physical traits, psychological temperaments, and abilities of different races (Morning, 2011). Thus,
psychological essentialism of race can be, and often is, equated with
socially or biologically deterministic beliefs about human difference (Keller, 2005; Rangel & Keller, 2011).
Unsurprisingly, then, it is not uncommon for researchers to
operationalize racial essentialism as the belief that race is biological, or the belief that racial difference is genetic, or the belief that
races are natural kinds, or the belief that races possess different
cultural essences. Indeed, racial essentialism can mean so many
different things in the literature on racial beliefs that it is important
to carefully dene the beliefs that constitute this construct if one
intends to discuss it (Glasgow, Shulman, & Covarrubias, 2009).
Social-psychologists (e.g. Bastian & Haslam, 2006; Haslam,
Rothschild, & Ernst, 2002; Williams & Eberhardt, 2008) that study
the social consequences of essentialist thinking on variables such as
stereotyping, prejudice, and interethnic socialization, operationalize essentialism with two sets of beliefs: natural kind thinking and
entitative thinking. Studies employing factor analysis also provide
empirical support for the conceptual distinctness of these two
components of essentialist thinking (Haslam, Rothschild, & Ernst,
2000).
The natural kind dimension of essentialism involves the belief
that races are biological categories. To be precise, the natural kind
component includes beliefs about the discreteness, immutability,
naturalness, and stability of racial categories as well as beliefs about

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

the necessary characteristics that a person must possess to be a


member of a race (Haslam et al., 2000). Discreteness is the belief
that some categories, like racial categories, have sharper boundaries than others. Immutability, on the other hand, is the belief that
membership in a race is not easy to change because it is xed.
Stability is the belief that racial categories are more stable over time
than other social categories, and thus, have not changed much
throughout history. Finally, necessity is the belief that each race
possesses necessary features or characteristics, and without these
characteristics, someone is not a member of that race.
In contrast, entitativity involves beliefs about the uniformity,
inherence, exclusivity, and informativeness of racial categories
(Haslam et al., 2000). Uniformity is the belief that racial categories
contain members who are very similar to one another. Consequently, racial category membership is informative about an individual because such categories allow people to make many
judgments about the unknown characteristics of their members.
That is, racial categories permit one to stereotype individuals
accurately. Inherence is the belief that racial categories have an
underlying reality, which means that although members of a racial
group may have similarities and differences on the surface, underneath they share the same essence. Finally, exclusivity is the
belief that some categories do not allow their members to belong to
other categories. It is the belief that you can only be a full member
of one race.
A critical consequence of essentialism is that it causes individuals to perceive less variation within racial groups and more
variation between racial groups (Chao, Hong, & Chiu, 2013;
Hirschfeld, 1998; Prentice & Miller, 2007). These perceptual biases
also increase the tendency of people to engage in race-based categorizations (Chao et al., 2013). This means that essentialist beliefs
cause individuals to make more categorical distinctions when
perceptual input involves continuously varying phenotypic features
(e.g. skin color). Thus, psychological essentialism is a cognitive bias
that facilitates social stratication based on race, because it causes
individuals to categorically differentiate humans into discrete races.
Experimental studies show that the belief that social categories
possess cultural or biological essences is implicated in prejudice,
stereotyping, and perceived group homogeneity (Keller, 2005;
Rangel & Keller, 2011). Essentialist beliefs can cause one to hold
prejudiced positions on racial issues, and conversely, the desire to
rationalize prejudiced positions on social issues can lead one to
essentialize race. Thus, the causal relation of prejudice and essentialism is bidirectional (Keller, 2005; Rangel & Keller, 2011).
Experimental research with undergraduates also demonstrates
that natural kind beliefs about race attenuate interest in interethnic
socializing whilst also decreasing concern about racial disparities
(Williams & Eberhardt, 2008). Furthermore, entitativity is predictive of anti-black attitudes (Haslam et al., 2002). For example,
representative studies of the White American population indicate
that the extent to which White Americans believe that genes
explain race differences is predictive of anti-black attitudes
(Jayaratne et al., 2006). Other studies suggest that 20% of non-Black
Americans believe that genes explain economic disparities between Whites and Blacks (Brueckner, Morning, & Nelson, 2005).
Even amongst European-American elementary school children, the
belief that races are natural kinds is an important predictor of racial
stereotyping (Pauker, Ambady, & Apfelbaum, 2010).
Last but not least, psychological essentialism appears to be
associated with misunderstanding of intraspecic variation. Science education researchers have argued that psychological essentialism leads individuals to devalue within-species variation and
fail to understand natural selection (Evans et al., 2010; Opfer, Nehm,
& Ha, 2012; Shtulman & Schulz, 2008). Opfer et al. (2012), for
instance, found that essentialist biases were negatively correlated

with proper understanding of differential survival, variation, and


heredity in a sample of undergraduate participants (N 320) in a
series of introductory biology courses. They also found that belief in
essentialism was positively correlated with evolutionary misconceptions and negatively correlated with end of semester biology
grades. Because human races are understood as biological subdivisions of the human species (i.e. sub-species) (Lieberman, 1997),
and because essentialism causes individuals to ignore intra-racial
phenotypic variation (Chao et al., 2013; Hirschfeld, 1998), psychological essentialism is arguably an impediment to a proper understanding of intraspecic genetic variation in humans.
Thus, knowledge of psychological essentialism is the rst piece
of SMK required to teach about race, because, one ought to know
that essentialist beliefs are predictive of racial bias and misunderstanding of intraspecic diversity if one wishes to educate students
about human genetic variation without increasing prejudice.
Naturally then, the science educator might be tempted to explain to
students that biological data does not support the idea that races
are biologically real. Because if races are not real, the reasoning
goes, then there is no logical basis for discriminating on the basis of
race. Indeed, science educators may have heard that there is more
average genetic variation within racial groups than between racial
groups, and therefore, race cannot be real. But even though psychological essentialism of race (or other living categories for that
matter) is inconsistent with biological theory and data (Donovan,
2015; Ereshefsky, 2010; Mayr, 1982), educators should be careful
about appealing to data to claim that race is not biologically real.
For, biologists can apply different methods to the same human
genetic data and infer different things about the reality of race.
2. Component two
There are essentially four things that a science educator ought to
know about population genetics if they want to teach about race: (i)
there is more average genetic variation within populations than
between populations, which has been interpreted by scholars to
mean that race is not biological, or not real; (ii) there is a population
structure in humans that differentiates populations (often dened
linguistically or geographically) into genetic clusters, which has
been interpreted by scholars to mean that race is biologically real;
(iii) the previous two ndings can be veried by investigating the
same data using different population genetic methods (Winther,
2014); (iv) these ndings lie at the crux of most contemporary academic debates about the reality and nature of race.
Lewontin (1972) demonstrated the rst point and Edwards
(2003) argued the second point using the data, methods, and
ndings of Rosenberg et al. (2002). Consequently, this debate about
race could be named the Lewontin-Edwards debate and much has
been written about the methodological and normative substance of
it (e.g. see Kaplan & Winther, 2013a; 2013b). The heart of the debate
has to do with the fact that when geneticists average gene frequencies across loci and then compare populations they nd that
there is more average genetic variation within populations than
among populations. This means that if we randomly pick two individuals from two different racial groups and compare them to
two randomly picked individuals from the same racial group, we
can expect that the former will only be slightly more genetically
different from one another, on average, than the latter. Furthermore, through the method of diversity partitioning (Winther,
2014) that Lewontin (1972) used, one can arrive at a precise
quantication of the amount of average genetic difference across
continents commonly associated with racial groups (e.g. BlackAfrica, White-Europe, Asian-East Asia, Native AmericansdThe
Americas, etc.). For example, 93.2% of genetic differences between
randomly chosen individuals occurs within populations, 2.5% of

