1. Historical Considerations

Organisms were, at one time, the best representatives of individuality. First, consider how naturalist and evolutionist Julian Huxley professes his views to the philosopher in the preface of his book The Individual in the Animal Kingdom (Reference Huxley1912):

Huxley believed that organisms were more individuated than nonanimate crystals. Organism boundaries were definite: their size and form were defined by a scheme of architecture in contrast to inorganic crystal systems growing without limits. Organisms were more independent in their self-determining qualities (Reference Huxley1912, 51). That is, their agency – their capacity to self-sustain and repair against perturbations – is what made organisms proper objects of biology. For Huxley, they were not only the best representatives of individuality, but organisms were central to navigating differences among organic and inorganic materials.

In contrast, consider the physician Sir William Osler’s Ingersoll Lecture (Reference Osler1904) when he discussed the meaning of death against lessons of embryology:

Osler is not denying organismality’s existence, but rather the significance of individuatedness it’s supposed to represent. Osler’s view draws from Weismann’s germ-soma distinction identifying the germ plasm as central to heredity across generations. He isolated the germline from developmental events of the individual organism’s life cycle (Reference Richmond and LaubichlerRichmond 2001, 169). Organismality for Osler, then, is ontologically secondary to the eternal generational thread.

The historical considerations above yield one lesson about organismality’s significance: Huxley prioritized organismality as an entry point to access relevant features of individuality for life’s evolution (versus changes in nonliving, inorganic material). However, Osler emphasized continuity of genetic lineages for which organismality was just a vessel. As products of their time in the early twentieth century, for Huxley and Osler organismality was representative of individuality, yet as an organizing principle it functioned differently in their approaches.

2. Conceptual Contrasts

How organismal parts work together in the larger system sustaining life was historically conceptualized in relation to structural constitutions of inorganic systems, like crystals and “habits” of minerals exhibiting change and structural order. What distinguished organisms as alive prior to nineteenth century biology was the unobservable, nonmaterial substance elan vital or “the force of life.” However, a post-Newtonian scientific world demanded rejection of mysterious qualities in favor of mechanical descriptions, that is, in favor of how parts function together to produce system-level effects.

Organisms as complex systems were analyzed into component parts by different naturalists including anatomists, physiologists, embryologists, and so on (see Reference HullHull 1978, 336). Reference RuseRuse (1987, 225) argued that individual organisms can be fragmented into structurally various parts functioning together interdependently to sustain the entire unit. At the same time, he acknowledged the complexities of decomposing organisms into discrete characters based on function and ancestry. In an evolutionary context, decomposition matters for building phylogenies and classifying taxa, which sort organisms across the Linnean hierarchy.

However, decompositional approaches are often contrasted with holist goals. Nuño de la Rosa (2010, 290) explains that Organicism – a holistic tradition regarding organisms defined as functionally-integrated and autonomous systems – has more ancient and historical roots than Darwinian theory. But at least two traditions can be distinguished for conceptually analyzing organismality.

On the one hand, under Darwinian traditions in the shadow of modern synthesis orthodoxy, “organisms are included in the more general category of biological individuals, defined as those entities (not only organisms but also genes or species) on which natural selection acts” (Nuño de la Rosa 2010, 290). Continuing the critique of organisms as mere vessels of adaptative characters: “organisms are conceived of as a non-problematic kind of individuals composing populations, and their distinct parts [their characters] are abstracted as adaptive traits that assure [an organism’s] reproductive success within specific environments” (290). In other words, organisms matter for more than their role as adaptation bearers, a role that atomizes and isolates parts as theoretically primary.

On the other hand, Nuño de la Rosa argues that in fact there are non-evolutionary morphological or physiological theories that prioritize organisms as integrated wholes through their developmental lifetime. By appeal to organicism’s longstanding history of varied views emphasizing connectedness and integration, she argues that strong theoretical grounds persist from Aristotle and Kant to the experimental embryology and developmental biology of the late nineteenth and twentieth century.

Sometimes the organism concept is used to synthesize intellectual traditions just discussed. For example, Reference HunemanHuneman (2017) offers a conception of organismality to support evo-devo traditions combining developmental and adaption-focused views. One maps onto epigenetic self-production of parts within a viable whole, and the other explains design of the whole by natural selection. In sum, organismality has been conceptually considered according to approaches that decompose organisms into their adaptive character traits, approaches that consider their developmental features as living cycles, and combined approaches.

3. Etymological Analysis

Finally, “[o]rganisms are so called because they are literally organized” (Reference Simpson, Roe and SimpsonSimpson 1958, 519). The term ‘organism’ has a long history. Etymology reveals ‘organic’ in reference to natural organization occurred around the late 1600s to early 1700s. The suffix ‘ism’ denotes a distinctive practice or system of some kind: organ-ism in its literal sense refers to a form of organization adapted for use in natural (i.e., non-artificial) contexts. Reference CheungCheung (2006, 319) traces first appearances of the term in the life sciences and its usage in different settings. In the later 1700s, ‘organism’ became an ordering principle and a “generic name for individuals as natural entities or living beings” (Reference Cheung2006, 319). However, living order as a mechanical product of an organism’s parts working together needs more historical context.

Historian Jessica Riskin explains that the ancient model of living machinery persisted through the medieval Scholastics. By the mid-1600s it was as familiar as “automata on clocks and organs in churches and cathedrals” (Reference Riskin2018, 159). When Descartes wrote the Treatise of Man in the 1630s, an anatomical treatise, he applied a different method from his predecessors in ancient and medieval anatomy (Reference Riskin2018, 144). Riskin states that the analogy about mechanistic clockwork,

And so, the intelligibility of organism function was realized in terms of its material parts – its anatomy, which was not Cartesian machinery in the pejorative sense.

Three vantage points, historical, conceptual, and etymological, were just considered as depicted in Figure 1 below.

Figure 1 Summary of organismality as an organizing concept

So, organismality’s epistemic value as an organizing principle can be summarized as follows.

First, organismality was a conceptual lens for understanding how living (versus nonliving) systems function. Organismality was considered in contrast to inorganic systems like crystals and compared with artificial systems and machines.

