The Problems with Genetic Essentialism, Determinism, and Reductionism

The concepts of genetic essentialism, genetic determinism, and genetic reductionism (hereafter referred to as the GEDR concepts) have previously not been clearly distinguished in the scientific literature. These concepts refer to different ideas, and failing to distinguish among them can contribute to “semantic muddiness … and to rancour and miscommunication in vital scientific and ethical debates” (Harden, 2023, p. 203). Moreover, it is possible in theory that while one or two of these ideas could be useful to scientists, policymakers, and the public, the other(s) might have detrimental implications. In that case, it would be important to not embrace all three ideas as a single unit that cannot be disentangled.

In addition to the poorly specified relations between the GEDR concepts, the concepts themselves represent simplified or even misleading understandings in which genes are considered more important in organism development than warranted by the scientific evidence. Therefore, these concepts can form the basis of misconceptions, resulting in two main problems. First, the literatures on genetic essentialism (e.g., Donovan, 2022; Donovan et al., 2024; Stern & Kampourakis, 2017) and genetic determinism (Castéra & Clément, 2014; Gericke et al., 2017) suggest that these concepts have fostered discrimination among humans based on genetic arguments. Examples include sexism (e.g., the claim that men are overrepresented in STEM fields because they are smarter in these domains than women; see Geary, 1998; Kimura, 1999) and racism (e.g., the claim that IQ scores show people of different races to be differentially intelligent; see Herrnstein & Murray, 1994). Second, these concepts relate to the frequently misunderstood relationship between genes and phenotypes (Lynch et al., 2019), which has been recognized as the most central aspect of genetics (Portin, 2009) and genetics education (Duncan & Reiser, 2007; Gericke & Smith, 2014; Jamieson & Radick, 2017; Kampourakis, 2017). An accurate account of this relationship should be based on a model in which “nature” and “nurture” are understood to be inseparable contributors to phenotypic outcomes, contributors that interpenetrate one another at every level of analysis (Lickliter, 2009) and that are also under the influence of stochastic events (Lewontin, 2011). Contemporary scientific models show that how genes function in biological contexts is more complex than initially expected by, for example, the proponents of the Human Genome Project (Kampourakis, 2021). To mitigate these two negative consequences while accurately communicating information about genetics to the public and in education, we need to understand the GEDR concepts, to differentiate among them, and to recognize when they are being misused.

We define these concepts as follows (based on Kampourakis, 2017, p. 6, and Kampourakis, 2021, pp. xvii-xviii):

• Genetic essentialism: Genes are fixed entities that constitute the essence of what we are. They achieve this by specifying phenotypes from which their existence can be inferred.

• Genetic determinism: Genes invariably determine phenotypes, so that the outcomes are just a little, or not at all, affected by the changing environments in which individuals live.

• Genetic reductionism: Genes provide the ultimate explanation for phenotypes, and so the best approach to explain these is by studying phenomena at the level of genes.

Thus, these concepts refer, respectively, to three important properties usually attributed to genes: (1) that they are fixed essences that specify who we are; (2) that they alone determine characteristics notwithstanding individuals’ developmental environments; and (3) that they best explain the presence of characteristics.

Each GEDR concept is important, although for different reasons. As noted above, genetic essentialism can underpin values used to discriminate among people based on genetic arguments, for instance, by erroneously considering specific groups (e.g., a racial or ethnic group) as genetically homogeneous and genetically distinct from others (Donovan, 2022; Roth et al., 2023). In contrast, genetic determinism can leave people feeling as if their choices, behaviors, or environments do not have an impact on their life outcomes (Willoughby et al., 2019). Further, genetic reductionism can interfere with scientists’ and policymakers’ abilities to understand how interventions at higher levels of organization (i.e., socioeconomic, educational, etc.) can potentially be just as effective as interventions at lower levels (i.e., biological). In the case of the Human Genome Project, for example (Gannett, 2022; Vicedo, 1992), the ideological commitments of its proponents resulted in a focus on molecular factors only, rather than on the complex systems responsible for normal and pathological phenotypes (Gottlieb, 2002; Moore & Lickliter, 2023; Oyama et al., 2001).

Harden (2023) recently attempted to clarify the meaning of and relationships between the GEDR concepts. We appreciate Harden providing some clarification in her paper; it has drawn attention to these concepts and the distinctions among them. Nonetheless, there are relevant, crucial facts about these concepts that were not addressed in Harden’s paper. By addressing these facts here, we hope to provide a better understanding of pitfalls associated with the GEDR concepts.

