Skip to main content
SearchLoginLogin or Signup

Irreducibly Social: Why biocriminology’s ontoepistemology is incompatible with the social reality of crime

Published onMar 28, 2022
Irreducibly Social: Why biocriminology’s ontoepistemology is incompatible with the social reality of crime


Professing interactionist bio + social terminology, contemporary biocriminology asserts a break from its biologically essentialist past. Assurances notwithstanding, whether biocriminology has undergone a decisive paradigm shift rejecting notions of biological criminals and bad brains remains uncertain. Unfortunately, discussions of biocriminology’s assumptions are mired in politics, obscuring important scientific issues. Motivated to clarify misunderstanding, I address the ontoepistemology of biocriminology from a scientific realist perspective. Drawing on familiar notions of crime as a social construction, I explain how and why biocriminology’s ontoepistemology is inconsistent with the social reality of crime for scientific not ideological reasons. I explain that recognizing crime is a social construction does not imply that crime is not real or objective and cannot be studied scientifically. On the contrary, the irreducibly social nature of crime requires that scientific realists reject assumptions of ‘biological crime’ as well as the biologically reductionist epistemology on which biocriminology depends.

Keywords: biocriminology, biological essentialism, social constructionism, reductionism, genetics

Scholarly efforts to link biology to crime have once again been gaining steam. Departing from earlier approaches, most contemporary biocriminology adopts interactionist (bio + social) language, emphasizes the end of nature versus nurture models, and acknowledges that genes are not destiny and biological determinism is flawed human science (Rafter et al., 2016). Consequently, 21st century biologically informed criminology is positioned to be more influential than earlier notorious iterations (e.g., Lombroso, 1876/2006, Hooton, 1939). In this era, we are assured, the biopathological “criminal type” is widely recognized to be a fallacious concept, and bio-cautious social scientists can let down their guard knowing that today ‘born criminal’ explanations would be quickly dismissed if not ridiculed.

Assurances notwithstanding, whether biocriminology has undergone a decisive paradigm shift rejecting notions of born criminals, bad brains, and criminogenic genes remains uncertain. Several critics have argued that biocriminology continues to adopt a biologically essentialist (i.e., biologistic) ontoepistemology with its biological criminal type (Carrier and Walby, 2014). However, the persistence of this biologistic ontoepistemology is obscured by flashy new technologies and methods, such as GWAS and polygenic scores, and hyped claims about the power, precision, and potential of biologically informed analyses and interventions (Meloni, 2014).

To evaluate science of biocriminology, we need to have clarity about the presuppositions of the worldview undergirding this approach. Unfortunately, most debates around biocriminology are mired in politics, obscuring important scientific issues. On the one side, biocriminologists accuse critics of letting their liberal political sensibilities guide their research rather than adherence to a value-free science (see Walsh and Wright, 2015; Walsh, 2011). Critiques are thus cast as political-ideological (“environmentalist ideology” or “equalitarian dogma”) and dismissed as an “ideological (i.e., non-scientific) denial[s] of biological inequality” (Carrier and Walby, 2014: 12). On the other side, some critics have resorted to ad hominem, imputing (‘bad’) political motives to biocriminologists and connecting contemporary biocriminology with horrific eugenic and/or racist motives or aims of the past. The result is that sociopolitical and ethical debates have displaced scientific ones. Critics and proponents speak at not with each other; consequently, controversy and misunderstanding persist (Panofsky, 2014).

This is a mistake. Contemporary biocriminology continues to suffer from underexamined scientific inadequacies (Carrier and Walby, 2014). Yet, these deficiencies remain on the margins of criminology in large part because they are not well understood. It is much easier to ignore critiques we do not understand. Here I aim to address this deficit by elucidating biocriminology’s scientific inadequacies, with a specific focus on illuminating reality-incongruent assumptions and explaining why these thwart biocriminology’s contributions to criminology and controvert its claims to scientific superiority.

Central to my discussion is the widely agreed upon fact that crime is socially constructed (e.g., Duster, 1990; Schur, 1979; Carrier and Walby, 2014). Here I aim to explicate what this constructionist ontology implies for the biocriminological enterprise. In particular, why irreducibly social crime is incompatible with biocriminology’s reductionist epistemology has not been clearly explained and remains unappreciated. The result is that some scholars seem to view the recognition that crime is socially constructed to involve a ‘radical postmodern, anti-science stance’, which implies that ‘crime is not real or objective’. What is missing from this literature is a detailed, accessible explanation of why the irreducibly social nature of crime derails the biocriminological project for scientific not sociopolitical reasons. Addressing this gap is the aim of this work.

In what follows, I explicate the underlying assumptions of biocriminology and their implications from a scientific realist perspective. I adopt an ontologically pluralist approach, which rejects biological reductionism as the aim of social science. As I discuss, this pluralist view recognizes the existence of different things in the world at different levels—things which need not and often are not reducible to entities at lower levels (such as biology and physics). This approach rejects a hierarchical view of science where the social sciences are subordinate to biology, which is subordinate to chemistry and then physics (Cartwright, 1999; Dupré, 2001). Following others, I argue that crime’s irreducibly social nature is incommensurate with a biologically reductionist ontoepistemology. Moving beyond assertion and setting aside ideology and politics, I explain how rejecting ‘biological crime’ is not just compatible with scientific realism, it is justified by it. I conclude by arguing that scientific realists must recognize the irreducibly social nature of crime and, therefore, reject the ontoepistemology of biocriminology as a means of advancing useful knowledge on crime and criminality. Although philosophical, these issues are not idle or frivolous; they are of direct relevance to the research process and products—the science—of criminology.

Notably, this is a ‘critical’ account that interrogates the assumptions of biocriminology and explores whether these are commensurate with the reality of crime. This is not an ideologically driven hit piece calling for scientific censorship of a program whose findings may not be politically correct. I have no interest in the motivation of biocriminologists (which I have no reason to believe are any different than my own—understanding crime to prevent crime and reduce human suffering). I do not discuss either the epistemic virtues or contributions of the biocriminological program. Fortunately, these are inscribed in most biocriminological articles (e.g., Walsh, 2009; Wright and Cullen, 2012). Finally, I do not claim any special position to adjudge the merits of any scientific endeavor; rather, my goal is to stimulate discussion via critique.