Second, organismality’s epistemic value is evidenced by its role in shaping biology’s early scientific status in a post-Newtonian era. Mechanical function and decomposition of machines informed and constrained analysis, even organicist critiques that challenged the mechanical–vitalism dichotomy. Mysterious qualities to explain organisms as living systems were rejected in both decompositional and organicist accounts.

Third, organismality served as both a primary and secondary analytic entry point: for prioritizing quasi-closed and autonomous systems as agents of change (as per Huxley) and as a mere vessel for continuity (as per Osler’s rendition of the eternal thread). However, there’s more to biological individuality than organisms alone.

1. Tracing the Motivations of S-A-I

Why argue that species are individuals? The S-A-I thesis was proposed in response to ancient, pre-Darwinian views that species are static categories of nature.

As Reference Ereshefsky and ZaltaEreshefsky (2022) points out, since Aristotle species have been the main examples of natural kinds (i.e., of natural categories independent of our classification schema) with essences. In pre-Darwinian contexts, species were created (by the gods or later, God), each endowed with essential characteristics – an essence – signaling species membership. Classifying species taxa occurred by shared qualitative characteristics, which were unique to a species and necessary in that all and only members of a species have them. However, even Linnaeus had difficulty determining a species’ essence, and evolutionary theory explains why: Forces like selection, mutation, recombination, and random drift can cause traits to disappear over time (Reference Ereshefsky and ZaltaEreshefsky 2022, 2001).

Early on, Reference HullHull (1965) explained negative effects of essentialism on taxonomy, what he called “two thousand years of stasis” in response to Ernst Mayr’s paradox. Mayr pointed out that while taxonomists accepted evolution, they still adhered to conceptualizing species as static entities. Reference HullHull (1965, 316) worked to unpack what he and others viewed as the problem’s crux. Essentialism’s residue was responsible for the conflict taxonomists faced. In particular, Mayr’s paradox was due to essentialist views of species as natural kinds or classes defined by shared essences precisely because evolutionary change precludes species taxa as static, unchanging entities.

Reference HullHull (1978) contrasted metaphysical notions of natural kinds and classes with individuals. Classes are groups of entities that can function in scientific laws, whereas individuals are historical entities that occupy particular space-time regions (Reference Hull1978, 337). Members of a certain class belong to that class because of the attributes they share. In modern contexts, most common examples to illustrate natural kinds and their essences draw from chemical kinds on the periodic table, for example, all instances of gold have the atomic number seventy-nine. Consider the following three features of classes.

First, classes serve nomothetic aims by providing a stable, reliable base for induction; laws generalize over features of classes; reliable inferences can be made about how members behave under certain conditions. For example, pure gold melts at 1948 degrees Fahrenheit in standard atmospheric pressure, pressure which is defined at sea level. The melting point of gold is reliably inferred not only by its chemical constitution but also by how that constitution behaves under specific conditions.

Second, classes are spatiotemporally unrestricted or “forever open” meaning that members can in principle re-appear at different times and places, whereas individuals are spatially and temporally located with beginnings and endings in time.

Third, members of classes share similar attributes and do not exist in part–whole relationships with other members of their class. Parts of an individual need not be similar, for example, an individual organism can be fragmented into structurally various parts that function together interdependently to sustain an entire organism (Reference RuseRuse 1987, 225).

So, if species are not classes, this implies: (1) It’s possible for inferences to fail. There is no guarantee for species behavior (i.e., genetically, morphologically, or behaviorally) in certain conditions. (2) The same species cannot go extinct and re-emerge later because species taxa are unique to specific times and places.Footnote ⁴ And (3) not all organisms in a species will necessarily share an essential “core” set of attributes.

Later, Reference RuseRuse (1987) argued that most philosophers discussing species as natural kinds were not in touch with biological reality. While Ruse raised objections against the S-A-I thesis, he clarifies motivations behind S-A-I. Typological views infused with static isolation and unchangeability dominated pre-Darwinian thought about species. We want to say that species are real, Ruse argues, but also that they can change. And so, one theoretical motivation driving S-A-I was its promise to designate species as tangible, concrete, and changing entities.

While Reference HullHull’s 1965 paper characterized the problem of species as natural kinds in light of evolution, that was one year after S-A-I’s initial formulation emerged from Reference GhiselinGhiselin (1966, 208–209) who proposed that biological species are “in the logical sense” individuals. He argued that to think otherwise is a category mistake. Species names are proper nouns and species must be individuals, metaphysically speaking, in order to evolve. Individuals, Reference GhiselinGhiselin (1987, 128) claimed, are single things, including compound objects made up of parts, while classes, as previously defined, are unrestricted to definite locations in space and time, so their names may designate any number of objects – including none at all. Hull (1976) took Ghiselin’s view further.

2. Hull on S-A-I: Interpretation & Clarification

With motivations in hand, it’s time to analyze Hull’s 1976, Reference Hull1978, and Reference Hull1980 papers. After some interpretations are proposed and criticized.

One important distinction from “Are species individuals?” (Reference Hull1976) concerns the species category versus species concepts. Species concepts organize species taxa according to some set of criteria, while the species category is defined in contrast to other Linnean classifications like genus, family, and order. Consider Figure 2 below.

Figure 2 The received view that species are a class of classes versus Hull’s development of Ghiselin’s view that species taxa are individuals

According to Hull (1976, 174) the old view of species defined the species category as a class of classes, that is, species taxa were classes or natural kinds. He did not contest the species category as a class but instead outlined how species taxa are individuals in the same way organisms are. He says, “the relation which an organ has to an organism is the same as the relation which an organism has to its species” (Reference Hull1976, 181). Evolution as a selection process requires relations among organisms in a species to be one of continuity: organisms reproduce themselves over time and as such are integrated into species, as historical entities, by descent. This why species taxa cannot be classes so long as classes remain ahistorical; class membership is unrestricted by history of time and place. For example, for something to be gold, it does not depend on its location along the American River in Coloma, California in the mid nineteenth century – gold membership is unrestricted by time and place, and thus discontinuous. Consider the following reconstruction of Reference HullHull’s 1978 argument:

1. 1. Evolution requires historical entities

2. 2. Classes are not historical (they are spatiotemporally unrestricted) –––––––––––––––––––

3. 3. Species cannot be classes

4. 4. Individuals are historical entities by definition–––––––––––––––––––

5. 5. Species are individuals

Species taxa, the basic units of classification, are the basic units of evolution – individuals in evolution – because they are historical, which made the search for qualitative similarity in essentialist approaches to classification a red herring (1978, 348). Species taxa are individuated from one another on the basis of continuous descent and cohesiveness, and not by similarity or essence (355–356).