There is value in attempting to define and distinguish among the GEDR concepts, but it can be misleading to clarify and distinguish ideas that do not reflect veridical reality; clarifications and distinctions of such ideas are best accompanied by clear statements conveying those ideas’ shortcomings. In our view, Harden’s (2023) paper includes statements that misrepresent (1) how genetic factors contribute to trait determination and (2) the value of strongly reductionistic approaches to biology. Consequently, her paper might inadvertently generate further confusion about the relationship between genotypes and phenotypes. The primary goal for the present paper is to correct persistent misunderstandings about the role of DNA segments in phenotype development. In this way, we hope to contribute positively to Harden’s effort to create a common ground for using the GEDR concepts in the genetics and education literatures.

We are comfortable with Harden’s definitions of the GEDR concepts. She defined “genetic essentialism” as “a particular type of essentialist thinking, where the essence that constitutes what a thing or person ‘really’ is … is some real or imagined DNA sequence” (Harden, 2023, p. 200). She defined “genetic determinism” by writing that “a phenotype is governed by genetic determinism if, and only if, given a specified genotype, the way the phenotype develops thereafter is fixed as a matter of natural law” (Harden, 2023, p. 198). Finally, Harden defined genetic reductionism as “the claim that higher-level phenomena [such as psychotic experiences or depression] can, in theory, be entirely explained by knowledge about lower-level parts and processes [such as genes and genetic processes], and that, accordingly, phenomena are more fruitfully investigated at those lower levels” (Harden, 2023, p. 201). Despite our alignment with Harden on these definitions, we believe her paper might leave readers with significant misunderstandings. In what follows, we address these misunderstandings in turn.

Harden’s discussion of genetic essentialism is the least problematic of the three major sections of her paper. She identifies determinism as “a philosophical term about the causal structure of the universe” (Harden, 2023, p. 198) and (quoting Rosenberg, 2001) she defines reductionism as “a metaphysical thesis” as well (Harden, 2023, p. 202). A similarly metaphysicaldefinition of essentialismwould entail that biological entities are characterized by essential features that make those entities what they are. Such metaphysical essentialism is an ancient idea rooted in Plato’s idealism, yet it remains controversial among philosophers (Hull, 2006). Regardless, rather than discussing metaphysical essentialism, Harden departs from her approach to determinism and reductionism by instead focusing on psychological essentialism, which – as Medin noted when distinguishing metaphysical and psychological essentialism – “refers not to how the world is but rather to how people approach the world” (Medin, 1989, p. 1477).

Harden’s concern with psychological essentialism is evidenced in her statements that essentialism is a “belief that things have essences” (Harden, 2023, p. 199, italics added) and that “essentialist thinking is a cognitive bias that emerges very early in development” (Harden, 2023, p. 200, italics added), even appearing in preschoolers (Gelman, 2005). Harden’s essay accurately represents the findings from the psychological literature and makes no claims about genetic factors actually constituting a person “essence.” The fact is, as Medin noted, “psychological essentialism is bad metaphysics” (Medin, 1989, p. 1477); just because people believe something about how the world works does not mean the world works that way! After all, before the Copernican Revolution, most people erroneously believed the earth was at the center of the universe.

Of course, some people erroneously believe that genes are the source of a person’s “essence” and that genetically influenced traits are prescribed, unchangeable, and fundamentally responsible for what the person thinks or does (Dar-Nimrod & Heine, 2011; Heine, et al., 2019). It is important to acknowledge this error because as Harden makes clear, numerous studies (with some exceptions) have found forms of essentialist thinking about gender, sexual orientation, or race, for example, to be related to prejudice, stereotyping, and tolerance of discriminatory policies (Mandalaywala, 2020).

We believe any form of genetic essentialism is easily defeated, as people’s DNA alone cannot be what makes them “essentially” who they are; it is uncontroversial to acknowledge that experiences play critical roles in who we become. Moreover, we think everyone would agree that people lacking some normal human genes are nonetheless still “essentially” human; for example, we doubt anyone would argue that people with Williams syndrome, who lack genes normally present on the seventh human chromosome (Martens et al., 2008), are somehow lacking a human essence!