The following discussion proceeds in two parts, which present interrelated yet distinct explanations. First, I discuss the ontological assumptions of biocriminology and their implications, specifically focusing on how biocriminology necessarily biologizes crime and what this means given that crime is irreducibly social. In the second part, I explain why the social nature of crime is incommensurate with biocriminology’s epistemology.

Biocriminology’s Essentializing Ontology

All theoretical frameworks contain assumptions, which presuppose particular understandings of reality linked into a coherent worldview (Ellis, 2001; Williams and Arrigo, 2006). In criminology, ontological assumptions about the reality of crime and of human behavior are principal and shape not only the questions asked but the methodologies appropriate to answer them (i.e., epistemologies). To the extent that biocriminology starts from a reality-incongruent ontology (i.e., view of its subject and the basic constitution of the world in relation to crime), the scientific value of the program is seriously undermined. Understanding and assessing the adequacy of biocriminology’s ontology is thus an important task for theoretical criminology.

Biologized Crime

In biocriminology, crime and criminality are conceived as biological types or ‘kinds’.1 Of course, that crime involves biology is both true and completely banal because all human actions necessarily involve biological processes. However, biocriminology conceives of crime not as involving biology in this prosaic sense, but in the sense of biological differentiation. On this view, crime is a natural object (event or behavioral category) with intrinsic biological features (or biological essences) that distinguish it from non-crime (see also Carrier and Walby, 2014). Similarly, ‘criminals’ and ‘non-criminals’ are distinct biological kinds or types of persons, with distinguishing biological properties. The assumption is that the crime – non-crime distinction is ontologically dependent on—as existing in part due to—a biological difference.

This biologizing ontology undergirds the aim of identifying biological differences between criminals and non-criminals and/or the biological elements of criminality. After all, if crime and criminals were not biological kinds with differentiating biological characteristics, the search for biological differences would be nonsensical. Notably, biocriminology’s ontology does not require or imply that all differences between criminals and non-criminals are biological only that the two groups can be demarcated based in part on differentiating biological factors (e.g., genes). To quote biocriminologists, this ontology “does not mean that all members of a class of people are identical only that they share essential properties that mark them as one thing [criminals] rather than another [noncriminals]” (Walsh and Wright, 2015: 131).

To summarize, biocriminology necessarily assumes that crimes are biologically distinct from non-crimes as a product and expression of a distinct biological pathology—what Carrier and Walby (2014) deem the ‘biopathologizing gaze’. All theories are constructed on a foundation of assumptions; however, to be useful they must be consistent with reality and empirical evidence (see Dupré, 2001).

Natural Essences and the Reality of Crime

The deepest scientific difficulty with the biocriminological paradigm is its biologizing ontology. Crime is a social, not a biological, kind. Surely the most influential product of the 1960’s and 1970s radical/critical criminology was the permanent disruption of the assumed naturalness of crime (e.g., Becker, 1963; Chambliss, 1964; Duster, 1970; Schur, 1971). Over the past several decades, critical criminology has fostered a near consensus that rather than being natural and inevitable, crime is socially created—inherently contingent, relative to time, place, and situation. Yet strikingly, this is usually ignored in biocriminological scholarship and assessments of this work (but see Carrier & Walby, 2014; Duster, 1990; Fallin et al., 2019), and the implications for biocriminology are widely misunderstood.

In what follows, we consider what it means to say that crime is a social construction; illustrate the construction of crime; consider biocriminology in light of the social reality of crime; and then discuss what this implies for biocriminology’s reductionist approach and empirical findings.

The Social Construction of Crime

As labeling theorists highlighted roughly a half century ago, crime is a social construction. Using Haslanger’s (1995) terminology, crime is a constitutive social construction, which means the classification of crime is inherently social. Crime is not something that exists in nature waiting to be discovered. As Becker (1963) famously noted, we create crimes by identifying acts that we deem unacceptable in a particular context, by prohibiting them via a socio-political process, and by labeling (certain) offenders who commit such acts as criminals (or outsiders).

Constitutive social constructions are created by what Searle (1995) calls “Y status functions”; this is the human imposition of a status Y on an object X, a status that X does not possess based on its material (physical or biological) constitution. Y status functions create new social categories (properties, objects, events) without adding any biophysical features to the world. Because the biophysical features of X are insufficient, the resulting Y status must be constituted by collective social agreement (or by coercion) (Searle 1995).

In its most basic form, Y status functions are created as X counts as Y in C, with C being contextual factors (Searle, 1995). For example, X (smoking marijuana) is Y (crime) in C (Alabama). The criminal status of X, then, is created by the socio-political act of criminalizing marijuana in the state of Alabama not by the physical or biological properties of X (the act of smoking marijuana), which is why X does not count as Y (crime) in the state of Washington (if over age 21). Such constructed Y categories, which are called social or nominal kinds, exist only because we create them, invariably layered on top of other social creations (e.g., nation-states, legal institutions). In so doing, we make distinctions among things that exist irreducibly in a matrix of social relationships rather than things that simply exist in nature awaiting discovery (Hacking, 1999; Popper, 1974).

In contrast to social kinds, natural kinds, such as hydrogen, clouds, biological sex, and coral, exist out there in nature awaiting our recognition and identification (Ellis, 2001; Searle, 1995). That is, whether something is or is not a member of a natural kind is decided by natural features not by social agreement or fiat (Ellis, 2001; Searle 1995). For example, copper is copper (an X object composed of specific chemical features) whether it is in New York City, New Guinea, or Mars. The element that exists in nature that we call copper exists as X in nature without regard to C (even if what we label it may vary across C).

Criminals are clearly not a natural kind. Like college students, bachelors, and jocks, the category of criminals does not exist in nature waiting to be discovered. Although the social character of criminals and the institutional dependence of criminal facts is largely, if not wholly, uncontested, the implications of this constructionist ontology are often misunderstood. One extreme interpretation is that it implies a rejection of the existence of an objective or independent reality making scientific study of crime an impossibility. This does not follow from such a recognition. I do not take such a position (and very few do; but see Hulsman, 1986; Muncie, 1998). To acknowledge that crime is a social kind does not mean that crime is not real or without meaning (see e.g., Walsh and Wright, 2015 for that interpretation). Ontological realists can recognize both socially constructed kinds and the existence of an objective and knowable reality (see Searle, 1995). Indeed, I argue that scientific realists must recognize the social construction of crime.