While historical continuity is a necessary condition for individuality, maintaining a sufficient level of unity or cohesiveness is also required to form an individual unit. How Hull thought of unity at this juncture is complicated: while some entities have sufficient unity to compete with one another but not the open-ended organization to evolve, such as organisms, others are capable of open-ended change but might lack sufficient unity, such as higher taxa (184). While sufficient unity seems underspecified, intrinsic and external causes are identified as responsible factors (183–184). For example, if asexual forms specialize, become adapted, and shift their adaptations, then asexual species lack intrinsic mechanisms for promoting their evolutionary unity, and as such form units entirely by the unifying effects of external causes (183).

Regardless, individual units are the objects of processes, and so one might presume that genes are the unit of mutation, organisms the unit of natural selection, and species the unit of evolution (181). However, Hull acknowledges complex views of biologists: processes can occur across wide ranges of organizations. Mutations can consist of single nucleotide alternations or as the loss or gain of entire chromosomes. Plus, Reference LewontinLewontin (1970) argued that natural selection occurs across levels from macromolecules to populations. So, Hull clarifies in the later 1980 paper:

Hull argued that entities at various levels of organization (e.g., genes, cells, organisms, species, etc.) can function as units of selection, which implies that natural selection occurs at and across various levels.Footnote ⁵

Species are individuals for Hull, though, because they are units of evolution: to be units of evolution species must form lineages where natural selection causes those species to evolve (see Reference Ereshefsky and ZaltaEreshefsky 2022). In other words, species form lineages of “evolutionary unity” that can be individuated by spatiotemporal location and continuity (1978, 344). As historical entities, species form lineages just like genes and organisms form lineages, and they persist “while changing indefinitely through time” (341). Hull briefly distinguishes between units of selection and units of evolution, but he thought that both are individuals (Reference Hull1978, 338). In other words, species as units of selection was a contentious idea, but even if one denies species are units of selection, they are still spatially and temporally continuous, and as such are historical entities that evolve as a result of selection at lower levels.Footnote ⁶

So, the term ‘evolutionary individual’ can be ambiguous if we do not necessarily mean the objects of natural selection, but rather units of evolution. As we’ll see, contemporary usage of ‘evolutionary individual’ in the next section concerns individuals in natural selection. However, Hull’s evolutionary individuals (e.g., with species as the exemplar evolutionary units) are entities that become adapted or shift their adaptions, and not the bearers of adaptation like individuals in selection are. Whether species can be objects of selective processes – as both individuals of evolution and of selection – was the point of contention though.

Rejecting Alternative Interpretations. One reviewer noted that Hull’s 1976 paper is explicitly an ontology of the species category, and that it would be misguided to emphasize Hull’s earlier work as not focusing on the species category’s ontological status. To this point, I disagree. Hull did not explicitly focus on the species category, rather his focus is that species taxa are, ontologically speaking, not classes but individuals. Hull concedes (to Ernst Mayr in footnote 1 on p. 175) that the species category remains a class, however, he doesn’t provide a positive argument in that paper for why the species category remains a class. In his own words, Hull strategically accepts that his analysis does not affect the species category’s class status:

This means that the species category and species taxa are treated differently: Hull remarks that as a class, the species category can continue to be identified in the “usual,” that is, non-contested, way (175).Footnote ⁷ The usual, received view of the species category was that it is a class. This leaves open the possibility that, if life exists on other planets, the species category would apply there as well. In contrast, and given the historical nature of evolution, one could explore the species category as itself an individual too pertaining to the tree of life on Earth specifically (and not on other planets). However, opening that pandora’s box is a task for another day.

Additionally, that same reviewer claimed that Hull described three kinds of evolutionary individuals (in the 1976 and 1978 papers) – units of mutation, selection, and evolution – by using the method of theoretical individuation. By identifying the role these types of evolutionary individuals play in theory, namely, three theoretical roles, there are three kinds of evolutionary individuals. Furthermore, Reference HullHull (1980) then distinguishes units of selection into two types: replicators (i.e., entities that retain their structure largely intact through descent) and interactors (i.e., entities that cohesively interact with their environment in a way that has a unitary effect on constituent replicators).Footnote ⁸ So, they summarize, there are four types of evolutionary individuals for Hull: units of mutation, two units of selection, and units of evolution.

First, I find this characterization of how Hull views ontology and theory to be too simplistic, or at least it must be drawn out. It’s well known that in 1992, Hull grounded his evolutionary treatment of biological individuality on a theoretical basis (i.e., the theory of evolution by selection) in the absence of competing physiological or morphological theory.Footnote ⁹ However, Hull is explicit in 1980 that he focuses on characteristics of processes, and how entities perform with regard to those processes. He identifies “ontological status” as referring to the differences between class-inclusion, class-membership, and part–whole relations (Reference Hull1976, 181, footnote 6). Specifically, ontological status concerns the relations between entities as they pertain to those logical types. He says, “ontological status is theory-dependent” (in that same footnote) because he’s referring to evolution as requiring species taxa to be historical entities, which classes cannot be. What the theory of evolution determines is that species taxa cannot be classes, which leaves the part–whole relations of individuals as the better logical type.

Units of mutation, selection, and evolution across levels and as objects of processes are not merely manifestations of theoretical roles. Ontology – by way of logical relation types listed above – is theory-dependent for Hull because, as he says, nature does not come with logical types written on its face. Evolutionary processes are not theoretical posits, and so the objects of those processes are not either. Even when distinguishing two units of selection (i.e., replicators/interactors), Reference HullHull (1980) is consistent in that he starts with processes. In that regard, function is not merely a theoretical role that defines categories from normative expectations set by theory.