In contrast to Harden’s treatment of genetic essentialism, her treatment of genetic determinism focuses not on a cognitive bias, but on a metaphysical claim about genetic causation of phenotypes. As a result, empirical observations can more helpfully be brought to bear on this question. Because the idea that genes can determine phenotypes is a claim about how things actually work in the world, it is important to ensure that anything implied in a published text is consistent with those empirical observations.

Genes (i.e., segments of DNA used to produce functional RNA molecules) clearly have a causal impact on phenotype development. Even so, the nature of that impact and how it occurs is widely misunderstood. A common misconception, reinforced by media reports and textbooks, holds that genes can be the sole causes of phenotypes (Carver et al., 2017; Gericke et al., 2014). This is an inappropriately simplified conception of the role of gene expression during development, one that is inconsistent with what molecular biologists currently understand. In fact, virtually all theorists who have written about how genotypes contribute to phenotypes have concluded that phenotypes should not be considered immutable and that dichotomous conceptions of genetic and environmental factors are ill-considered; these theorists include developmental biologists (Gilbert, 2002; Gilbert & Epel, 2015), evolutionary biologists (Jablonka & Lamb, 2005), physiologists (Noble, 2017), geneticists and molecular biologists (Lewontin, 2000b; Strohman, 2003), philosophers of biology (Griffiths & Gray, 2001; Robert, 2006; Stotz, 2006) and of psychology (Moore, 2008), and behavioral neuroscientists (Lickliter & Honeycutt, 2003; Moore, 2015). Many scientists trained as behavior geneticists, as Harden was, also acknowledge that genes cannot determine complex human traits (Turkheimer, 2011).

As detailed in many prior publications (Kampourakis, 2021; Moore & Shenk, 2017; Tabery, 2014), one reason some people still think genes can cause phenotypes is because genes are, in some cases, “difference-makers,” factors that make a difference in phenotypic outcomes, all else being equal. However, studies that show such effects do so by controlling non-genetic factors in order to reveal the difference-making effects of genes. In so doing, any potential effects of these non-genetic factors in natural contexts are hidden. This experimental approach can create the illusion that genes are determinative even when the phenotypes being studied actually develop because of how non-genetic factors co-act with the genes. An analogy might be helpful here: the invariably freezing temperatures at the South Pole could ensure that the only difference maker able to account for snowfall there is the relative humidity in the atmosphere on a given day. But just because temperature does not make any difference in this scenario does not mean freezing temperatures are unimportant for producing snow (Moore, 2001; for another analogy, see Hebb, 1970). Likewise, factors beyond DNA can play important roles in a phenotype’s development even if controlled studies make it look like genes alone determine the phenotype.

Harden defines genetic determinism as entailing phenotype development that is “fixed as a matter of natural law” by an organism’s genotype. Nevertheless, what makes genetic determinism wrong is that phenotype development is never “fixed” by an organism’s genotype; only by specifying the organism’s environment, too, would a scientist have any hope of predicting the organism’s phenotype (Lewontin, 2000a; Oyama, 1985). In fact, because an enormous number of factors in natural developmental environments cannot be guaranteed, phenotype development cannot be determined by a genotype. Gottlieb et al. (1970) captured this important idea when he described development as a process involving probabilistic epigenesis.

However, Harden continues by remarking that “having five fingers on each hand is genetically determined” (Harden, 2023, p. 198), explicitly miscommunicating to readers that some phenotypes are genetically determined, even though geneticists have understood for over a century that genes cannot determine phenotypes (Sturtevant, 1915). She explains her statement by noting that “if one has a specified genotype, then one will (almost) invariably develop five fingers on each hand as a matter of natural law” (Harden, 2023, p. 198). Notice here how Harden included the word “almost” parenthetically. On one level, this is an acknowledgment that this phenotype is not determined by genes alone. Yet by relegating the word “almost” to parentheses, Harden suggests that for all intents and purposes, we can safely ignore non-genotypic contributions to hand development. The truth, though, is that the characteristics of hands, like all phenotypes, are not genetically determined. They instead emerge from the complex coactions during development of both genetic and nongenetic factors (Lerner & Overton, 2017), including signaling molecules and other proteins such as sonic hedgehog and fibroblast growth factors (Raszewski & Singh, 2023), the functions of which always depend on yet other nongenetic factors such as ambient temperatures (Johnston, 1987) and the presence of various protein-folding chaperone molecules (Scriver & Waters, 1999; Zanna & West, 2014).