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

such differences occur across populations within continental regions, and 4.3% of the average genetic difference between any two
humans can be attributed to differences across the continental
regions (Rosenberg et al., 2002) associated with American census
racial categories (read Edge & Rosenberg, in press, for more on the
method of diversity partitioning).
But if one looks at the accumulated differences without averaging across loci something different occurs. Genetic clustering
techniques assign an individuals genome to a theoretical population using the accumulated information present in a genome
(Winther, 2014; Winther, Giordano, Edge, & Nielsen, in press). For
example, if the frequency of an allele, lets call it identity allele A, is
47% in Europe, 34% in Asia, and 40% in Africa, then knowing
whether a person possesses identity allele A does not permit a
condent inference about whether that persons ancestry is Asian,
African, or European. But imagine a scenario in which we discover
that our racially uncategorized person also possesses identity allele
B and that this allele occurs at a frequency of 60% in European
populations and at a frequency 40% in African populations and 44%
in Asian populations. With these two pieces of information we can
be slightly more condent that the ancestry of our imaginary individual is European because identity alleles A and B are most
prevalent in this region. When geneticists accumulate all of the tiny
frequency differences in genetic polymorphisms between populations without averaging across loci, then they are able to
differentiate humans into genetic clusters using probability models
similar to, although much more complex than, the previously
described heuristic (read Edge & Rosenberg, in press; Winther et al.,
in press). Moreover, when geneticists apply genetic clustering
techniques to 300 or more loci the possibility of misclassifying a
person into a population is effectively zero (Edwards, 2003;
Witherspoon et al., 2007). Thus, using the same data but applying
a different technique, population geneticists can demonstrate that
there is a population structure in humans.
Lewontin (1972) used the rst empirical nding produced
through diversity partitioning techniques to argue that racial
categorization is of virtually no scientic use and that given the
social dangers of biological thinking about race, scientists should
stop using racial categories in their research. Edwards (2003) disagreed with Lewontin (1972). Edwards (2003) used ndings from
genetic clustering analyses to argue that: (i) genetic data is of value
to racial classication; (ii) two individuals are not different just
because they are individuals, but also because they belong to
different races; (iii) you can predict someones race from their
genes. This does not mean, however, that the clusters identied by
clustering methods are natural populations of humans (read
Winther et al., in press). Or, that human populations are made of
individuals sharing the same underlying genetic essence. In fact,
only 7.53% of all human alleles are private to any single geographic
region and 46.6% of all alleles are present in every geographic region (Rosenberg, 2011). So, even though population geneticists can
assign a persons genome to a genetic cluster that corresponds with
a continental region this does not mean that human populations
are genetically discrete or that individuals within a population are
genetically uniform. Thus, genetic clustering methods do not
establish the validity of essentialist beliefs about race.
Nevertheless, in the apparent disagreement between Edwards
and Lewontin is the observation that researchers can apply
different methods to the same biological data and infer different
things about the reality of race (Kaplan & Winther, 2013a; 2013b).
This conundrum raises important questions about whether genetics can determine the reality of race. For instance, how much
average genetic variation across continents is adequate to dene
racial groups? If humans differ from chimpanzees in 0.1% of their
DNA, then is the current estimate of 4.3% (of the 0.1%) of

intercontinental variation enough to consider human races a


meaningful subdivision of the human species? If the mathematical
algorithms that produce genetic clusters demonstrate that humans
can be divided into 2, 3, 4, 5, or 6 genetic clusters (Rosenberg et al.,
2002), then which of these clustering schemes is the correct racial
classication of humans? Winther (2014) argues that the answers
to these questions are not found in the data but in our culturally
specic interpretations of that data. Thus, biology educators should
be skeptical when they encounter authors such as Wade (2014),
who make strong racialist claims by appealing to studies of genetic
diversity, because for every racial reading of data there is also an
anti-racial reading of the same data.
The key point is that scientic data is often used to reify, or
augment, previously held racial beliefs (Kaplan & Winther, 2013a,
2013b; Winther, 2014). Throughout history people have appealed
to the authority of science to justify the validity of their racial beliefs
(Appiah, 1996; Hacking, 2005; Morning, 2011). If teachers understand that biological data does not denitively answer the question
of whether human races are biologically real, then they will be able
to help students understand how seemingly conicting ideas about
the reality of race can be supported by the same genetic data set. In
sum, human population genetics does not denitively answer
whether race is real, rather it shows that there are biological differences between and within populations. The differences within
populations suggest that essentialism is biologically awed, because
individuals within a race do not each share the same genetic essence.
Nevertheless, teachers should be cautious about using biological
data to show students that race is not real. The reality of race is a
question more appropriately addressed through philosophy.
3. Component three
Philosophical arguments about the reality of race can be taxonomically categorized. For example, Spencer (2012) classies
metaphysical stances on race into views about racial anti-realism
and racial realism. Racial anti-realists1 claim that race is not real
whereas racial realists claim that race is real. The racial realists are
further subdivided into those who subscribe to biological racial
realism (BRR) and those who subscribe to social racial realism
(SRR). Crudely speaking, biological racial realists afrm the biological reality of race in one way or another, whereas social racial
realists tend to deny certain claims about the biological reality of
race while afrming the social reality of race. Haslanger (2008)
employs a similar taxonomy for grouping academic arguments
about race. She uses the labels of racial constructionism and racial
naturalism to distinguish realist conceptions of race (Haslanger,
2008). Racial naturalism is roughly equivalent to BRR and racial
constructionism is the analog of SRR. The key idea in the philosophical race debate is that philosophers who disagree about the
reality of race can each be concerned with eliminating racial
inequality. To understand this claim, however, one needs to know
how much conceptual variation exists within and between the
different realist positions on race.

1
In contrast to these realist positions on race is the anti-realist position. Racial
eliminativism claims that human races are not real and therefore race should be
eliminated from our social practices because it underwrites racism (Appiah, 1996;
Haslanger, 2008). Thus, racial eliminativism could be grouped with racial antirealism because this position claims that race is not real. Yet, one could be a
racial anti-realist without agreeing that race should be eliminated because it is
socially destructive. Hence, racial eliminativism and racial anti-realism is not
necessarily the same thing. Whether one believes race is real or not real, it should
be clear that racial equality is an implicit value in the eliminativist tradition of racial
anti-realism. However, because of the dangers of colorblindness (Neville, Lilly,
Duran, Lee, & Browne, 2000) it may be unwise to promote racial anti-realism in
the science classroom.