And second, what I take to be less important is the number of types (e.g., 3 or 4) to be distinguished from Hull’s work, at least for the purposes of what motivates S-A-I and why it matters. One significant motivation for Hull is that evolutionary theory demands species taxa be historical entities, rather than ahistorical natural kinds viewed as unhelpful pre-Darwinian residue. Hull was also engaged in consistent conversation with paleontologists like Simpson, Gould, and Eldredge over the status of species change in particular: whether species change indefinitely through time as merely the result of lower-level selection or whether species taxa form lineages, lineages which themselves evolve through macrolevel processes (e.g., see Reference Hull1980, 327). Hull did not endorse static types in the case of species taxa, and it is doubtful he would endorse its analog in the case of individuality.

In closing, S-A-I has normative impact later articulated by Hull concerning human species membership. Throughout history, many people were de-humanized as deviants from humanity. In response, Hull’s view implies one is human insofar as they are part of the human lineage, rather than satisfying some necessary (set of) features that all and only humans have. His work undermined the normative value of a “type-specimen” by taking polymorphism and polytypic representation seriously (Reference Hull1978, 351). The S-A-I thesis was not just theoretically significant, but socially conscious and non-exclusionary in ways that are sometimes overlooked.Footnote ¹⁰

Next, evolutionary individuality is considered in its contemporary sense before other types of non-evolutionary biological individualities. Thereafter, the resulting plurality is analyzed.

1. A Case Study & Some Background

Recall how Hull’s (1976) individuality view used organisms as paradigm examples of individuality with integrated organization sustaining their unity. However, does integrative unity require parts to be in the same vicinity? Candidate criteria for organized unity include physiological integration (i.e., working together), spatial contiguity (i.e., being within proximity or in contact), and autonomy (i.e., sustained independence from external environment). However, cases lacking one or more of these attributes undercut these criteria (Reference SantelicesSantelices 1999, 152). Consider the following.

The Case of the “Humongous Fungus.” In 1992, Smith et al. published an article in Nature claiming the largest and oldest living organisms were in the Armillaria genus. Armillaria bulbosa was identified as an individual occupying at least 15 hectares, weighing in excess of 10,000 kg, and retaining genetic stability for more than 1,500 years. Armillaria gallica extended up to 37 hectares of forest floor in Michigan’s Upper Peninsula, but was older than the original estimates, at least 2,500 years old (see Reference Anderson, Bruhn and KasimerAnderson et al. 2018). Because asexually reproducing organisms occur across kingdoms in a variety of taxa, distinguishing asexually produced genotypes was essential for understanding their population biology to define their clonal structure. It (they? – that’s the question) consists of nearly genetically identical fungi clusters often separated by trees, and in some cases entire forests. The clusters encompass tree root systems and exhibit stability of somatic mutations, which reflect historical growth patterns from a single point (2018). Why is this case a puzzle for individuality?Footnote ¹¹

In a 1977 landmark paper, evolutionary biologist Daniel Janzen argued that pre-theoretical intuitions misguide scientific work. Clonal cases – like the fungus above or dandelions, aspen groves, strawberries, and so on – are physiologically distinct in terms of their location (i.e., the mushroom that can be picked, the dandelion for our wishes, the aspen tree cut down). Barring intra-clonal variation, Janzen argued that physiological unity is insufficient, and in some cases not even necessary, to distinguish individuals in natural selection. Genetic and phenotypic similarity exhibited by clones means they do not have varying traits populations of individuals require to evolve. In other words, they are not unique from one another from selection’s point of view.Footnote ¹²

Janzen’s view motivates our fungi case: Armillaria in the upper peninsula of Michigan is one very large, spatially disparate evolutionary individual. It (no longer plural “they”) defies pre-theoretical expectations due to its age as older than Christianity, its weight at 400 tons, and its size as larger than roughly seven Yankee stadiums put together (Reference PennisiPennisi 2018). Janzen’s paper from 1977 mattered for challenging intuitions about individuality. He emphasized genetic identity and rejected the need for physiological unity: parts of an individual in selection need not be physically near and/or touching one another (i.e., spatially contiguous/continuous) to consider them as parts of that unit.

Evolutionary individuality, then, is where that theoretical machinery is put to use. Since some biological individuals are objects of natural selection, what’s known about that process must be revisited. Reference LewontinLewontin’s 1970 recipe is the received starting point.

According to Lewontin, three conditions must hold for evolution by selection. First individuals, the discrete units, must exhibit phenotypic variation. Second, those varying phenotypes must have corresponding rates of survival and reproduction, which indicate how varying traits make a fitness difference for the individual. Third, that individual’s fitness, their ability to survive and reproduce, is based on the heredity of those traits; traits must be transmitted to offspring. And so, individuals in selection – evolutionary individuals – exhibit varying heritable traits that make a difference to their fitness. As Reference LewontinLewontin (1970, 7) puts it: “the primary focus of evolution by natural selection is the individual.”Footnote ¹³ While the case above considered a clonal organism, as we’ll see, not all evolutionary individuals are organismal.

2. Darwinian Individuals, Reproduction, & Heredity

Reference HullHull (1980) and Reference Godfrey-SmithGodfrey-Smith (2009, Reference Godfrey-Smith, Bouchard and Huneman2013, Reference Godfrey-Smith2015) build from Lewontin who did not analyze reproduction specifically. While Hull distinguished two types of individuals in selection (i.e., replicators and interactors), he focused on evolutionary theory’s demand for historical entities. This included more than individuals in selection alone (i.e., units of mutation, of evolution). In contrast, Godfrey-Smith takes a narrower focus on individuals in selection, specifically the nature of reproduction as a mechanism for heredity.Footnote ¹⁴ Material overlap is too restrictive, and formal accounts of reproduction work well enough for Darwinian processes (2009, 83). Let’s unpack this.

Reference Godfrey-SmithGodfrey-Smith (2009) draws from his view of Darwinian populations: Darwinian populations are evolving populations in which novel variations arise. Understanding reproduction as a way to satisfy heredity is an aim;Footnote ¹⁵ how individuals reproduce in those populations matter. ‘Darwinian Individuals’Footnote ¹⁶ (his term for individuals in selection) need not be organismal: “genes, chromosomes, and other fragments of organisms can all form Darwinian populations” (2009, 85). Sometimes organisms don’t meet evolutionary criteria, such as sterile animals who cannot produce fertile offspring. These organisms can metabolize but are unable to reproduce. They resist forces of decay, but cannot be individuals comprising Darwinian populations (2013, 25).