Moreover, recognizing that crime is a social construction does not mean that there is no ‘objective crime’. Ontological constructionism does not preclude epistemological objectivity (Searle, 1995). We can thus accumulate objective facts about crime, as criminology does, even though such facts are not given in nature (like facts about water) but rather are created irreducibly via social processes. Explicating this distinction, Searle (1995: 1) notes that “there are portions of the real world, objective facts in them, that are only facts by human agreement,” and he distinguishes “social or institutional facts,” which require human institutions for their existence from “brute facts,” which require no human input for their existence.2

Crime is thus a social construction based not on any intrinsic biophysical characteristics but rooted in social factors, perhaps most obviously legal institutions, but also social context (time, place, situation) and social categories of individuals (e.g., minor, incapacitated). Likewise, criminals and criminality, as they are rooted in the social construct of crime, do not exist independent of these social conventions. The reality of this social construction of crime is reflected in the familiar adage that the most facile way to get rid of all crimes is to eradicate criminal laws (Michael and Adler, 1933).

Essentializing Errors and the Impossibility of Naturalizing Social Kinds

Biocriminology relies on a biologistic conception of crime; theorizes criminals as being a natural kind sharing biological essences; and assumes these differentiating bodily properties can be isolated without regard to context. This naturalizing ontology works well for natural objects, such as copper, and natural processes, such as meiosis, but not for social constructions like crime.

Although most biocriminological scholarship remains largely silent on this biologizing ontology, a few scholars address this topic in their work and endeavor to defend these assumptions. In a cogent discussion of the social construction of crime, Ellis and Walsh (2000: 7) acknowledge that much of crime is socially contingent, arbitrary, and varying, pointing to substance use and other mala prohibita crimes, which they refer to as “moving targets.” However, they contend that there is a “stationary core” of crime, which consists of acts “such as murder, assault, rape, and theft” that are “universally condemned offenses.” In this view, the stationary core offenses “all have certain features in common that distinguish them from nearly all other acts,” and presumably, this focus eliminates the thorny social construction aspect (Ellis and Walsh 2000: 7). Although appreciating their attention to constructionism, the extent of cross-cultural, historical, and situational consensus is overstated. The “stationary core” is still not a natural kind. To illustrate the irreducibly social nature of crime, I discuss a few allegedly “universally condemned offenses”.

Consider killing, which in its criminal form as murder is invariably deemed the most serious crime; one which is often pointed to as inherently criminal, immoral, unconventional, and universally condemned. Except it is not. Murder, like other crimes, is socially constructed. The intentional, even premeditated, killing of one person by another—generally the definition of criminal murder—can be expected (as in honor killings or duels), lauded (as in war or battles), excused or justified (as in self-defense), or even ignored (e.g., plantation owners killing ‘their’ slaves in the ‘Old South’). As Cooney (2009) discusses in his book “Is Killing Wrong,” the intentional killing of one person by another is not inherently criminal, but rather social factors—including when, where, and who is involved—determine which killings are prohibited and treated as murders (see also Hoebel, 1954). The act of intentionally killing another human – its materiality (physical or biological) – is insufficient for its criminal status, and any category of people who kill people would include a variety of individuals whose killings were not murders. Biological factors do not distinguish killing from murder.

Turning to the “stationary core” crime of rape, we can see clearly that this crime is also varying, contested, and socially contingent. Biocriminology’s ontology would presume that the formerly legal act of forcing sex on female slaves was biologically distinct from the illegal forcible rape of white women, which was also biologically distinct from the legal forcible sex of wives by their husbands in the 18th century. This example reveals most clearly how irreducibly social crime is. We would not only have to consider context (18th century) but also as part of that context, who is committing the act against whom (socially defined categories) in what relationship (e.g., marital or not). So for this example, we might see: X (forcible intercourse) counts as Y (crime) when committed by a white man against a white woman (socially constructed racial classifications) who is not his wife and given other characteristics (at this historical time, classification of an act as rape required that a woman physically resisted her attacker; there was corroboration of this both the force and the resistance; and she was known to be chaste). Clearly, X (forcible intercourse) is not essentially Y (crime). Thus, even for serious “stationary-core offenses” the demarcation of illegal from legal relies on social context. Such acts cannot be reduced to a biological essence or an act that is biologically different in kind.

Given the remarkable variation across human cultures, historical time, and place, presuming a context-invariant biological property linking this heterogenous group of expected, acceptable, and highly deviant and illegal acts flies in the face of common sense and empirical evidence.3 When categorization shifts based on national boundaries, court rulings, age, situation, etc., the essence of this categorization cannot be reduced to intrinsic biological features of the act or the actor (e.g., Duster, 2006; Garland, 1985; Schur, 1979).

Yet, some may argue that we – members of Western, Educated, Industrialized, Rich, and Democratic (WEIRD) societies (Heinrich et al., 2010) – have identified the real crimes and criminals. These beliefs flow from successful socialization into our culture (e.g., Quinney, 1970; Reiman, 1979; Searle, 1995) with perhaps a hefty degree of confidence in our moral righteousness (Carrier and Walby, 2015). Our cultural frames encourage us to consider what we treat as serious crimes as the truly serious crimes, such that we tend to view other societies with divergent conceptions of crime as misguided, even deficient or depraved. Yet ours is not a neutral or ‘The Right’ system from which to judge ‘Real Crime’. The point is not that there is a better way to do things, although I believe that is true, but that ours is a historically specific way of doing things that is neither natural nor inevitable but one of many possibilities (Currie, 1985; Quinney, 1975; Young, 2011).

As one final note in this vein, biocriminology’s ontology assumes the ‘criminal’ is clearly delineated and distinct from the ‘normal’ noncriminal; crime is an abnormal marginal category; and such a majority/minority binary is rooted at least partially in biological deficiencies (e.g., Walsh and Wright, 2015). Notably, however, research in WEIRD societies suggests not only that law violation is common but also that it is statistically normal behavior. Insofar as most of us have violated the law, identifying a unique (statistical minority, inferior or dysfunctional) biological criminal type or essence characteristic of persons—a pathological ‘Other’—is nonsensical.