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

3.1. Racial constructionism or SRR


Social constructionism explains how social artifacts come into
existence. For example, Haslanger (1995) writes that social artifacts
can be understood as (see pp. 97e100): (i) generic social constructions (which is the idea that something is a social construction
if it is an intended or unintended product of social practice); (ii)
causal constructions (which is the idea that something is a social
construction if and only if social factors play a causal role in
bringing it into existence or to some substantial extent, in its being
the way it is); (iii) constitutive constructions (which is the idea that
something is socially constructed if and only if in dening it we
must make reference to social factors); (iv) discursive constructions
(something is constructed just in case it is the way it is, to some
substantial extent, because of what is attributed and/or self
attributed to it); (v) pragmatic constructions (when something is
constructed through a classicatory apparatus whose use is
determined, at least in part, by social factors).
Since racial classication schemes vary across cultures and
through time (Morning, 2011) race could be considered a pragmatic
construction. Racial disparities are also the intended and unintended products of American social practices (Morning, 2011).
Thus, races could be generic social constructions. Likewise, race
might be causally constructed because American society discriminates on the basis of biological traits like skin color. And, since the
threat of being stereotyped can make racialized individuals
behave in a manner consistent with racial stereotypes (Steele &
Aronson, 1995) race could be a discursive construction as well. In
general, constructionists argue that race is real because it was
brought into existence through historical acts and because it is
maintained by social and political practices today.
This does not mean, however, that all forms of constructionism
deny biological differences between races. Take, for example,
Hackings (2005) argument that race is constructed through the
interplay of human culture and cognition, social taboos against
miscegenation, and imperialism. Although he never explicitly
states that race is a social construction, Hacking (2005) argues that
each of the previously mentioned factors construct race. And, while
Hacking (2005) would appear to agree that racial essentialism is
not supported by biological data (see pp. 104e105), he still argues
that there are biological differences between races that are statistically signicant, biologically meaningful, and medically useful. For
instance, Hacking (2005) argues that race-specic bone marrow
registries are ethically justied because human leukemia antigens
differ across races. Thus, Hackings argument exemplies the
notion that one can agree that race is socially constructed and that
there are biological differences between races.
It is also important to note that there are two reasons why social
constructionism is not cultural essentialism. First, constructionists
do not believe that every member of a race shares an underlying
cultural essence that makes the race uniform in their behavioral
characteristics. Second, as Haslanger (2000) claims, constructionism
is concerned with developing accounts of race that are effective in
ghting injustice. Cultural essentialism, on the other hand, is not
concerned with this goal. It employs the concept of a cultural essence
to rationalize the intransigence of social inequalities between groups
(Rangel & Keller, 2011). The main point is that individuals who subscribe to social constructionism are social racial realists who are
concerned with eliminating (not justifying) racial inequality.
3.2. Racial naturalism or BRR
In contrast to the constructionist position, biological racial realism (BRR) (Spencer, 2012), or racial naturalism (Haslanger, 2008),
argues that races are biologically real. BRR asserts that there is a

stable mapping between socially dened races and genomic measures of races (Kaplan & Winther, 2013b). Some biological racial
realists assert that genetics or biology justify racial inequalities
observable in society (Kaplan & Winther, 2013b). Thus, BRR could
be used to support inegalitarian claims that racial disparities are
natural (hence, the term racial naturalism). Even so, BRR is not the
same thing as biological essentialism and it too can be concerned
with eliminating inequality.
As Spencer (2012) points out, there are different conceptions of
BRR. Some philosophers dene BRR as the conception of objectively
real natural kinds, which are kinds that are said to exist independent of human cognition. Other philosophers have dened BRR as
the conception of an inductively useful natural kind, which is the
idea that racial categories allow for inductive generalizations in
biological science. Still other positions maintain that BRR refers to
useful kinds (races are merely those kinds which are studied by
biological science), pragmatic kinds (races are valuable for research
in a scientic context in so far as the utility of races for research is
not outweighed by the potential damage race might play in contexts outside of scientic research), real biological kinds (races
occur independently of human interest and individuals within
races share common ancestry), and real principled biological kinds
(races have a principled biological basis). To these conceptions
Spencer (2012) adds his idea that races are genuine kinds because
they are useful for underwriting observational laws, theories, and
presuppositions in a well-ordered scientic research program.
Kaplan and Winther (2013b) also argue for an additional form of
biological racial realism called, bio-genomic cluster racial realism. It
is based upon the nding that there is a population structure in the
human species that can be assessed through genetic clustering
studies. However, bio-genomic cluster realism does not assert that
such clusters are evidence for the genetic inferiority or superiority of
different racial groups. Nor does it assert that there is a genetic
explanation for racial inequality. Like Spencers conception of races as
genuine kinds, bio-genomic cluster realism is not concerned with
explaining the naturalness of racial inequalities. Rather it is merely
committed to the idea that there is a population structure in humans.
Thus, a major difference between biological essentialism and
some forms of BRR is that some scholars writing about the biological reality of race do not presume racial inequality is a natural
consequence of biological differences between races. For example,
after proposing a populationist concept of race (PRC) for use in
biological research, Hardimon (2012) concludes:
One striking disadvantage of the denial of the existence of biological races in human beings is that it makes it impossible to
say that biological race is unimportant (existence being a
precondition of unimportance). Given the salience of differences
in patterns of visible physical characters associated with differences in geographical ancestry and the enduring widespread
propensity to racialize themdto treat them as correlated with
humanly important traitsdthis is a notable shortcoming. The
PRC, on the other hand, provides us with the discursive means of
asserting the unimportance of biological race. That is no small
thing. (p. 274)
Clearly, Hardimons (2012) population concept of race conicts
with the essentialist belief that genetic essences are responsible for
humanly important differences between races. Hardimons
(2012) conception of BRR is not an attempt to justify the inevitability of racial inequality by claiming that races differ in humanly
important ways because of biology. Rather it states that the biological differences that exist between races, such as variation in
skin color, are unimportant. These differences do not, contrary to
popular opinion, serve as markers of deeper intellectual or