There are different forms or modes of reproduction. Simple reproducers, such as cells, make more of themselves by machinery internal to them. They often reproduce on their own in environmental contexts allowing for nutrients and energy to do that work. Scaffolding reproducers, like viruses and chromosomes, reproduce by means of structures or mechanisms external to them (2009, 88–89).

One mode receives special attention: collective reproducers have simple reproducers as parts, for example, human sexual reproduction and gamete fusion, but there can be collectives of collectives, such as eusocial insects containing multicellular organisms (2009). As a result of de-Darwinization, collectives form when evolution of lower-level entities is suppressed by evolution occurring at higher levels of organization. However, evolutionary activity at lower levels can disrupt stability of a collective unit, such as the rapid generation of cancer cells – simple reproducers – within a eukaryotic organism. Evolutionary processes at lower levels become re-Darwinized sometimes at the host’s mortal expense.

Three parameters are identified that together set a gradient for collective reproducers: paradigm-minimal-marginal Darwinian individuality. How biological entities score on the following three criteria determines their place on that gradient.

First, Darwinian individuals exhibit degrees of physiological integration with the mutual dependence of parts serving different functions (2009, 93).

Second, an individual’s parts exhibit a special division of labor, reproductive specialization, such as germline cells responsible for the capabilities of an organism to produce a new organism.

Third, there must be some mechanism by which the production of something new (i.e., re-production) is distinguished from growth of the same. Reproductive bottlenecks mark generational divides where genetic variation is reduced. In other words, reproductive bottlenecks force the process of growth and development to start anew (2009, 91). Consider how some organisms begin as one-celled zygotes that flourish to many cells: that narrowing is a “bottle neck” shape depicted in Figure 3 below.

Figure 3 From many to few and back and again. New token individuals distinguished by a narrowing reproductive bottleneck

What do puzzle cases look like on this view? Phenomenally individuated units of plants and fungi have narrowing runners that result in new clonal fragments. Reference Godfrey-SmithGodfrey-Smith (2009, 92) says, “considering again the cases with ramets and runners: the thinner the runner – especially in relation to what is to come – the less the new structure is a mere continuation of the old.” This is a nonbinary view: “thinner” indicates a gradation from growth to the reproduction of something new.Footnote ¹⁷

Something “new” is the reproduction of offspring, which matters if fitness is measured according to numbers of viable offspring.Footnote ¹⁸ Tracking parent–offspring relationships – lineages formed – is difficult in cases involving symbiotic relationships; associations with two or more partners from different species. Symbiotic associations vary in their nature, for example, how mutually beneficial the relationship is to all parties, whether there’s continuous physical integration or containment, and so forth.

On the one hand, Godfrey-Smith’s view accommodates vertical transmission. For example, aphid-Buchnera associations include bacteria that are transmitted vertically through a lineage running in tandem with that of the aphid (Reference Godfrey-Smith, Bouchard and Huneman2013, 31). “Vertical” transmission concerns reproductive lineages of all symbiont partners forming patterns of parent–offspring relationships arranged together – the Buchnera are maternally passed to aphid offspring.

On the other hand, there’s dispute over cases of symbiotic associations recurring horizontally, for example, by uptake of partners from the environment. For example, while Hawaiian Bobtail Squid and its Vibrio bacteria form a consortium, a potential adaptive unit, there is a many-many parent–offspring relationship between symbionts creating complicated parent–offspring transmission networks.

So, microbial symbionts can be acquired vertically from host–parent to host–offspring (e.g., like aphid-Buchnera), as well as horizontally from other host organisms and the environment during development (e.g., like squid-Vibrio). Reference Godfrey-SmithGodfrey-Smith (2015, 10123) argues that while the latter cases of horizontal transmission result in multispecies metabolic collectives recurring through actions of several Darwinian individuals, these are collaborations that coevolve, but do not combine into single reproducing objects. That is, coevolutionary theory accounts for those cases.

Godfrey-Smith’s account of reproduction illustrates formal characteristics of lineage formers and their degrees of “tightness” or obligation to one another in parent–offspring networks. His concept of reproduction is satisfied “as long as we know who came from whom, and roughly where one begins and another ends” (Reference Godfrey-Smith2009, 83). In that sense, passing on material structure is unnecessary (or too restrictive) because there are different modes of trait transmission satisfying heredity as summarized in Figure 4 above.

Figure 4 Material overlap is unnecessary for reproduction. Different forms of reproduction satisfy heredity, that is, there are different reproducers – Darwinian individuals – like simple, scaffolding, and collective, which each scale along a marginal–minimal–paradigm gradient of individuality. Parameters are set for collective reproducers: reproductive bottlenecks, reproductive specialization (germ/soma), and integration of specialized parts. Sometimes material overlap is insufficient too.

In some cases, like the squid-Vibrio example, material overlap is insufficient for forming reproductive objects (i.e., Darwinian individuals) tracked to determine a population’s response to fitness differences.

The complicated parent–offspring networks of horizontal transfer cases confound fitness measures of reproductive success; viable offspring need to be distinguished from parents (i.e., re-produced). However, one criticism is that “the epistemological problems associated with tracking lineages of holobiont parts should not lead to confusion about the fact that such patterns could be determined in principle” (Reference BoothBooth 2014, 667). In other words, there may be relevant ancestry, even if traditional concepts of parenthood are challenged by complex (and reticulated) lineage patterns.

And while Godfrey-Smith argues that degrees of tightness matter for heredity in multispecies units, others argue that inheritance can occur by recruitment. Reference Ereshefsky and PedrosoEreshefsky and Pedroso (2013, Reference Ereshefsky and Pedroso2015) considered biofilms candidate evolutionary individuals.Footnote ¹⁹ Biofilms are singular or multispecies communities of microorganisms; viscous collections existing in containers of water, streams, and rivers, and on living and nonliving surfaces. Cells exist together within an extracellular polymeric matrix, which facilitates both chemical signaling (i.e., quorum sensing) and lateral exchange of genetic material (i.e., by transformation or uptake of exogenous DNA, and by conjugation through “bridges” via plasmid DNA transfer).