In the next paragraph, Harden correctly notes that educational attainment cannot be considered genetically determined because “people who have the same genotype can have numerous different educational outcomes” (Harden, 2023, p. 198). However, since phenotypes are never genetically determined, we should expect to find individuals with the same genotype varying on any phenotype. In fact, individuals with the same genotype can be born with different numbers of fingers (i.e., polydactyly), too, because of variations in developmental events occurring in utero (Hwang et al., 2016; Peterson & Rayan, 2004). The always-present influence of non-genomic factors on development means that individuals with the same genotype can have different phenotypic outcomesregardless of the phenotype under consideration. Even a Mendelian disorder like Huntington disease, which Harden explicitly calls a “genetically determined” disorder (Harden, 2023, p. 198), is more complex than is usually recognized. This disease is not caused in a straightforward way by any single genetic variant (Moore, 2013), and monozygotic (MZ) twins with identical DNA in the relevant chromosomal regions have been observed to develop different behavioral abilities and symptoms (Anca et al., 2004; Georgiou et al., 1999). Furthermore, MZ twins have phenotypes that diverge across their lifespans, as their differing experiences lead to the epigenetic up- and down-regulating of different portions of their genomes (Fraga et al., 2005). Finally, note that non-genomic variables that significantly influence phenotypes are not limited to dramatic events like thalidomide exposure (to use Harden’s example, discussed further below); instead, phenotypic outcomes can also reflect what Waddington (1957) – and Lewontin (1983), after him – called “developmental noise” (Waddington, 1957, p. 39).

Harden concludes her initial section on genetic determinism by revisiting the parenthetical “almost” in her earlier sentence. She notes that this “qualifier is necessary, because there are environmental exposures that might alter morphological development beyond one’s genotype, for example, exposure to thalidomide in utero” (Harden, 2023, p. 198). For reasons that mirror our primary concerns about genetic determinism, Harden backtracks here on the claim that finger number is genetically determined. But rather than revising the earlier appearing text to indicate that there are no such things as genetically determined phenotypes, Harden doubles down on the notion of genetic determinism by appealing to “our intuition that the number of fingers is genetically determined” (Harden, 2023, p. 198) … as if our intuition that the sun revolves around the earth – a reasonable hunch, given our everyday experience! – should alter the scientific conclusion that the earth in fact revolves around the sun.

This is not quibbling over words. The fact that genotypes cannot determine phenotypes is important for several reasons. Genetic determinist arguments were used in the 20th century to justify the sterilization of over 60,000 Americans considered socially “undesirable” (Kevles, 1995; National Human Genome Research Institute, n.d.) and to justify the extermination of millions of people in Nazi Germany (Lerner, 1992; Lifton, 1986; Müller-Hill, 1998; Proctor, 1988). Such arguments are still deployed to justify some hate crimes today (Harden, 2023; Sternberg, 2020, 2024). Genetic determinism clearly has social consequences. In addition, this way of thinking can leave us believing that a person’s “genetically determined” pathologies can only be treated with genomic alterations. However, this is not the case. For example, although phenylketonuria (PKU) is a disorder often associated with a genetic abnormality (Al Hafid & Christodoulou, 2015), once the mechanism underlying the development of PKU was understood to involve specific genetic and environmental factors, a dietary treatment became available (Scriver & Waters, 1999). This reflects the fact that even Mendelian disorders like PKU can be “multifactorial and complex” (Scriver & Waters, 1999, p. 271). And consistent with the statements about Huntington disease and polydactyly above, even siblings who share identical abnormal PKU-related genes sometimes “have greatly different cognitive and metabolic phenotypes” (Scriver & Waters, 1999, p. 268).

Failing to acknowledge that nongenetic factors always play important roles in phenotype development is a mistake, insofar as it can blind us to interventions that could improve lives. Because of human ingenuity, we need not accept our environments as we find them; instead, we can purposefully vary developmental circumstances that otherwise do not normally vary, and in so doing, alter phenotypes in helpful ways. This was Dobzhansky’s point in 1955 when he wrote “the existing variety of environments is immense, and new environments are constantly produced. Invention of a new drug, a new diet, a new type of housing, a new educational system, a new political regime introduces new environments” (Dobzhansky, 1955, p. 75). Once we understand the complex developmental processes that cause a pathological phenotype – as opposed to merely knowing what DNA segments are correlated with that phenotype – we will always have multiple ways to intervene in those processes and thereby forestall the development of pathology.