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

behavioral differences between races that make racial inequalities a


natural and unxable problem.
At the same time, some forms of BRR appear to question the
possibility that racial inequality can be eliminated. Sesardic (2010),
for example, argues that racial differences in psychological traits
(such as intelligence) and behavioral traits (such as criminality) are
attributable to genetic differences between races. About genetic
clustering data Sesardic (2010) writes, with these new studies it
becomes harder to accept the widespread but often unsubstantiated claim about the biological meaninglessness of race (p. 154).
He goes on to claim that clustering studies of phenotypic traits
show that races are morphologically different. Finally, he argues
that, some of the psychological differences between the races .
are at least partly due to genetic differences (Sesardic, 2010, pp.
157e158). To be fair, Sesardic (2010) never argues that racial inequalities are natural in his 2010 publication on BRR. In fact, he
states that the essentialist belief that races can be characterized by
properties that are shared by all and only the members of a race
(Sesardic, 2010, p. 146) is unjustiable. Nevertheless, Sesardics
(2010) particular conception of BRR would be well received by an
essentialist thinker who is unaware of the ecological fallacy and
who wishes to justify the intransigence of racial inequality through
biology. Sesardics (2010) conception of BRR is clearly different from
Hardimons (2012) conception of BRR even if Sesardic disagrees
with essentialism.
The key insight for science teachers in the philosophy of race is
that while scholars might disagree about the reality of race they can
still be concerned with eliminating racial inequalities. Social constructionists, such as Haslanger (2000), and biological racial realists, such as Hardimon (2012), have developed accounts of race
that are useful for ghting racial injustice. Knowledge of their work
is useful because if teachers know that there different conceptions
of race that are concerned with the elimination of racial inequality
then they will be able to offer students alternatives to an essentialist view of human racial difference. Furthermore, this knowledge provides a framework for helping students understand how
individuals can infer different things about the reality of race from
the same genetic data. That is, such people might be interpreting
the data with differing prior assumptions about race.
4. Component four
Inevitably, however, the teacher of race will encounter a student
who expresses the essentialist opinion that racial inequality is
immutable because of biological differences between races. This
teacher may also encounter a student who believes that science has
authoritatively established this truth (perhaps because he read
Wade, 2014). To educate this student the science teacher needs to
know more than components two and three. They need to be able
to point out the weaknesses in this students claim. Thus, the
teacher must know that scientic research into human behavior
cannot be used to convincingly argue for the immutability of racial
inequality because: (i) there is no scientic concept of behavior to
support strong stereotypical claims about the general behavioral
predispositions of a race; (ii) ndings from human behavioral
research do not provide a comprehensive account of the causes of
human behavior.
4.1. Problems with the concept of behavior
The rst reason there is no denitive scientic reason to believe
that racial inequality is immutable is that there is no agreed upon
denition of behavior in science. For example, Levitis, Lidicker, and
Freund (2009) investigated how 174 members of three behaviorfocused scientic societies understood the term behavior. They

found that behavioral biologists could not agree upon a single


denition of behavior. High rates of inconsistencies in the denitions of behavior among novices and experts in behavioral biology
were also found. A similar argument has been made that psychologists possess no shared denition of behavior (Bergner, 2011).
Even racially stereotyped behaviors, such aggressiveness, have no
consistent operational denition in human behavioral research
(Longino, 2013). Additionally, Longino (2013) shows that when
scientists overcome the problem of dening aggression there is still
the issue that only a subset of aggressive behaviors are ever studied
and these are often disorders that affect only a small portion of the
human population. As a consequence, studies on aggressive
behavior are usually uninformative of the general human aggression that research seeks to understand (Longino, 2013).
If Longino (2013) is correct, then it is tenuous to claim that
science has anything denitive to say about stereotypical claims
about the behavioral predispositions of a race. For example, if scientists cannot agree upon the denition of aggressiveness and if
researchers do not operationalize aggressiveness in the same
manner, then how can anyone use science to conclusively support
the claim Black people are inherently aggressive and that racial
disparities in the criminal justice system are therefore immutable?
If a person uses research to support the strong claim that Science
shows Black people are inherently aggressive then they have
overstated their case. For, if one were to investigate how aggressiveness is dened in the studies used to support such a position it
will be clear that such studies have not investigated general
aggressiveness.
What the person might be able to claim is a highly qualied
statement, such as: amongst individuals racially classied as Black
in study X, which investigated population Y, researchers found a
correlation between a gene variant and a psychological instrument
known as Z. Or, the person might be able to say that in a study that
investigated people from population A, which varied in measured
characteristics such as BeG, an average difference on the psychological instrument Y was found between racially categorized individuals. Neither statement comes close to the wording Science
shows that Black people are inherently aggressive and both
statements are correct only if the studies that support them met the
necessary statistical assumptions to produce an internally valid
correlation between the variables of interest. Indeed, if one were to
read further into the literature to evaluate the claim Science shows
that Black people are inherently aggressive they would likely
encounter papers questioning whether the studies used to initially
support this claim met the requisite statistical assumptions. Or,
they might nd a study which used a different measure of
aggressiveness and which showed no difference between racial
groups or no correlation between genes and aggressiveness.
The point is that the conceptual incoherence of the behavior
concept in scientic research makes it difcult to appeal to science
to support strong stereotypical claims about the general behavioral
predispositions of a race. For, if there is no scientically agreed
upon denition of behavior then there cannot be general behavioral predispositions of a race. Furthermore, if there is no agreed
upon denition of behavior in science then it is difcult to make
sense of contradictory ndings pertaining to race and behavior.
Such difculties make it nearly impossible to use scientic ndings
to denitively evaluate the veracity of racial stereotypes. Even so,
one might argue from this weakened position that average
behavioral differences between races are still socially signicant
even if the research only applies to a small population and a limited
form of behavior. One might also make a deterministic argument
that such differences are caused by biological or by cultural factors
that are not subject to change. In order to see why this kind of
argument is tenuous from a scientic perspective it is important to

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

understand the different approaches of human behavioral research


and their limitations. What follows is a brief summary of some of
these approaches (using the terminology of Longino, 2013).
4.2. Approaches to behavioral research
Quantitative behavioral genetics (QBG). QBG partitions human
phenotypic variation into additive genetic factors, the environment,
geneeenvironment interactions, and geneeenvironment correlations (Sober, 2001). Typically QBG collects data on human behavior
through rst or third person methods. For example, a researcher
might have people complete psychological instruments that target
anti-social personality disorder or they might gather teacher
observations on anti-social behavior. The population of study is
almost always twins reared together or apart, adopted siblings
reared in the same environment, or adopted siblings reared in
different environments. The underlying reasoning of the QBG
method is to compare the phenotypic similarity of such dyads to
determine how behavioral variation correlates with the shared or
non-shared environments. This gives researchers an idea of how
much phenotypic variation might be attributable to underlying
genetic variation. For example, if a high degree of concordance in
aggressive behavior between identical twins reared apart is
discovered then this would suggest that aggressiveness is heritable.
However, this conclusion can only be reached provided the instruments are valid measures of aggressiveness. Furthermore, QBG
estimates of heritability are unbiased only if the researchers
establish that there are no gene-environment interactions or genee
environment correlations inuencing the expression of the trait
(Sober, 2001). Both of these assumptions are difcult to meet, but
such requirements do not preclude the possibility of establishing
the amount of phenotypic variation in a trait that is heritable
(Sober, 2001). Importantly, however, heritability is not the same
thing as genetic determination (Block, 1995). It does not tell researchers which genes cause a behavior. Rather it tells researchers
the proportion of behavioral variation attributable to genetic
variation.
Molecular behavioral genetics (MBG). The strength of the MBG
approach is that it can identify and evaluate whether particular gene
variants correlate with a behavior. Thus, MBG investigates questions
about the genetic determination of traits because it seeks to identify
the genes that arguably have a causal relationship with behavior.
The general approach of researchers in MBG is to identify and
sequence genes that are hypothesized to affect a given behavior. The
hypothesized geneebehavior correlations studied in MBG are usually derived from gene-knockout studies carried out in animal
models. The candidate genes that cause a behavior in animal model
studies are then identied in humans. Next, researchers try to
correlate the presence or absence of a particular genotype with the
presence or absence of a particular behavioral phenotype in
humans. Because the hypothesized gene variants are likely to be
proportionally low in the human population, samples for MBG
studies are purposefully selected in order to increase the representation of individuals who possess the phenotype of interest
(Longino, 2013). Like QBG twin studies, MBG studies are not representative of the general population. For example, researchers might
conduct their analysis on a kinship group that shows high heritability of an X-linked disorder to improve the statistical power of the
correlational analysis. If a reliable correlation is found between a
gene variant and a phenotype then researchers tend to appeal to
animal model experiments to argue about the putative causal
relationship between the studied genotypes and phenotypes. Thus,
a causal interpretation of the correlational ndings in MBG assumes
that animal model ndings are generalizable to humans. Furthermore, the correlational approach employed in MBG designs means