Reference DoolittleDoolittle (2013, 372) said that distinguishing inheritance and recruitment matters for determining microbial communities like biofilms as bearers of adaptations, and thus as individuals in selection.Footnote ²⁰ However, I specify recruitment as an inheritance mechanism. There are at least two different processes through which traits are transmitted, such as (1) via recruitment, which I take to be the horizontally convergent nodes within webs of parent–offspring lineages, and (2) the modes of reproduction that entail a vertical “tightness” of parent–offspring lineages as identified by Godfrey-Smith. However, while some satellite cells contain genetic material from the parent biofilm, it’s debatable whether adaptive traits (e.g., like antibiotic resistance) exist at the level of the entire biofilm and are passed on.

Reference 85ClarkeClarke (2016) objected that biofilms don’t have the characteristics of lineage formers: they lack a unitary character because of aggregation, which means (1) they neither reliably pass on their structure to offspring biofilms (i.e., not sufficiently resembling them) nor (2) remain bearers of adaptations due to lack of stability. The frequent shuffling of genetic material makes it unlikely that “offspring” biofilms have traits resembling the parent biofilms.

In response, Reference PedrosoPedroso (2017) maintained that overall function and corresponding phenotypes remain across successive generations. Aggregation yielding increased genetic variability through horizontal transfer does not mean that phenotypes themselves are unstable (e.g., such as antibiotic resistance, signaling patterns, potential for virulence, motility, etc.). Furthermore, biofilm recruitment is not an open-ended free-for-all. There are patterns of recruitment based on signaling – co-aggregation is a genetically controlled mechanism (129–130). Parent–offspring lineages are products of ecological succession. Their bottleneck patterns are formed by ecological surroundings, but that doesn’t mean varying, inherited traits are absent (2017, 131).

At this juncture, Reference Doolittle and BoothDoolittle and Booth’s (2017) patterns of lineage formation are relevant. Functional patterns (e.g., metabolic and developmental interaction patterns), rather than the material (i.e., the taxa) responsible for them, are the units of selection. Material transfer is not needed for lineages formed by sequences of interaction patterns. While Godfrey-Smith emphasized modes of trait transmission, Doolittle and Booth focused on abstract functional relationships lineages engage with and present as their patterns through time.Footnote ²¹ Instead of reproduction, it becomes reconstruction – a formal kind of reproduction with no crucial piece of matter or material to be made (Reference Doolittle and BoothDoolittle and Booth 2017, 16, also see Reference Doolittle and InkpenDoolittle and Inkpen 2018, 4007 on “re-production” meaning “created again”). While abandoning materiality, the abandonment of causality is denied. Patterns themselves are causal in the reconstruction of a new evolutionary individual.Footnote ²²

3. Inheritance Satisfied Functionally vs. Its Material Conditions

It is now time to draw a conclusion. Formal accounts of heredity emphasize functional profiles and patterns of inheritance. But there’s varying attention to material conditions of trait transmission. Reference Doolittle and BoothDoolittle and Booth (2017, 6) say, “put metaphorically, what matters is the song, not the singer. The song, to flesh out the metaphor, is the pattern of interactions (metabolic, structural, or developmental) between partner lineages (the singers).” They argue that instances of interaction patterns pass on traits to later instances, leading to the differential persistence of the overall pattern’s type. Those patterns, rather than material overlap and constitution, are sufficient to fulfill the criteria for evolutionary individuality. That is, heredity is fulfilled formally, and without appeal to inheritance’s material conditions for making more individuals.

That “more-making” capacity, as Reference 87GriesemerGriesemer (2016, 807) puts it, has constraints though: “material overlap means that reproduction involves bonds of material continuity, not merely resemblance or formal information transmission.” Reference Griesemer, Jones and CartwrightGriesemer (2005) previously argued that both information copying and formal relations are “problematic as stand-alone concepts of inheritance in abstraction from the material conditions of reproduction. The latter, not the former, determine the causal pathways of heredity relations” because the flow of genetic information depends on material connections between senders and receivers. Moreover, inheritance is a process where evolved mechanisms of development are “propagated in reproduction,” which must include both epigenetic and nongenetic mechanisms (2016, 807). Development is important because it is the “recursive acquisition, refinement, or maintenance of the capacity to reproduce” (2016). The driving point is to avoid considering evolutionary individuality in isolation from materials of developmental systems; it must be recast in research spaces of evo-devo, eco-devo, and developmental biology. There is too much focus on formal or overall structural patterns of inheritance, a critique best understood by the following example.

In a recent publication titled, “The Information Theory of Individuality” (Reference Krakauer, Bertschinger and OlbrichKrakauer et al. 2020) the authors formalize how information is transmitted through an information-theoretic lens, while drawing from both environmental dependence and inheritance. Adaptive aggregations can be multi-scale without physical boundaries like cell walls or tissue, yet still visible to selection. They are evolutionary individuals in the sense discussed presently.

If information is only meaningful in the context of material systems, though, then the channel through which information is passed – the material medium – should not be overlooked. That is Griesemer’s point.Footnote ²³ Otherwise, what story is there about reproduction, if there is no medium between the sender and the receiver? Reference Krakauer, Bertschinger and OlbrichKrakauer et al. (2020, 210) construe individuality without relying on material boundaries. They aim to capture fluidity and porousness of aggregates and associations, only the propagating forward of information though time is needed (214). Quantity, closure, and autonomy are defined by formal boundaries of functionally individuated systems emerging at different scales (220). Not required on their view is the kind of physical contact, overlap, and raw materials involved in causal pathways of heredity relationships.

The key takeaway is that evolutionary individuality facilitated conceptual expansions of reproduction into more abstract or formal (versus material) senses. Reference Godfrey-SmithGodfrey-Smith (2009, 84) argued that nothing about inner logics of Darwinism preclude a purely formal account. Others proposed inheritance by recruitment; reconstruction of new offspring individuals distinguished by ecological bottlenecks, which comprise parent–offspring lineages. The functional shape of reproductive patterns was taken further; transmission patterns – their information – transcend physical boundaries yet cause variation remaining visible in in its form to selection (see Figure 5).