In sum, and quite simply, crime is not, and by virtue of its grounding in a socio-political definition cannot, be essentialized to the biological (genetic, hormonal, brain) level. A biocriminological explanation requires biological differentiation. Crime is irreducibly social. The point is not that there is no place for science in relation to crime. On the contrary, the point is that a scientific approach to crime must recognize that crime is a social construct.

The Epistemology and Alleged Superiority of the Biocriminological Approach

Some biocriminologists advertise their approach as more truly scientific than mainstream (“sociological”) criminology (e.g., Walsh and Wright, 2015; Wright and Morgan, 2015; Wright and Cullen, 2012), with implicit and explicit assumptions about what it means to be more scientific. This includes a commitment to essentialism, discussed above, as well as reductionism, which Walsh and Wright (2015: 128) define as “the process of taking causal explanations from fuzzier levels to deeper more precise levels” (emphasis added). Some biocriminologists explicitly privilege the biological over the sociological—where the sociological is “fuzzy,” “stale,” and “ideological”, the biological is “more precise,” “fresh,” and “objective” (e.g., Walsh and Wright, 2015; Wright and Cullen, 2012). Biology is proposed as a “more scientific” foundation for anchoring criminological knowledge. Indeed, in several instances, scholars explicitly align biocriminology with “science” and disparage those who question biocriminology as “anti-scientific” and motivated by political correctness and ideological biases, specifically “radical liberal philosophy” (Walsh and Wright, 2015: 137).

In this section, we examine biocriminologists’ claims about ‘superior science’ by scrutinizing the bio-reductionist epistemology they confidently embrace. I begin with a discussion of what we understand by reductionism and its characteristics that I claim make it inappropriate for criminology.

Biocriminology’s Reductionist Epistemology

Perhaps the most misconstrued aspect of biocriminology is its reductionist epistemology, defined roughly as the position that the proper aim of all science is a unification such that higher-order phenomena are explained by the composition and structure of the lower-level entities which compose them (Nagel, 1961; Fodor, 1974). To quote biocriminologists on this issue: “Phenomena may find their meaning in holistic regions, but they are explained by lower level mechanisms”; “Fundamental knowledge… is reductionist” (Walsh and Bollen, 2012: 1-2; emphasis in original). According to this view, “[t]here is nothing in the whole that is not contained in the parts” (Walsh and Wright, 2015: 129).

Although reductionist approaches have contributed significantly to knowledge advances at certain times and in certain fields, it does not follow that this is the superior way of advancing all knowledge. Philosophers of science continue to debate the merits of reductionist approaches and their limits (e.g., Dupré, 1993; Gallagher and Appenzeller, 1999; Kitano, 2002), and longstanding efforts to encourage scientists to move “beyond reductionism” continue (Dupré, 2012; Kaiser, 2011). In what follows, I aim to provide a perspicuous account of what scientists mean when they talk about reductionism given its various usages, a discussion that provides explanation for the longstanding critique that biological reductionism is not an appropriate epistemology for criminology.

Reductionism: What Is All the Fuss About?

Scientists often speak at each other in discussions of reductionism, a pattern which is understandable given the different usages of the term and varying degrees of acceptance (e.g., Ayala, 1974; Kaiser, 2015). The two principal types of reductionism—methodological and explanatory—are overlapping, and biocriminology subscribes to both. Methodological reductionism is a thesis about the proper practice of research, and proponents of this approach claim that entities should be studied by applying reductive methods (Ayala, 1974), which involves focusing on lower-level parts and isolating their influence from other factors (Kaiser, 2011). Explanatory reductionism is the product of the methodological approach—the explanation. In form, this consists of the replacement of the higher-order explanans (i.e., the explanatory factors) with lower-level entities, such that the higher-order explanandum (i.e., phenomenon to be explained) is fully explained with lower-level parts and the laws governing these parts (Oppenheim and Putnam, 1958; Fodor, 1974; Schaffner, 1967). Although these definitions are rather straightforward, the basis for biocriminologists’ argument that this represents the proper way to do science requires a bit more explication.

Reductionism and its requisite essentialist ontology (here biological essentialism) are bridged by assumptions of ontological monism and the unity of science (Dupré, 2012; 1993; Fodor, 1974). Ontological monism is the thesis that the world (reality) is ultimately composed of only one kind of stuff (the basic particles of physics), and the doctrine of the unity of science presumes that science is unified by the study of the laws of this one kind of stuff (Dupré, 2001). These assumptions undergird the epistemology known as classic physicalist reductionism—that scientific knowledge about large complex things could ultimately be derivable from the laws of physics (Dupré, 1993; Nagel, 1961). This approach assumes a hierarchical relation among the subjects of sciences—social, biological, chemistry, physics—and the completeness in principle of an explanation of everything in the universe through their unification, with the idea that reality can be explained by the scientific laws of physics through explanatory reductionism (Cartwright, 1999; Dupré, 2012; Nagel, 1986). These epistemologies are associated with the view that the lower-order sciences and their elements are more scientific and more precise (i.e., closer to the truth of physics), than the higher-order ones. On this view for example, sociology is scientifically subordinate to biology, which is subordinate to chemistry, and then physics.

At present, most scientists reject classic physicalist reductionism as a practical aim of science (see Dupré, 2001; Kaiser, 2011). The limited broad success of reducing higher-order entities to the laws of physics as well as the recognition of the impracticability of reaching such a goal, given both the complexity of the world and the different kinds of things in it, has produced a revision in the reductionist narrative and goals. Classic physicalist reductionism was substituted with a more realistic, softer version: “hierarchical or gradual” instead of “precipice” reductionism (Dawkins, 1986); “uncompromising reductionism” versus “compromising reductionism” (Weinberg, 1992), and, in the most famous form, Dennett’s (1995) widely referenced “preposterous” or “greedy” versus “bland” reductionism. These softer neo-reductionist approaches do not require the assumption that everything can be reduced, in practice, to the particles of physics. Rather, this neo-reductionist epistemology proceeds with lower-level replacement such that explanations refer exclusively to entities located on a lower level than the explanandum and do not mention factors on a higher level (Kaiser, 2011; 2015).4 This is the “so-called methodological reductionism” adopted by biocriminologists, which involves reducing crimes or criminality to the biological level (Walsh and Wright, 2015). Despite its compromises, this softer reductionism is still not a viable strategy for criminology.