there is always the risk of spurious genotypeephenotype correlations caused by omitted environmental variables.
Gene-environment interactionism (GEI). This approach integrates
experimental neuroscience and genetic epidemiological approaches to study human behavior. Caspi and Moftt (2006)
describe the basic approach as: (1) identify the neural substrate
involved in the behavior; (2) identify how the environment affects
the neural substrate associated with the behavior; (3) identify how
the genotype affects the neural substrate. The neuroscience allows
one to investigate the association between genetic factors and
neural activity associated with a human behavior. The epidemiological approach allows one to investigate whether there are genee
environment interactions contributing to variation in the behavior
between individuals or populations. Together, this approach can
begin to tell researchers about the biological mechanisms behind a
behavior and how they are moderated by the complexity of humaneenvironment experiences. Interactionism is an attempt to
overcome the limited external validity of MBG. Additionally, it addresses some of the internal validity issues with MBG because
neuroscientists can experimentally manipulate environmental
stimuli amongst humans of varying genotype (instead of lab rats) to
see how it affects neural activity associated with the behavior of
interest. The combination of large sample sizes and carefully
controlled experimental manipulations in laboratories suggests
that interactionism produces a more valid map of the relationship
between genes, environment, and behavior than either QBG or
MBG. However, one might question the external validity of the
environmental inputs studied in neuroscience. How often does the
human environment involve being pinned down in an fMRI machine as one responds to carefully crafted psychological tasks?
Furthermore, the behaviors investigated in this approach tend to be
psychiatric disorders that only a small proportion of humans
exhibit (Longino, 2013). So, it is unclear whether many GEI studies
say anything about everyday human behaviors in the general
population.
Social-environmentalism (SE). The distinguishing feature between SE and the previous approaches is that it is solely concerned
with social and environmental causes of human behavior. It is not
at all concerned with biogenetic or developmental inuences on
behavior (Longino, 2013) even if such causes are theoretically at
work (see Shiao, Bode, Beyer, & Selvig, 2012). SE research is usually
carried out in laboratory settings (i.e. think social psychology). Like
QBG, it employs clinical observations of behavior and psychological
instruments to quantify behavior. Studies may be experimental or
observational. The experimental studies can produce causal estimates of the relationship between a socialeenvironmental cause
and a behavior. However, these same experimental studies may be
incapable of identifying the mechanism that links the cause to the
behavior. Studied populations in SE are usually undergraduates or
preschoolers and their parents in western, industrialized, educated,
rich democracies (Henrich et al., 2010). SE ndings, consequently,
do not generalize to human groups living outside of such societies
(Henrich et al., 2010). And, as with the previous approaches to
research, observational/correlational studies in the SE tradition
produce causal estimates of the relationship between two variables
only when certain methodological assumptions are met (e.g. no
omitted variable bias).
4.3. The immutability of inequality revisited
So, what is the science educator of race to make of these
different approaches to the study of human behavior and the
essentialist argument that racial inequality is immutable because of
biological determinism? The answer to this question depends on
ones interpretation of the limits of scientic knowledge.

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

One way of viewing behavioral research is to believe that it


provides a comprehensive explanation of human behavior. This
view has two different avors: monism and integrative pluralism
(Longino, 2013). The monist argues that a single approach to
research will inevitably produce a correct and comprehensive
explanation of behavior. That is, our knowledge of a given behavior
will ultimately reduce down to the knowledge produced by a single
eld of research, such as MBG. Someone committed to integrative
pluralism would argue that each approach to the study of human
behavior provides a partial explanation of human behavior that, if
integrated, will give a comprehensive account of behaviordone that
is not reducible to the knowledge produced by any single approach.
An alternative to both views is ineliminable pluralism. This position
maintains that differing approaches each produce partial knowledge of the causes of human behavior. But because each approach
investigates different (sometimes mutually exclusive) causes, and
because different approaches often dene the same causes differently, it is unclear whether the study of behavior can be unied to
give a comprehensive explanation of human behavior.
Ineliminable pluralism is the idea that each approach to the
study of behavior is in fact investigating different phenomena and
different causes of those phenomena. Thus, ndings from human
behavioral studies in different elds of research cannot be integrated into a comprehensive view of human behavior. The latter
view is supported by at least some evidence (Longino, 2013). First,
different approaches to the study of human behavior: (i) fail to
explain a comprehensive amount of variation in behavior; (ii) use
different operational denitions of behavior; (iii) investigate a
different causal space; and (iv) tend not to read each others work to
inform their own. Thus, at present, a monist or integrative explanation of human behavior appears elusive. Longino (2013) summarizes her position on ineliminable pluralism as follows:
Epistemologically, the relation among the different research
approaches is neither competitive, reductionist, nor additive but
is best understood in a pluralist framework. Each approach offers partial knowledge that need not be congruent or
commensurable with knowledge produced in another approach,
even when they have overlapping subject matter. Ontologically,
there is no single concept of behavior. Furthermore, the operationalization of particular behaviors reects folk psychological
and moral values or political concerns as much as objective
properties of the phenomena. Socially, there is not enough
epistemologically productive interaction among these approaches, and the patterns of research uptake distort the
perception of what scientic investigation can reveal.
The point is that research into human behavior is incredibly complex. The problems are complex, the methods are complex, and
consequently, the interpretation of the research is cognitively
complex. Because of this complexity, it appears that some people
will always be able to justify inegalitarian beliefs by appealing to a
select body of scientic ndings about human behavior. That is,
people will always be able to nd some scientic evidence to
support a biologically or culturally deterministic explanation of
behavioral differences between races. But, if Longino (2013) is
correct, if there really is no comprehensive explanation of human
behavior because differing approaches to research are epistemically
and ontologically irreconcilable, then science cannot denitively
answer the question of whether or not racial inequality is immutable because of biological determinism. Each approach to scientic
research offers only a partial view of human behavior. Thus,
appealing to science to support inegalitarian political agendas
rooted in deterministic claims about the intransigence of
inequality is a partial reading of science that distorts the limits of
scientic knowledge.