Although Harden clarified what genetic determinism is and how it differs from genetic essentialism and reductionism, she left intact the idea that some traits are genetically determined. This mistaken perspective might be a consequence of the methods favored by behavior geneticists, who ordinarily rely on statistical analyses of correlational data. But the experimental approaches utilized by molecular and developmental biologists have clarified that genes cannot single-handedly determine phenotypes. Consequently, Harden’s essay risks perpetuating misunderstandings, sowing confusion about how phenotypes are related to genotypes. The idea that genes can determine phenotypes is an important error that requires correcting.

Finally, there is the idea that phenomena at the level of an organism (i.e., traits) can be fully explained with knowledge only about phenomena at “lower” levels of organization (e.g., DNA segments). This way of thinking, called reductionism, holds that there is a causal chain that begins at “lower” levels of biological organization and proceeds through “higher” levels – the levels of cells, tissues, organs, etc. – to produce a phenotype at the level of the whole organism. Different forms of “reductionism” have engendered varying degrees of consensus among philosophers of biology (Griffiths & Stotz, 2013). One of these, methodological reductionism, holds that the most fruitful strategy for studying a biological system is to begin by decomposing it into its component parts. Another form, metaphysical reductionism, entails a commitment to the idea that an organism is merely the sum of its physical parts (Griffiths & Stotz, 2013). Consistent with this latter idea, scientists broadly agree that “people, and their minds and their behaviour, ultimately consist of physical states and processes; there is no extra-material or spiritual realm” (Harden, 2023, p. 202). Nonetheless, by the 1980s, philosophers of biology had nearly arrived at an anti-reductionist consensus, based on the rejection of stronger forms of reductionism that posited no role for higher level factors in the operation of lower level factors (Griffiths & Stotz, 2013).

Methodological reductionism has proven useful in research. For example, the Human Genome Project arguably improved understanding of the complexities of development, including the facts that genes are typically pleiotropic and that diseases are typically polygenic and multifactorial. In contrast, strong forms of metaphysical reductionism – forms that insist that some genes cause phenotypes independently of the hierarchical systems in which they are embedded – remain problematic, given how biological systems work (Noble, 2006; Noble & Noble, 2023). Understanding how these systems work requires an integrative approach and a focus on the process of development (Lickliter & Honeycutt, 2003; Lickliter & Moore, 2023; Moore & Lickliter, 2023).

Several forms of genetic reductionism allow that complex systems have properties that emerge from interactions between system components, properties that do not characterize the components themselves (Jones, 2000). These properties often emerge only when a system operates at a sufficiently large scale. An example from chemistry is illustrative. Liquid water is composed of H₂O, but a collection of H₂O molecules that numbers fewer than 6 does not behave as a liquid, because only when a sixth molecule is present does the compound settle into the kind of three-dimensional structure responsible for the properties of water (Liu et al., 1996), such as being able to wet a solid surface (Coghlan, 1997). Thus, small numbers of H₂O molecules do not have the properties we associate with liquid water; these properties only emerge at larger scales.

Human beings likewise have properties that do not characterize the molecules that constitute our bodies, so these properties are emergent; in this sense, people are obviously more than the sum of our parts. However, if the characteristics of lower-level components like molecules permit complete explanations of higher level phenomena like social behavior, then the presence of emergent, social phenomena need not weigh against strong reductionist approaches. This is one reason why most students of human behavior do not hold extreme anti-reductionist positions (Harden, 2023). We agree with Harden that “a complex phenomenon, such as human behavior, can be understood from multiple, overlapping perspectives, and [that] scientific studies that differ in their level of analysis (ranging from the actions of molecules within cells to the actions of governments within nations) can provide complementary information” (Harden, 2023, p. 202). Even so, there remains a problematic aspect of strong reductionism that is insufficiently highlighted in Harden’s article, an aspect that risks leaving readers with a more reductionist orientation than is warranted when considering human behavior.