Lest I be misunderstood, I am not arguing that the epistemological reductionism is inappropriate in all cases. Scientists have made some extraordinary advances in addressing how things work in a reductionist fashion. however, extended into a general worldview applicable across all sciences, it is misguided. This is due in part to the fact that the assumptions undergirding reductionism—both the doctrine of the unity of science and the ontological monism it presupposes—are myths, in both senses of the term myth: that they are literally false (not supported by empirical evidence) and serve certain functions for those who adhere to it (Dupré, 2001; 2012; Searle, 1995; Skolimowski, 1974); we discuss both next.

The Myths of Ontological Monism and the Unity of Science

A monistic metaphysics assumes that real things, the proper subject of science, are out there in nature, awaiting identification and explanation via replacement by their lower-level constituents. The assumption is simply that causality lies in the parts and their structures, and that whether in a lab, on Mars, or in Times Square, the lower-level causal forces identified in the explanans will explain the explanandum (Kaiser, 2011).5 Biocriminology, implicitly via commitment to explanatory reductionism, privileges explanations rooted in intrinsic, structural properties over explanations appealing to extrinsic properties, such as environmental or situational factors.

As the earlier ontological discussion makes clear, a reductionist biocriminology immediately runs into the problem of translation across levels. Entities at higher levels do not necessarily (and often do not) correspond to entities at lower levels. Biocriminology assumes that there is a physical (biological) realization of crime (versus non-crime) that can be fruitfully explored with reductionist methods. That is, the materialization of crime depends on the biological such that for any criminal act there is a specific biological state that constitutes it, and wherever we were to find an identical biological state we would find an identical criminal state (Dupré, 2001). However, socially constructed objects such as crime contravene such causal fundamentalism – the idea that “the macro is the way it is in virtue of how things are at the micro” (Jackson and Pettit, 1992: 5). The social kinds of criminology—rape, murder, theft – are not translatable to lower-level biological entities because their criminal (Y) status invariably relies on numerous contextual factors – not merely the materialization of biological potentialities. As discussed, rape is not determined by the biological features of the act itself; similarly, theft, requires among other things the adoption of a specific set of institutional rules about private property. Criminological explicanda are socially defined entities that differentially apply to the same bio-physical realization (X) across time, place, and situation.

According to a reductive monistic metaphysics, reality is composed of one kind of stuff, and the behavior of higher-level phenomena can be explained by the operation of its constituent parts. If entities are not reducible to their lower-level components (biology, physics), they are not amenable to scientific study (and possibly are not ‘real’). Crime, as discussed, is not reduced to the operation of lower-level biological components. As such, we have a dilemma: either crime is not real (an eliminativist position on crime) and cannot be studied scientifically, or a reductive monistic metaphysics is not consistent with reality (i.e., is false).

Crime is real; we do not always agree on the definition6; and it varies across space and time, but we have acts classified as crimes that are committed in the material world. Thus, crime is not reducible to lower biological levels, but it is real. Following others (Cartwright, 1999; Dupré, 2001; Popper, 1974), I believe a monistic metaphysics is false, but this does not require a retreat from science or a rejection of the possibility of objective knowledge (e.g., Walsh and Wright’s (2015) suggested “powerful anti-scientific forces” including “post-modernism and radicalism”). Instead, this dilemma can be resolved with a pluralist ontology.

Ontological Pluralism and Scientific Parity

In contrast to a reductive monism, a pluralist ontology recognizes the existence of multiple kinds of things in the world that exist on different levels (Anderson, 1972; Dupré, 2001; Popper, 1974). A fundamental feature of ontological pluralism is the distinction between social and natural kinds. Although many things in the world are reducible to, even explained by, their lower-level components and can profitably be studied in a (neo)reductionist fashion, many other real things, such as crime, are not directly translatable to lower-level entities, and can still be scientifically studied, just not usefully in a reductionist fashion. Humans carve up the world in a variety of ways for various purposes, and these social kinds (properties, objects, and events) are still real and are studied scientifically. Importantly, this pluralist perspective is consistent with a materialist stance. Reality can be composed out of the particles of physics, but because social constructions are often not commensurate with schemes of classification at lower levels (biological, chemical), they do not correspond with or neatly map onto – and thus are not reducible to – their lower-level entities.

Criminological objects, such as criminal theft, rape, and incarceration, are real things. We count crime, we fear it, and we spend careers trying to predict and explain it. To suppose that social kinds like crime are appropriate for biological reductionist explanations is to assume, quite without empirical or theoretical warrant, that nature provides us with the category of ‘crime’ that we in fact construct (Dupré, 2001; 2012; also Duster, 2006; Searle, 1995). Rejecting a monistic ontology allows us to recognize that objects of proper scientific study can exist that are not reducible to lower-level entities, requiring a concomitant rejection of a commitment to a reductionist epistemology. If real things exist and they cannot be reduced to lower levels, then reductionism is not an appropriate epistemology for such things.

Furthermore, ontological pluralism rejects the idea of a hierarchy of science in a unified form that prioritizes lower-level ‘precise’ explanations over ‘fuzzy’ higher-level ones based on a one-way causal flow from the parts to the whole. Rather, it recognizes that many higher-level entities have causal powers that are not simply consequences of the structural composition of their lower-level components, as new causal capacities and potentialities emerge with increasing complexity, from cells to tissues and organs and individuals, neighborhoods, and nations (Dupré, 2012). The study of social life entails the study of emergent phenomena that arise, for example, out of the relationships between the social phenomena (Thorpe, 1974). Significantly, this pluralist ontology gives social science autonomy from biology, and the biological sciences autonomy from the physical ones (Dupré, 2001; Popper, 1974). Real things exist on different levels, and biological truths are not inherently closer to reality than social ones.

What are the practical (theoretical and methodological) implications of this philosophical discussion? A pluralist ontology, which neither privileges lower-level explanations nor accepts reductionist approaches as the way to do science and ‘as explanation’, subverts the notion that biology and reductionist explanations are closer to the truth or more scientific than social science explanations. It follows that the claim by some prominent biocriminologists that incorporating biology and reductive methods into criminology will provide a more scientific foundation for the discipline is recognized to be propaganda dependent on empirically unsupported and reality-incongruent assumptions (see also Carrier & Walby, 2014). Perhaps more important, seen in this light, the guiding assumption behind biocriminology – the notion that there must be something biologically essential about crime because it is real and real phenomena are biologically differentiated – is seen to be misguided. Again, biological elements and processes are always involved in human actions, but these processes do not distinguish social kinds like crimes. Finally, acknowledging that biological essentialism and reductionism is inappropriate for criminology also means confronting the myth of ceteris paribus guiding biocriminology. The idea that we can isolate biological elements (e.g., genes) as primary, context-independent causes of nominal kinds like crime is specious.