Figure 5 Schematic of debate over Lewontin’s heredity condition: formal modes and material conditions of reproduction. On the left, reproduction is formally analyzed by abstract functional relationships lineage-formers engage in. On the right, there is dissent from a material standpoint targeting abstract, formal accounts. Insofar as evolutionary individuals must pass on varying, inherited traits, how that occurs is analyzed both formally and materially

However, pre-occupations with purely formal aspects abstract from reproduction’s material conditions, which risks occluding matter – the medium – that both determines the causal pathways of heredity relations and the effects of development on reproductive systems (Reference 87GriesemerGriesemer 2016, 201). Reproduction’s material processes intertwining both development and recursively (i.e., in the form of complex life cycles) generated propagules must not be overlooked.

Relevant material conditions could be further analyzed by drawing from Lynn Margulis’ (e.g., 1998) work. Margulis argued that symbiogenesis – the merging and diverging of associations – drives evolutionary change. For Margulis, cell structure, biochemistry, and geological context are material conditions that matter for increasingly complex levels of individuality. New adaptations arise not only from random mutations but also the merging of two separate organisms in those conditions (see Reference GoldscheiderGoldscheider 2009, 44). This is a task for later work.

To summarize, reproduction as a heredity mechanism for individuals in selection was analyzed according to both formal and material accounts. Next, individuals in immunology, ecology, and metabolism are considered.

1. Immunological Individuality

In a 1992 chapter, David Hull stated that if physiology were theoretically developed enough, it could be used to determine individuality in biology due to its emphasis on mechanics and function of multi-part systems and wholes. Reference PradeuPradeu (2010, Reference Pradeu2012) takes up this challenge. Functionally integrated parts are pervasive in literature on biological individuality, even of the evolutionary sort: Godfrey-Smith’s Darwinian individuals must be integrated and exhibit relevant divisions of labor among parts that support their mutual dependence. Reference HullHull (1978) argued that functional integration is a necessary feature. Providing specificity to what “integration” means, Pradeu draws from physiology, in particular the field of immunology.Footnote ²⁴

Roughly, immunology is the study of how bodily systems fend off threats, but how to define immune response is contentious. For Pradeu, immunogenicity, or the triggering of immune response, occurs in the presence of strong discontinuous molecular difference rather than exogenecity specifically, that is, rather than a foreign “non-self” source (137). Reference PradeuPradeu (2012, 143–144) specifies that immune response does not concern just any discontinuity, but a strong discontinuity of ligands with which immune cells interact. Relevant factors of discontinuity include features of antigens or substances that induce immune response, such as antigen quantities, speed of antigen appearance, degree of molecular difference, and regularity of antigen presentation.

A referee requested that because molecular interactions are foundational to Pradeu’s view, a note on genetics should be provided. However, it is imperative not to conflate genetics-based analyses with physiological study of molecules and receptors in, on, and around the surfaces of (immune) cells.Footnote ²⁵ Genetics is not all there is to molecular activity – cellular and molecular physiology concern biochemical interactions and dynamics. Insofar as genetics concerns the study of inherited differences in evolutionary contexts, a main framing of Pradeu’s book is Hull’s challenge to provide a well-developed physiological, rather than evolutionary, theory of individuality and identity in biology. Moreover, Pradeu distinguishes between the continuity theory at the genetic level versus its application concerning molecular patterns recognized by receptors mostly located on the surface of immune cells (see Reference Pradeu2012, 178).

In contrast to previous discussions of evolutionary individuality, Pradeu’s work highlights maintenance conditions, rather than reproductive conditions for new token individuals. For example, if cancer cases constitute breakdowns of cohesive functioning with parts proliferating at the expense of the whole, then Pradeu’s immunological account explains when and how individuality fails: one-way individuality breaks down is through poor boundary maintenance and control over parts. With new technologies in cancer therapeutics drawing from immune response research, close attention ought to be paid to immunological individuality.

Considering an evolutionary context Pradeu centers organismality; “it is necessary to examine the physiological processes produced in the organism to arrive at a precise definition of what, in each case, counts as an evolutionary individual” (Reference Pradeu2012, 260). While organisms are not always individuals in selection, starting with the heterogeneous organism helps to determine what evolutionary individuals are.Footnote ²⁶ An immunological account identifies organisms as the most well-defined individuals, rather than merely one individual among other types (Reference Pradeu2012, 264).

For a historical take, consider Reference MedawarMedawar (1957), a zoologist and comparative anatomist, who investigated the uniqueness of individuality. He explored issues with surgical skin grafting and transplantation, that is, why intolerance occurs when borrowing from members of one’s own species in humans. Medawar also considered immunological reactions (e.g., to bacteria and viruses, but also allergens). He disagreed with philosophers that the distinction among individuals is of a difference in kind or even of degree (Reference Medawar1957, 154). Instead, he proposed that difference among individuals amounts to something combinatorial: one individual differs from others not because of unique endowments, but because of the unique combination of endowments (Reference Medawar1957). Medawar investigated how combinatorial factors of one individual are retained, for example, what the antigens are and what keeps the immunological reaction maintained. Indeed, immunological individuality is intertwined with medical consequences.Footnote ²⁷

2. Ecological Individuality

Are communities in ecology mere assemblages? Or individuals in their own right? Reference LeanLean (2018, 520) argued that ecological communities are often not biological individuals because they lack causal boundaries – individuality is not the natural end point of all biological interactions.Footnote ²⁸ Others maintain that ecological communities are individuals in contrast to arbitrary sets of things (e.g., see Reference HunemanHuneman 2014a and Reference 88Huneman2014b). Complex interactions within communities are “stronger” than interactions between communities and the external environment. Huneman explores different theoretically driven definitions of “strength” because “within a theoretical domain [they allow] us to partition the assemblies into “individuals” and “non-individuals” (2014, 361). Ecological individuality is a continuum: strong and weak individuals are distinguished by the connections among their constituents. Interactions exist within a set of formally defined parameters in terms of how likely they are to occur from more intimate interactions (e.g., aphid-Buchnera consortia) to interactions with lower intimacy (2014, 370). Engineering ecosystem interactions involve several organisms from many species over a significant timescale. When considering ecosystems, the living or “biotic” criterion appears less stringent.