Just as reductionist approaches can explain emergent phenomena like the properties of water, such approaches could, in theory, explain emergent human behavior; this is why studies of genetics can enhance understanding of psychological processes. But in an attempt to embrace “the middle ground of explanatory pluralism” (Harden, 2023, p. 202), Harden also writes that “what makes something an ‘ability’ is measured at the level of behavior and defined by social conventions” (Harden, 2023, p. 203, italics added). Now, social conventions do establish understandings of “ability” and such things are best measured at the level of behavior. Nevertheless, Harden’s approach leaves open the possibility that regardless of how a society defines “ability,” the behaviors that contribute to an “ability” can be completely explained by referring to genes. In this way, Harden’s “middle ground” fails to reject strong reductionism.

Strong reductionism warrants rejection because information flows bidirectionally in complex biological systems. Reductionism can tolerate emergent phenomena per se but when higher level processes feedback in ways that influence lower level processes, then a systems approach becomes necessary (Noble & Noble, 2023). It is not enough to acknowledge that emergent, higher level phenomena are best studied at that higher level; instead, we must also recognize that events at higher system levels influence the functioning of components at lower system levels. Harden acknowledges the existence of higher level phenomena that can fruitfully be studied at both higher and lower levels of analysis, and she suggests that knowledge of genetics can help explain these phenomena, presumably because events at lower levels affect events at higher levels. What is missing from her portrayal is the important ways in which events at higher levels affect events at lower levels.

In a schematic image published over 3 decades ago, Gilbert Gottlieb (1991) quickly conveyed the basic principle that complex biological systems are characterized by a bidirectional flow of information (see Fig. 1). In this representation of the developmental systems perspective, Gottlieb captured the important roles that higher level system components and processes play in phenotype development. As his figure indicates, genetic activity is influenced by system components that are present – and events that are taking place – at higher levels of organization. Again, phenotypes are never “fixed” by genotypes, and Gottlieb’s drawing clarifies why this is so. The way DNA is used during development depends on an organism’s experiences, specifically on how those experiences alter the contexts in which DNA is employed by cellular processes.

Studies of experiential effects on the epigenetic regulation of DNA have produced data consistent with Gottlieb’s representation (Moore, 2015, 2017). Factors such as diet (Kucharski et al., 2008; Waterland & Jirtle, 2003), maternal behavior (Provençal et al., 2012; Weaver et al., 2004), social interactions (Cole, 2009; Tung et al., 2012), and drug use (Maze & Nestler, 2011) alter the functioning of genomes in many species, thereby influencing phenotypic outcomes in ways that cannot be predicted from DNA sequences alone (Moore, 2015). For example, immune system functioning is influenced by nonhuman primates’ dominance rank in their social groups (Tung et al., 2012) and by people’s subjective perceptions of how lonely they are (Cole, 2009). These chronically stressful experiences can compromise individuals’ immune systems, leaving them with pathological phenotypes (e.g., cardiovascular diseases associated with inflammation, or [at least in macaque monkeys] reproductive dysregulation). Similarly, being abused early in development increases stress reactivity and epigenetically downregulates BDNF (brain-derived neurotrophic factor) gene expression in rodents (Roth et al., 2009; Weaver et al., 2004). Early abuse has likewise been associated with the epigenetic downregulation of glucocorticoid receptor expression in the hippocampi of human suicide victims (McGowan et al., 2009). Such data undermine genetic determinist arguments, but they also reveal weaknesses in genetic reductionism. Because higher level phenomena like social interactions influence what the genome produces, a reductionist approach that tries to explain higher level phenomena in terms only of lower level system components is bound to fail.

Reductionism can blind us to another issue because it can draw attention to genetic structures at the expense of biological processes. To Harden’s credit, her first paragraph on genetic reductionism mentions “processes” (Harden, 2023, p. 202) repeatedly. She recognizes that it cannot be the structure of DNA molecules alone that influences phenotypes; events involving those molecules must also be considered. These processes invariably involve molecular interactions that reflect the spatial and temporal relations that characterize the molecules involved. Since at least the 1970s, many biologists, psychologists, and philosophers have cogently argued that a relational, process-oriented approach to the study of development is the best way to escape the fruitless nature-versus-nurture debate (Allen & Bickhard, 2013; Gottlieb, 1992; Lehrman, 1970; Lerner, 1978; Lickliter & Moore, 2023; Overton, 2015; Oyama, 1985; Spencer et al., 2009; Witherington & Lickliter, 2017), and that this approach can also neutralize a potential pitfall of genetic reductionism, namely, the tendency to think it is the genome’s structure that is of paramount importance in biological outcomes.