The Impossibility of Ceteris Paribus

An essential characteristic of the epistemology of biocriminology, and an inherent feature of reductionism, is the examination of constituent parts in isolation (Kaiser, 2011). The focus is squarely on factors that are internal to the higher-order phenomenon, ignoring external factors such as contextual ones (Dupré, 2012; Kaiser, 2011). This reductionist procedure simplifies by making the investigation manageable and thus identification of parts and their connections within the system easier (Bechtel & Richardson 2010). However, while simple may be easier, we should not confuse simple and easy with precision and truth.

Crime emerges from a situation embedded in a specific social context. Recognizing this, biocriminological accounts generally rely on the notion of ceteris paribus—holding constant contextual features to explore the effects of individual biological properties. This model works well for highly stable systems and controlled laboratory experiments; however, human development and criminal behavior are neither highly controlled nor stable. We cannot isolate constituent parts of the crime or ‘the criminal’ outside of a given context because both are defined by their very context—from prenatal and family influences and cultural overlays to the normative and political structure of our institutions (also Fallin et al., 2019). Human social behaviors are always embedded in a web of social relations, which give meaning to and impose contingencies on biological parts.

Both the importance of and misleading nature of the assumption of ceteris paribus can be illustrated with a focus on genetic causation. Recent sociogenomics work emphasizes the causal primacy of the gene based on the idea that genes (and genetic influences on crimes) preexist social influences and are fixed at conception. On this view, often valid criminological concerns of reverse causation are irrelevant because genetic influences are always causally prior to social ones. However, in addition to the fact that postgenomic knowledge advances reveal clearly that genes are not fixed charges exerted independently of environmental contexts (Griffiths and Stotz, 2013), individual genotypes are not causally prior to social factors. Social structures and social networks—as matrices of unequal, non-arbitrary relationships—predate conception, and these macro-structures sort individuals—and their genotypes—into different and unequal social positions from birth.

Our socio-political system creates specific goals and rules and funnels individuals (through specific policies and social processes) who carry unique genotypes into different contexts with different, even incomparable, opportunities and constraints. These social opportunities, rules, and goals are created not arbitrarily or blindly, but as an outcome of a matrix of unequal power relations that influence what biopsychosocial capacities are privileged (e.g., impulse control versus attention shifting; mental versus physical stamina) (Burt, 2018). These structures also criminalize some harmful actions and not others (e.g., Quinney, 1975; Schur, 1969). For example, genetically influenced traits such as skin pigmentation impose significant constraints on development (and the likelihood of crime) for reasons that are neither caused by nor reducible to genes but significantly shaped by social relations that can only be understood in a sociohistorical context (Burt and Simons, 2014).

By mistakenly focusing as the genome as the source of primary causal power, the fact that preexisting social structures both sort genotypes into separate and unequal contexts and give (different) meaning to phenotypes based on social position is neglected.

Perhaps the clearest example of the context-specific effects of biological (here genetic) influences as is found in the famous experiments by Clausen et al. (1940) on the plant Achillea millefolium. For seven different individual plants (genotypes), three clones were created; one of the clones for each plant was planted at low elevation, intermediate elevation, and high elevation. The results of these experiments showing how genetic clones grew in different environments is shown in Figure 1, arranged horizontally in order of how well they grew at the lowest elevation. Immediately, one can see that it is not possible to select a ‘best growth’ plant based on the results at one altitudinal context. We cannot properly answer the question “Which genotype produces the best growth,” without specifying the altitude, and as Lewontin (2001) noted, even averaging over the environments is not very informative.

***Figure 1 about here***

While not diminishing the importance of altitudinal context and the complexity of plant height, socially constructed behaviors like crime are surely more context-dependent across both physical and social layers. For example, in our society some individuals are born into financial stability (and the consistent resources and safety such wealth provides) while others are born into severe disadvantage, where resources are scarce, safety is unassured, and prospects for upward mobility are bleak. The gap between these groups is large and growing, and, in a real sense, the notion that we can consider these “similar contexts” of development from which we might gain generalizable information about genetic causation is dubious and empirically unjustified. Add to this the well-known fact that the harmful behaviors of the disadvantaged are subject to more penal social control as well as the fact that they have less privacy and are subject to greater surveillance that those who are more well off (Currie, 1985; Schur, 1969, Reiman, 1979), and this notion that we are comparing like to like, much less achieving ceteris paribus, is increasingly difficult to defend.

Although heterogenous, our society is unique in many ways, which influence, among other things, our high rates of crime and incarceration. Mainstream U.S. values such as maximizing personal profit at the expense of others – whether by unlivable wages for workers, pricing necessary goods far beyond their cost (e.g., Epipens); or harmful polluting to make a profit – are accepted (if not encouraged) (see Reiman, 1979; Currie,1985; Messner and Rosenfeld, 1994). As noted earlier, ours is not a neutral context from which to isolate biological influences any more than that of Japan (similarly developed, drastically lower crime), Sweden, India, or New Guinea, Israel, or Ethiopia. The biocriminologist program of isolating biological causes from both contextual influences and social meanings misrepresents the reality of crime and is undergirded by the empirically unwarranted myth of ceteris paribus. Crimes and criminals emerge from a social matrix defined by an irreducible complex of Y status functions shaped by power (Quinney, 1970; Reiman, 1979). Like all other biological factors, genes are not causally primary nor are they principal movers of developmental processes, and social structures that preexist our birth and which we enter in a particular social position at a historical time and place are always relevant to development because we do not live in a world of all else equal.