Ineliminable pluralism does not mean that there are no valid


explanations of behavior, or that only some approaches produce
valid explanations of behavior. But it does eliminate the possibility
that basic research will illuminate general behavioral models of
humans (Longino, 2013). Consequently, it is hard to see how science
will ever produce a general behavioral model for Blacks or
Whites, or any other social group for that matter. And if so, then
essentialist beliefs about racial categories will never be validated by
basic scientic research into human behavioral variation. Furthermore, ineliminable pluralism implies that there are many truths to
human behavior and the task of society is to sort out which truth is
useful for which task in society (Longino, 2013). In some situations,
biological explanations of human behavior may be irrelevant to
social policy. Thus, we must be careful about which knowledge base
we use to evaluate the potential effectiveness of social policies
aiming to redress racial disparities.
Take, for example, QBG studies that report that academic ability
is heritable (see Shiao et al., 2012). In light of papers on the genetic
basis of racial differences in IQ (e.g. Jensen, 1969), these QBG
ndings have been used to claim that racial inequalities in education are not subject to change for genetic reasons. Consequently,
social policies aimed at reducing the racial achievement gap in
education are doomed to fail. In order to understand why QGB may
be irrelevant to social policies addressing the achievement gap one
must rst dene academic ability. Academic ability is frequently
dened in the literature as grades, or standardized test scores, or as
grade point average. But what does it take to earn good grades? To
earn good grades students must coordinate many different behaviors. Students need to get to school on time and they need to not
do things that cause them to be distracted from learning once in
school. Outside of school students need to engage in many behaviors to get their homework done. When the homework is done,
students need to engage in behaviors that lead them to eat well and
sleep so they can function cognitively and regulate any emotions
that might impair grade-earning behaviors. In addition there are
factors operating at higher levels of social organization that affect
grades, such as school funding and resources, which differ by the
racial composition of schools (Darling-Hammond, 2010).
If the biological essentialist wishes to use science to argue that
genes for academic ability prevent social interventions from
reducing academic inequalities between races then they would have
to show that genes are the difference makers when it comes to academic ability. To support the claim that genes are the difference
makers in academic ability, the biological essentialist would need a
body of MBG studies identifying the specic genes correlated with
each of the behaviors outlined above and they would need QBG
studies showing that the heritability of such behaviors is high and
that the behaviors differ by race. They would also need to have a
body of SE studies showing that: (i) none of the academically
important behaviors can be changed through social interventions;
(ii) the gap in academic ability between races does not respond to
any social intervention. Then they would need to explain away the
SE studies that show that social interventions can reduce the
achievement gap (e.g. Aronson, Fried, & Good, 2002; Walton &
Cohen, 2011). After ruling out these rival arguments, the biological
essentialist would need to identify a body of GEI ndings establishing that genes do not interact with the environment to produce
variation in academic ability. Then, they would need to explain away
studies showing that geneeenvironment interactions reduce heritability statistics for cognitive outcomes in economically impoverished populations (e.g. Turkheimer, Haley, Waldron, DOnofrio, &
Gottesman, 2003). After all is said and done they would still need
to show that the studies supporting the inegalitarian thesis dened
the behaviors contributing to grades in a commensurate way. Then
the biological essentialist would need to show that across these

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

studies a comprehensive amount of variation in grades is explained


by such genetically determined behaviors.
It is unlikely that the biological essentialist will ever be able to
accomplish this formidable task because there is, at present, no
general behavioral model for humans (Henrich et al., 2010). Instead,
he will be forced to acknowledge that there are limits to scientic
knowledge that make strong scientic claims about the immutability of racial inequality highly tenuous. Until a unied eld theory
of human behavior emerges in science it will be difcult to evaluate whether monism, integrative pluralism, or ineliminable
pluralism is the correct way of viewing human behavioral research.
And, until scientists explain how research across elds on any given
behavior maps across disciplinary grounds, ineliminable pluralism
appears to be the most defensible way of viewing human behavioral research. If so, then scientic research on human behavior
does not point us to any racial differences that are relevant for
policy purposes so the basic ethical assumption of racial equality
(present in BRR and SRR) should govern our attitudes toward policies that target social disparities between races (i.e. economic,
educational, and criminal justice disparities). Put differently, science provides no denitive reason to think that racial disparities
are immutable because of biology and therefore science provides us
with no racial reason to believe that disparity-closing policies are
doomed to fail for biological reasons.
5. Conclusions
In summary, the SMK for teaching about race in school biology
is: (i) psychological essentialism is associated with a misunderstanding of human genetic diversity and the belief that racial
inequality is immutable; (ii) biologists can apply different methods
to the same human genetic data and infer different things about the
reality of race; (iii) philosophers who disagree about the reality of
race can each be concerned with eliminating racial inequality; (iv)
scientic research on human behavior cannot be used to conclusively support the essentialist claim that racial inequality is
immutable for biological reasons.
If the preceding arguments are sound, then the educator who
understands this knowledge and who wishes to teach about race
for humanistic purposes faces a dilemma. Teachers will not be able
to claim that there are no biological differences between races
because of component two. Because of component three teachers
will not be able to claim that racial equality is threatened by the
view that race is biologically real. Both limitations might seem
difcult to swallow in light of the problem of biological essentialism outlined in component one. Indeed, if science teachers look
for a denitive scientic answer to the question of whether genes,
environment, or genes and environment, is the best explanation for
the racial disparities observable in society they will not nd a
conclusive answer because of component four. Upon rst glance
then, the prospect of teaching about race in school science for
humanistic purposes appears gloomy and many educators might
justiably choose not to engage in this task.
But, the gloominess pre-supposes that the possibility of racial
equality is tied to the authority of scientic knowledge. Why should
it be? Science educators need not appeal to the authority of science
in their teaching about race. Rather teachers can introduce students
to the concepts of SRR and BRR. Then, they can use this framework
to teach students about the deeper philosophical assumptions that
may have led Lewontin and Edwards to look at the same human
genetic data and conclude different things about the reality of race.
They can teach students how some modern scientists, especially
those in biomedicine (e.g. Risch, Burchard, Ziv, & Tang, 2002),
defend BRR in the face of the history of scientic racism, precisely
because they are committed to reducing racial inequality in the

health care system. Or they can teach students how SRR is used to
justify the use of racial categories in health-related research in order to reduce racial health disparities (Lorusso & Bacchini, in press).
This SMK will allow educators to teach students that racism is not
simply a product of science. Rather it is a product of how society
interprets science. And, if teachers understand ineliminable
pluralism, then they will be able to help students understand how
scientic research on human behavioral variation is often misinterpreted by society.
Additionally, this SMK can help teachers navigate the difcult
terrain of discussing race. In the context of a classroom discussion
on race, the knowledge in component three allows the teacher to
condently state that we can disagree about the nature of race but
still agree that racial inequality should be eliminated. This knowledge helps to establish the norm that, when we discuss race in our
science classroom, we can assume the best about those with whom
we disagree. We can assume that our peers are still concerned with
eliminating racial inequality even though we might not see to eye
with them about the reality of race. Establishing this norm in the
science pipeline is surely valuable in light of the often-heated debates about race in science (Foster, 2009).
The point is that when these four pieces of SMK are transformed
into pedagogical content knowledge they will allow science educators to design and teach curricula on race. Consequently, the
knowledge outlined here can help science educators accomplish
the race-specic goals laid out in the science education literature.
First, Zeidler et al. (2002) advocated teaching about the history of
cultural prejudice in scientic research on criminality and intelligence in order to challenge racial stereotypes. It is hard to imagine
how teachers can accomplish this goal without knowledge of
ineliminable pluralism (component four). Second, Castra et al.
(2013) and Puig and Jimnez-Aleixandre (2011) argue that school
science should challenge genetic determinism in order to undermine racism and ethnocentrism. An understanding of psychological
essentialism (component one), GEI (component four), and human
population genetics (component two) would equip teachers with
the knowledge for designing curricula to accomplish that goal.
Third, Donovan (2015) argued that race should be directly discussed as a topic of biology textbooks in order to increase understanding of human evolution and decrease racial prejudice. Such a
curriculum would be well complemented by a teacher with a strong
understanding of these four knowledge components. Finally,
Willinsky (1998) argued that it was irresponsible for the biology
curriculum to be colorblind in light of sciences contribution to
racial prejudice. Teaching students that scientic research on human behavior cannot conclusively establish the immutability of
racial inequality (component four) can help students understand
why 19th century racial science, which sought to establish the
naturalness of racial inequality, was so utterly awed. And, if students are exposed to the idea that there are multiple conceptions of
race concerned with the elimination of inequality (component
three) it could help students understand that they do not need to be
color-blind to value human equality.
In conclusion, the knowledge outlined here may not sufciently
address the goal of challenging racism through school science, but
it provides a necessary foundation for beginning this important
task. In essence, to put humanity back into the teaching of human
biology is to educate students about the limits of scientic
knowledgedthat science helps humans understand the world but
it need not lead us to compromise humanistic values.
Acknowledgments
This manuscript beneted from the feedback of two anonymous
reviewers and from scholarly discourse with Carlos Andres