Instead, a focus on what happens during development is more likely to yield actionable insights about phenotype development. Such a process perspective (Lickliter & Moore, 2023; Moore, 2009; Moore & Lickliter, 2023) can provide insights about the functions of a complex system’s component parts, insights that would remain hidden to scientists focused on the parts themselves. In discussing emergent phenomena that arise from complex systems, Bar-Yam (2011) notes that a door key’s structure alone cannot be informative about whether the key will open a lock because the key’s functionality depends on the relation between the key’s structure and the structure of the lock. This perspective suggests that scientists ought not to conclude that structural aspects of DNA per se are what matter in phenotype development. Instead, we should recognize that what matters are the relations between these factors and the nongenetic factors with which they coact during development (Lerner & Overton, 2017). This approach would help avoid a problematic aspect of reductionism.

Behavioral geneticists need to be careful about how they discuss the results of their studies, so as to not promote the mistaken impression that the genotype-phenotype correlations they discover reflect genetic causes of phenotypes. Such explanatory reduction would be inappropriate, given how nongenetic factors regulate the activity of DNA. In addition, like Harden, we remain concerned that genetic reductionism can steer financial and human resources toward studies of genetics, when relational, process-oriented studies that also consider higher level factors could be just as beneficial and informative about the origins of human behavioral phenotypes. For similar reasons, it behooves researchers creating genetics education and communication tools to be wary of conveying outdated ideas about genetics to the students reading their work.

Research in molecular biology has consistently shown that genetic determinism does not accurately represent how phenotypes develop, and because essentialism, determinism, and reductionism are related ideas, wariness of all three is justified. Genomic sequence information is never the sole cause of a phenotype, so its influences should be considered in DNA’s broader cellular, organismal, and environmental contexts. DNA segments can only be expressed with the aid of numerous proteins and nucleic acids, and many of the products expressed when DNA segments are activated regulate or affect the expression of other DNA segments in turn; therefore, there is no reason to conceive of DNA segments as having more control over developmental outcomes than these other products of gene expression (see Griffiths & Gray, 1994, for a more inclusive version of this argument). Biological characteristics emerge from developmental processes that occur at various levels of organization and that influence one another. This is why phenotypes are not genetically determined but are rather the outcome of the coactions of multifarious genomic and non-genomic factors. Although genes are implicated in development, they do not cause or produce anything on their own (Kampourakis, 2021; Moore, 2001).

There are good reasons for carefully considering the meanings of and relationships between the GEDR concepts, because they are different (if related) ideas, despite their superficial resemblances. In addition, Harden is correct that “human genetics is contested science” (Harden, 2023, p. 203), so we agree that it is important for researchers to be clear about the probabilistic nature of the relationship between genotypes and phenotypes. Nevertheless, while Harden (2023) helpfully drew attention to the distinctions between the GEDR concepts, deeper problems inherent in these ideas remained unexplored in her text. When a paper published in a prestigious journal is focused on genetic essentialism, determinism, and reductionism, that paper should carefully consider and explicitly acknowledge what molecular biologists know about the veracity of the ideas under discussion.

Specifically, it is of the utmost importance to convey that genetic essentialism fails because individuals are as they are in part because of the contexts in which they develop, that genetic determinism fails because all phenotypes depend on both genomic and non-genomic factors for their development, and that genetic reductionism fails because emergent properties above the level of the genome can feed back and influence the subsequent functioning of that genome. It is especially important to convey these understandings because ignorance of the central role of development in phenotype origins – and of the nature of the relationship between genotypes and phenotypes – remains widespread. Scientific discourse will be improved by work that provides semantic clarity about the GEDR concepts. Nonetheless, elucidating the meaning of these concepts without providing strong arguments for rejecting them – and, in some cases, perhaps inadvertently supporting these misleading concepts – could be likened to the proverbial effort to rearrange the deck chairs on the Titanic.

The authors would like to thank Dawn Michele Moore, PhD, for her assistance in formatting and submitting this article.

This paper did not require work with human or animal subjects, cell lines, or any identifiable information, materials or data.

The authors declare no conflicts of interest.

None of the authors received any funding in connection with this article.

K.K., N.G., and D.S.M. conceptualized the article and reviewed and edited all drafts. K.K. and D.S.M. each wrote an original draft.

Human development considers “data” to include “the literature consulted” in the development of a paper. All data considered in this article are included herein. Further inquiries can be directed to the corresponding author.