In sum, an epistemological reductionism rooted in a monistic metaphysics is incommensurable with our ontologically pluralist world (Dupré, 2012). To be sure, reductive explanations are alluring. Research suggests that when given a choice of two explanations, people tend to prefer the explanation that refers to lower-level components or processes even when the information about the lower-level is irrelevant (Hopkins et al., 2016). This is perhaps especially true for crime. The conclusion offered by biocriminology – it’s not us, it’s them – both frees us/society from complicity while promising a scientifically precise and certain key to crime problems. Despite this appeal, there is no theoretical or empirical justification for taking a biological reductionist stance toward crime. Epistemological reductionism is clearly inappropriate for criminology because: 1) crime is not a biological construct (ontological monism is a myth), and 2) the causes of crime are not reducible to a mechanistic biology (causality is not unidirectional from parts to wholes). Given the absence of any compelling evidence, it seems sensible to accept the inapplicability of a reductionist approach to socially constructed behaviors like crime.

Notably, to call a biological reductionist criminology epistemologically unsound neither implies a blank slatist or social determinist model nor is it to “demonize biology,” as Walsh and Wright (2015) presume. The evidence is clear that we are not blank slates; genetic differences between individuals do – in a context-sensitive way – influence development and phenotypes including aptitudes and susceptibilities in concert with social-environmental factors. Biological influences on human social behavior (e.g., genes) are but one of many factors that comprise the complex human developmental system that is always operating in and defined by its social context. This rejection of explanatory reductionism is not based, as some biocriminologists assert, in the “concern of social scientists…that explanation at the lower levels entails the elimination of higher levels of explanation” (Wright and Walsh 2015: 129). Rather, it is a recognition that an essentialist, reductionist biocriminology is not congruent with reality. To say this is in no way to dismiss all reductionism, but to specifically debunk a biologically reductionist account of crime and the unsound science to which it naturally leads. The ontology of crime is extended and relational—the act is not located in a person’s brain or body, but in emergent relations with others (objects and persons in context). The actions of humans are more than the sum of their biological parts, which is how we can fly across the ocean and watch sports rematches with uncertainty about the outcomes.


The scientific deficiencies of biocriminology’s ontoepistemology are generally relegated to the margins of criminology. Arguing that this is due in part to widespread misunderstanding of what the irreducibly social nature of crime implies, this paper focused on identifying and explaining the implications of biocriminology’s ontoepistemology. In so doing, I addressed why these reality-incongruent assumptions thwart biocriminology’s contributions to criminology and controvert its claims of scientific superiority. I elucidated the program’s biologically essentialist ontology of crime and the criminal, which presumes an empirically and theoretically unwarranted biological differentiation of ‘crime’ vs. ‘non-crime’ and ‘criminals’ vs. ‘normals’. Throughout I highlighted how the social reality of crime is incompatible with biocriminology’s reductionist epistemology, scientifically undermining the approach.

The reason for devoting this much attention to critic the philosophical assumptions of biocriminology is because they are often misunderstood and hold deceptive popular appeal. Moving forward, an understanding of the flaws in biocriminology’s ontoepistemology may inform a germinating ‘biopsychosocial criminology’, which aims to usefully harness biotechnical advances without biologizing crimes or criminals. Whether, and if so how, criminologists can incorporate biology without biologizing crime and criminals is an important outstanding question. In my view, a prerequisite for a biopsychosocial criminology is a thorough grasp of the flaws with a biologistic approach. Understanding our scientific mistakes is necessary to avoid repeating or perpetuating them.

Invariably, biocriminological discussions are misunderstood and/or misrepresented. Before closing, I want to briefly reiterate what this paper does not suggest or imply. I am not arguing any of the following: a) biology has no role in criminology or crime, b) humans are born blank slates without bio/genetic differences, c) bio/genetic differences can never influence crime, or d) serious, repeat offenders are not more likely to have mental disorders. The evidence is clear that some biological factors do influence social behavior; however, these differences do not map neatly onto the social construct of crime and cannot be examined isolated from social contexts, which influence development and give meaning to biologically influenced differences.

Crime is inescapably social. Our classification schemes, our distinction between crime and non-crime, and our judgment of criminals and non-criminals, are inevitably influenced by many different social factors. These judgments are not tracking any natural or innate facts but are instead capturing (and perpetuating) socially created constructions. Assumptions of innate ‘biological criminals’ are misleading, and natural distinctions between crimes and non-crimes are a chimera. “It is of course a difficult matter of judgement when we should conclude that such a project is irredeemably flawed, but it is surely a judgement we must sometimes be willing to make” (Dupré, 2012: 37).

Acknowledgements: I am extraordinarily grateful to Kara Hannula for her patience and feedback on numerous earlier drafts of this paper. I appreciate those who commented on earlier (and much longer) drafts of the paper, including David Cohen, Mark Cooney, Chuck Lanfear, Michael Jackson, Mike Jones, Jay Joseph, Ron Simons, and several anonymous reviewers. The arguments presented here are my own and do not reflect those of funders or those who provided feedback.

Funding: This scholarship was supported by funding from a career development award from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (5K01HD094999).


Bechtel W and Richardson RC (2010) Discovering Complexity: Decomposition and Localization as Strategies in Scientific Research. 2nd ed. Cambridge, MA: MIT Press.

Becker HS (1963) Outsiders: Studies in The Sociology of Deviance. New York: Free Press.

Burt CH (2018) Racial discrimination and cultural adaptations: An evolutionary developmental approach. Advances in Criminological Theory: Building a Black Criminology 24: 207-252.

Burt CH and Simons RL (2014) Pulling back the curtain on heritability studies: Biosocial criminology in the postgenomic era. Criminology 52(2): 223-262.

Carrier N and Walby K (2014) Ptolemizing Lombroso the pseudo-revolution of biosocial criminology." Journal of Theoretical & Philosophical Criminology 6.1: 1.

Carrier N and Walby K (2015) For sociological reason: crime, criminalization, and the poverty of biosocial criminology. Journal of Theoretical & Philosophical Criminology 7(1): 73.

Cartwright N (1999) The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge UP.

Chambliss WJ (1973) The saints and the roughnecks.” Society 11(1):24–31.

Clausen JD, Keck D and Hiesey WM (1940) Experimental studies on the nature of species. I. Effects of varied environments on Western North American plants. Carnegie Institute of Washington, Washington, DC.

Cooney M (2009) Is Killing Wrong. Charlottesville, VA: University of Virginia Press.

Currie E (1985) Confronting Crime: An American Challenge. New York: Pantheon Books.

Dawkins R (1986) Sociobiology: The New Storm in a Teacup.” Pp. 61–78 in Science and Beyond, edited by S Rose and L Appingnanese. Oxford: Blackwell.