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

10

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11

Barragan, Michael D. Edge, Joshua Glasgow, Michael Hunter, Jonathan Kaplan, Helen Longino, Roberta Millstein, Noah Rosenberg,
Quayshawn Spencer, and Rasmus Winther. The writing of this
manuscript was supported by a Stanford Interdisciplinary Graduate
Fellowship.
References
Appiah, K. A. (1996). Race, culture, identity: Misunderstood connections. Tanner
Lectures on Human Values, 17, 51e136.
Aronson, J., Fried, C. B., & Good, C. (2002). Reducing the effects of stereotype threat
on African American college students by shaping theories of intelligence.
Journal of Experimental Social Psychology, 38(2), 113e125.
Bastian, B., & Haslam, N. (2006). Psychological essentialism and stereotype
endorsement. Journal of Experimental Social Psychology, 42(2), 228e235.
Bergner, R. M. (2011). What is behavior? And so what? New Ideas in Psychology,
29(2), 147e155.
Block, N. (1995). How heritability misleads about race. Cognition, 56(2), 99e128.
Brueckner, H., Morning, A., & Nelson, A. (2005). The expression of biological concepts of race. In Annual meeting of the American sociological association. Philadelphia, PA.
Caspi, A., & Moftt, T. E. (2006). Geneeenvironment interactions in psychiatry:
Joining forces with neuroscience. Nature Reviews Neuroscience, 7(7), 583e590.
Castra, J., Sarapuu, T., & Clment, P. (2013). Comparison of French and Estonian
students conceptions in genetic determinism of human behaviours. Journal of
Biological Education, 47(1), 12e20.
Chao, M. M., Hong, Y., & Chiu, C. (2013). Essentializing race: Its implications on racial
categorization. Journal of Personality and Social Psychology, 104(4), 619.
Condit, C., Parrott, R., Bates, B., Bevan, J., & Achter, P. (2004). Exploration of the
impact of messages about genes and race on lay attitudes. Clinical Genetics, 66,
402e408.
Coop, G., Eisen, M. B., Nielsen, R., Przeworski, M., & Rosenberg, N. (2014, August 8).
Letters: A troublesome inheritance. New York Times. Retrieved from http://
www.nytimes.com/2014/08/10/books/review/letters-a-troublesomeinheritance.html.
Darling-Hammond, L. (2010). Structured for failure: Race, resources, and student
achievement. In Doing race: 21 essays for the 21st century (pp. 295e318). W. W.
Norton & Company.
Donovan, B. M. (2014). Playing with re? The impact of the hidden curriculum in
school genetics on essentialist conceptions of race. Journal of Research in Science
Teaching, 51(4), 462e496.
Donovan, B. M. (2015). Reclaiming race as a topic of the United States biology
textbook curriculum. Science Education (in press)
Edge, M. D., & Rosenberg, N. A. (2015). Implications of the apportionment of human
genetic diversity for the apportionment of human phenotypic diversity. Studies
in the History and Philosophy of Science Part C (in press)
Edwards, A. W. F. (2003). Human genetic diversity: Lewontins fallacy. BioEssays,
25(8), 798e801.
Ereshefsky, M. (2010). Whats wrong with the new biological essentialism. Philosophy of Science, 77(5), 674e685.
Evans, E. M., Spiegel, A. N., Gram, W., Frazier, B. N., Tare, M., Thompson, S., &
Diamond, J. (2010). A conceptual guide to natural history museum visitors
understanding of evolution. Journal of Research in Science Teaching, 47(3),
326e353.
Foster, M. W. (2009). Looking for race in all the wrong places: Analyzing the lack of
productivity in the ongoing debate about race and genetics. Human Genetics,
126(3), 355e362.
Gelman, S. A. (2004). Psychological essentialism in children. Trends in Cognitive
Sciences, 8(9), 404e409.
Glasgow, J., Shulman, J. L., & Covarrubias, E. G. (2009). The ordinary conception of
race in the United States and its relation to racial attitudes: a new approach.
Journal of Cognition and Culture, 9, 15e38.
Hacking, I. (2005). Why race still matters. Daedalus, 134(1), 102e116.
Hardimon, M. O. (2012). The idea of a scientic concept of race. Journal of Philosophical Research, 37, 249e282.
Haslam, N., Rothschild, L., & Ernst, D. (2000). Essentialist beliefs about social categories. British Journal of Social Psychology, 39(1), 113e127.
Haslam, N., Rothschild, L., & Ernst, D. (2002). Are essentialist beliefs associated with
prejudice? British Journal of Social Psychology, 41(1), 87e100.
Haslanger, S. (1995). Ontology and social construction. Philosophical Topics, 23(2),
95e125.
Haslanger, S. (2000). Gender and race: (what) are they? (What) do we want them to
be? Nous, 34(1), 31e55.
Haslanger, S. (2008). A social constructionist analysis of race. In B. A. Koenig, S. S.J. Lee, & S. S. Richardson (Eds.), Revisiting race in a genomic age (pp. 56e59).
Rutgers University Press.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?
Working Paper Series des Rates fr Sozialund Wirtschaftsdaten. No. 139.
Hirschfeld, L. A. (1998). Race in the making: Cognition, culture, and the childs construction of human kinds. The MIT Press.
Hirschfeld, L. A. (2012). Seven myths of race and the young child. Du Bois Review:
Social Science Research on Race, 9(01), 17e39.