Dennett DC (1995) Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York: Simon & Schuster.

Dupré J (2012) Processes of Life: Essays in the Philosophy of Biology. New York: Oxford University Press.

Dupré J (2001) Human Nature and the Limits of Science. New York: Oxford University Press.

Dupré J (1993) The Disorder of Things: Metaphysical Foundations of the Disunity of Science. Cambridge: Harvard UP.

Duster T (2006) Behavioral genetics and explanations of the link between crime, violence, and race. Wrestling with Behavioral Genetics: Science, Ethics, and Public Conversation: 150-175.

Duster T (1990) Backdoor to Eugenics. New York: Routledge.

Duster T (1970) The Legislation of Morality: Law, Drugs, and Moral Judgment. New York: The Free Press.

Ellis B (2001) Scientific Essentialism. Cambridge, UK: Cambridge University Press.

Ellis L and Walsh A (2000) Criminology: A Global Perspective. Needham Heights, MA: Allyn & Bacon.

Fallin M, Whooley O and Barker KK (2019) Criminalizing the brain: Neurocriminology and the production of strategic ignorance. BioSocieties 14(3): 438-462.

Fodor JA (1974) Special sciences (or: the disunity of science as a working hypothesis). Synthese 28:97–115.

Gallagher R and Appenzeller T (1999) Beyond Reductionism. Science 284(5411):79.

Garland D (1985) The criminal and his science: A critical account of the formation of criminology at the End of the nineteenth century. The British Journal of Criminology 2:109–37.

Griffiths P and Stotz K (2013) Genetics and Philosophy: An Introduction. Cambridge University Press.

Hacking I (1999) The Social Construction of What. Cambridge, MA: Harvard University Press.

Haslanger S (1995) Ontology and social construction. Philosophical Topics 23(2):95–125.

Hopkins EJ, Weisberg DS and Taylor JCV (2016) The seductive allure is a reductive allure: People prefer scientific explanations that contain logically irrelevant reductive information.” Cognition.

Hulsman L (1986) Critical criminology and the concept of crime. Contemporary Crisis 10:63–80.

Jackson F and Pettit P (1992) In defense of explanatory ecumenism. Economics & Philosophy 8(1):1–21.

Young J (2011) The Criminological Imagination. Cambridge, UK: Polity Press.

Kaiser MI (2011) The limits of reductionism in the life sciences. History and Philosophy of the Life Sciences 33:453–76.

Kitano H (2002) Systems biology: A brief overview. Science 295(5560):1662–64.

Lewontin RC (2001) Gene, organism and environment. Pp. 59–66 in Cycles of Contingency: Developmental Systems and Evolution, edited by S Oyama, PE Griffiths and RD Gray. Cambridge: The MIT Press.

Lombroso C (1876/2006) The Criminal Man. Edited by M Gibson and N H Rafter. Durham: Duke UP.

Meloni M (2014) Biology without biologism: Social theory in a postgenomic age. Sociology 48(4):731–46.

Messner SF and Rosenfeld R (1994) Crime and the American Dream. Belmont: Wadsworth Publishing Co.

Michael J and Adler MJ (1933) Crime, Law, and Social Change. New York: Harcourt, Brace & Co.

Nagel T (1986) The View from Nowhere. New York: Oxford UP.

Oppenheim P and Putnam H (1958) Unity of science as a working hypothesis. in H Feigl et al. (eds.), Minnesota Studies in the Philosophy of Science, vol. 2, Minneapolis: Minnesota UP.

Panofsky A (2014) Misbehaving Science. Chicago: University of Chicago Press.

Popper KR (1974) Scientific reduction and the essential incompleteness of all science.” Pp. 259–83 in Studies in the Philosophy of Biology: Reduction and Related Problems, edited by FJ Ayala and T Dobzhansky. Berkeley: University of California Press.

Quinney R (1970) The Social Reality of Crime. Boston: Little, Brown and Co.

Quinney R (1975) Analysis and Critique of Crime in America. Boston: Little, Brown and Co.

Rafter N, Posick C and Roque M (2016) The Criminal Brain: Understanding Biological Theories of

Crime. 2nd ed. New York: New York UP

Reiman JH (1979) The Rich Get Richer and the Poor Get Prison: Ideology, Class, and Criminal Justice. Boston, MA: Allyn & Bacon.

Schaffner KF (1967) Approaches to Reduction. Philosophy of Science 34(2):137–47.

Schur EM (1969) Our Criminal Society. Englewood Cliffs, NJ: Prentice-Hall, Inc.

Schur EM (1971) Labeling Deviant Behavior. New York: Harper & Row Publishers.

Schur EM (1979) Interpreting Deviance. New York: Harper & Row Publishers.

Searle JR (1995) The Construction of Social Reality. New York: The Free Press.

Skolimowski H (1974) Problems of rationality in biology.” Pp. 205–23 in Studies in the Philosophy of Biology: Reduction and Related Problems. Berkeley: University of California Press.

Thorpe WJ (1974) Reductionism in biology.” Pp. 109–36 in Studies in the Philosophy of Biology: Reduction and Related Problems, edited by FJ Ayala and T Dobzhansky. Berkeley: University of California Press.

Walsh A (2009) Biology and Criminology: The Biosocial Synthesis. New York: Routledge.

Walsh A and Wright JP (2015) Biosocial criminology and its discontents: A critical realist philosophical analysis.” Criminal Justice Studies 28:124–40.

Williams CR and Arrigo BA (2006) Philosophy, crime, and theoretical criminology.” Pp. 1–38 in Philosophy, Crime, and Criminology, edited by BA Arrigo and CR Williams. Chicago: University of Illinois Press.

Wright JP and Cullen FT (2012) The future of biosocial criminology: Beyond scholars’ professional ideology. Journal of Contemporary Criminal Justice 28:237–53.

Wright JP and Morgan MA (2015) Human biodiversity and the egalitarian fiction. Pp. 55–73 in The Nature Versus Nurture Debate in Biocriminology, edited by KM Beaver, JC Barnes and BB Boutwell. Los Angeles: Sage.

Figure 1.

Figure 1. Growth of Achillea clones of seven genetically different

plants at different altitudes.

From an Introduction to Genetic Analysis by Suzuki et al. 1996 by

W.H. Freeman and Company.

No comments here
Why not start the discussion?