Jayaratne, T. E., Ybarra, O., Sheldon, J. P., Brown, T. N., Feldbaum, M., Pfeffer, C. A.,
et al. (2006). White Americans genetic lay theories of race differences and
sexual orientation: Their relationship with prejudice toward Blacks, and Gay
Men and Lesbians. Group Processes & Intergroup Relations, 9(1), 77e94.
Jensen, A. R. (1969). How much can we boost IQ and scholastic achievement. Harvard Educational Review, 39(1), 1e123.
Kaplan, J. M., & Winther, R. G. (2013a). Prisoners of abstraction? The theory and
measure of genetic variation, and the very concept of Race. Biological Theory,
7(4), 401e412.
Kaplan, J. M., & Winther, R. G. (2013b). Realism, antirealism, and conventionalism
about Race. Retrieved from http://philpapers.org/rec/KAPRAA.
Keller, J. (2005). In genes we trust: The biological component of psychological
essentialism and its relationship to mechanisms of motivated social cognition.
Journal of Personality and Social Psychology, 88(4), 686e702.
Kinzler, K. D., & Dautel, J. B. (2012). Childrens essentialist reasoning about language
and race: Language and race. Developmental Science, 15(1), 131e138.
Levitis, D. A., Lidicker, W. Z., & Freund, G. (2009). Behavioural biologists do not agree
on what constitutes behaviour. Animal Behaviour, 78(1), 103e110.
Lewontin, R. C. (1972). The apportionment of human diversity. Evolutionary Biology,
6(381), r398.
Lieberman, L. (1997). Gender and the deconstruction of the race concept. American
Anthropologist, 99(3), 545e588.
Longino, H. E. (2013). Studying human behavior: How scientists investigate aggression
and sexuality. Chicago, IL: University of Chicago Press.
Lorusso, L., & Bacchini, F. (2015). A reconsideration of the role of self-identied races
in epidemiology and biomedical research. Studies in the History and Philosophy
of Science Part C (in press)
Mayr, E. (1982). The growth of biological thought: Diversity, evolution, and inheritance.
Cambridge, Mass: Belknap Press.
Medin, D. L., & Ortony, A. (1989). Psychological essentialism. In S. Vosniadou, &
A. Ortony (Eds.), Similarity and analogical reasoning (pp. 179e195). New York,
NY: Cambridge University Press.
Morning, A. (2008). Reconstructing race in science and society: Biology textbooks,
1952e2002. American Journal of Sociology, 114, 106e137.
Morning, A. J. (2011). The nature of race: How scientists think and teach about human
difference. London, England: University of California Press.
Neville, H. A., Lilly, R. L., Duran, G., Lee, R. M., & Browne, L. (2000). Construction and
initial validation of the color-blind racial attitudes scale (CoBRAS). Journal of
Counseling Psychology, 47(1), 59e70.
Opfer, J. E., Nehm, R. H., & Ha, M. (2012). Cognitive foundations for science
assessment design: knowing what students know about evolution. Journal of
Research in Science Teaching, 49(6), 744e777.
Pauker, K., Ambady, N., & Apfelbaum, E. P. (2010). Race salience and essentialist
thinking in racial stereotype development. Child Development, 81(6), 1799e
1813.
Phelan, J. C., Link, B. G., & Feldman, N. M. (2013). The genomic revolution and beliefs
about essential racial differences a backdoor to eugenics? American Sociological
Review, 78(2), 167e191.
Prentice, D. A., & Miller, D. T. (2007). Psychological essentialism of human categories. Current Directions in Psychological Science, 16(4), 202e206.
Puig, B., & Jimnez-Aleixandre, M. P. (2011). Different music to the same score:
Teaching about genes, environment, and human performances. In T. D. Sadler
(Ed.), Socio-scientic issues in the classroom (Vol. 39, pp. 201e238). Dordrecht:
Springer Netherlands.
Rangel, U., & Keller, J. (2011). Essentialism goes social: Belief in social determinism
as a component of psychological essentialism. Journal of Personality and Social
Psychology, 100(6), 1056e1078.
Rhodes, M., Leslie, S.-J., & Tworek, C. M. (2012). Cultural transmission of social
essentialism. Proceedings of the National Academy of Sciences, 109(34), 13526e
13531.
Risch, N., Burchard, E., Ziv, E., & Tang, H. (2002). Categorization of humans in
biomedical research: Genes, race and disease. Genome Biology, 3(7), 1e12.
Rosenberg, N. A. (2011). A population-genetic perspective on the similarities and
differences among worldwide human populations. Human Biology, 83(6), 659e
684.
Rosenberg, N. A., Pritchard, J. K., Weber, J. L., Cann, H. M., Kidd, K. K.,
Zhivotovsky, L. A., et al. (2002). Genetic structure of human populations. Science,
298(5602), 2381e2385.
Sesardic, N. (2010). Race: a social destruction of a biological concept. Biology and
Philosophy, 25(2), 143e162.
Shiao, J. L., Bode, T., Beyer, A., & Selvig, D. (2012). The genomic challenge to the social
construction of race. Sociological Theory, 30(2), 67e88.
Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching.
Educational Researcher, 15(2), 4e14.
Shtulman, A., & Schulz, L. (2008). The relation between essentialist beliefs and
evolutionary reasoning. Cognitive Science: A Multidisciplinary Journal, 32(6),
1049e1062.
Spencer, Q. (2012). What biological racial realism should mean. Philosophical
Studies, 159(2), 181e204.
Sober, E. (2001). Separating nature and nurture. In D. Wasserman, & R. Wachbroit
(Eds.), Genetics and criminal behavior (pp. 47e78). Cambridge, UK: Cambridge
University Press.
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology,
69(5), 797e811.

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011

B.M. Donovan / Studies in History and Philosophy of Biological and Biomedical Sciences xxx (2015) 1e11
Turkheimer, E., Haley, A., Waldron, M., DOnofrio, B., & Gottesman, I. I. (2003).
Socioeconomic status modies heritability of IQ in young children. Psychological
Science, 14(6), 623e628.
Wade, N. (2014). A troublesome inheritance: Genes, race and human history.
Melboune, Australia: Penguin Press.
Walton, G. M., & Cohen, G. L. (2011). A brief social-belonging intervention improves
academic and health outcomes of minority students. Science, 331(6023),1447e1451.
Williams, M. J., & Eberhardt, J. L. (2008). Biological conceptions of race and the
motivation to cross racial boundaries. Journal of Personality and Social Psychology, 94(6), 1033e1047.
Willinsky, J. (1998). Learning to divide the world: Education at empires end.
Minneapolis, MN: University Of Minnesota Press.

11

Winther, R. G. (2014). The genetic reication of Race?: A story of two mathematical methods. Critical Philosophy of Race, 2(2), 204e223.
Winther, R. G., Giordano, R., Edge, M. R., & Nielsen, R. (2015). The mind, the lab, and
the eld: Three kinds of populations in scientic practice. Studies in the History
and Philosophy of Science Part C (in press)
Witherspoon, D. J., Wooding, S., Rogers, A. R., Marchani, E. E., Watkins, W. S.,
Batzer, M. A., et al. (2007). Genetic similarities within and between human
populations. Genetics, 176(1), 351e359.
Zeidler, D. L., Sadler, T. D., Berson, D. M. J., & Fogelman, A. L. (2002). Bad science and
its social implications. In The educational forum (Vol. 66, pp. 134e146).

Please cite this article in press as: Donovan, B. M., Putting humanity back into the teaching of human biology, Studies in History and Philosophy
of Biological and Biomedical Sciences (2015), http://dx.doi.org/10.1016/j.shpsc.2015.01.011