Ecosystems include organismal interactions with abiotic components like the surrounding landscape (e.g., soil, water, and so forth). Reference Millstein, Chen, Bueno and FaganMillstein (2018, 281) explores the concept of a land community as articulated by Aldo Leopold, a twentieth-century forester who, as Millstein argues, “did seem to think that the land community was an individual”: he didn’t use the word ‘individual’ specifically, though he did explore whether the land community is an organism.Footnote ²⁹ Since ‘organism’ and ‘individual’ were often used synonymously at the time, Leopold was likely probing a land community’s individuality status by exploring its organism-potential. Leopold spoke of interdependence among “soils, waters, plants, and animals collectively” and considered both the organization (and sometimes disorganization) of the land (2018, 281). What’s at stake for the land community is its status for moral considerability, that is, whether it can be an object of moral obligation and have intrinsic value. The moral motivation canvased by Millstein is a normative spin on individuality in the biological domain. However, the individuality status of land communities is difficult to establish.

Distinguishing boundaries is complex because open systems, that is, “systems where the spatial area of the densely interacting populations is larger than that of the dense matter/energy flow – or vice versa,” challenge traditional conceptions of boundaries as strictly physical or material barriers (2018, 291). While sometimes the spatial contiguity and continuity of an individual’s parts determine boundaries, in the land community case Millstein considers (and rejects) congruence as a requirement: parts need not be the same, nor need they always “agree” or be compatible in way that is strictly cooperative. As she puts it: “Individuality does not require location in the same space”: parts can be spatially disparate. Systems can be well-bounded or open, but still satisfy the metaphysical constraints as identified by Reference GhiselinGhiselin (1974) and Hull (1976), such as restricted to time and place, integrated parts such that the mutual dependence of causal interactions affects their shared fate, beginnings and endings in time, and continuity through time (Reference Millstein, Chen, Bueno and FaganMillstein 2018, 297). The problem, as Millstein (298) identifies it, is distinguishing an individual from “an abstract type or a mere set or a mere assemblage,” of which “organismality, internal regulation, being a unit of selection, and/or emergent properties” are not necessary for it.

Millstein points to features like internal regulation that could make an individual more robust but would not be necessary. While “robustness” carries connotations of resilience, in the context of ecological individuality, resilience against what? One might consider what defines individuality within a particular domain, such as an ecological individual’s ability to persist under various perturbations like pollution, poaching, and invasive species. In that sense, resilience is normatively cast as ecological health. Developing a concept of health according to both ecological and immunological individuality indicates the need to coordinate those domains. If different domains interact in important ways, overlap should be expected:

Thus, resilience produces robust individuality, which matters as a normative feature, one that arises in both immunological and ecological contexts concerning resilience against disease, poor function, and breakdown. Addressing heterogeneity and boundary conditions certainly matters, but healthy functioning in these contexts, however defined, reveals normative concerns about how ecological and immunological individualities should be.

In other words, while Millstein posits moral considerability as one normative element at stake, healthiness of ecological and immunological individuals is another. This is one angle for approaching the ecology of holobionts: microbes and their host-organism relationships occur in networks of interactions and dependence. Reference Gilbert, Sapp and TauberGilbert and Tauber (2012) argue that eco-immunology provides evidence for a holobiont’s individuality, which includes surveillance and response immune mechanisms. Those mechanisms play a critical role in regulating healthy social ecologies of holobionts, while acknowledging microbial activity in the immune system itself (2016, 846–847). Immunology and ecology are not all there is though. Threads of life and health arise in another context for biological individuality.

3. Metabolic Individuality

What are metabolic individuals? They might just be organisms, at least for Reference Godfrey-SmithGodfrey-Smith (2009) who distinguished between metabolic organisms and reproducing Darwinian individuals. Some organisms cannot reproduce (e.g., sterile mules or castes of insects). If organisms are defined immunologically or even ecologically, why bother discussing metabolism within the context of biological individuality at all? While Reference Dupré and O’MalleyDupré and O’Malley (2009) bring metabolism and replication/reproduction into the context of biological individuality, the issue is framed within the context of life.

Life is a complex concept with debate over necessity of metabolic function to consider something alive. For example, viruses may replicate themselves under the right conditions, but are metabolically inert and so are not considered alive (unless we focus on replication as a sole criterion for life). If one assumes that biological individuals must at minimum be alive (though the abiotic components of land communities considered previously might suggest otherwise), then how life is defined impacts what’s included under biological individuality’s scope. However, Reference Godfrey-SmithGodfrey-Smith (2016) points out that life has two sides: a metabolic side and another side concerning reproduction and evolution. To ask whether one is primary or sufficient is a misguided question. And so, we might consider viruses as biological individuals of a particular sort, even though they do not really have their own metabolism: they form Darwinian populations as they engage external machinery to make more copies of themselves (see Godfrey-Smith on scaffolding reproducers).

But let’s stay focused on biological individuality and metabolism. Metabolism amounts to conversion: the chemical processes and reactions involved in sustaining energy production, and the breakdown and synthesis of molecules and compounds needed to resist forces of decay. If considering individuality at a metabolic scale, then biochemical composition must be measured and tracked.

Reference Beebe and KennedyBeebe and Kennedy (2016) describe technology to comprehensively measure and track an individual’s metabolic profile, that is, metabolomics. Contrast metabolomics with genomics; “metabolomics” refers to the metabolic profile (i.e., the character or quality of an organism’s metabolic activity), rather than genomic profile, with the purpose of refining therapeutic interventions in precision medicine. Beebe and Kennedy offer a means of drafting blueprints – technical maps of molecular underpinnings of human individuals beyond genetics.Footnote ³⁰ Philosophers might recognize blueprinting as drafting the identity of metabolic individuality: what makes one individual unique from another in a biochemical sense.

However, organic individuation through the lens of metabolic reactions has been around at least for a while, such as C. M. Child’s Individuality in Organisms (Reference Child1915) where he develops a dynamic conception of individuality such that maintenance is possible in a changing environment. To be clear, Child’s solution takes a physio-morphological spin as he wrestles with concepts of unity and order. He searches for physiological evidence of metabolic gradients: The organic individual is fundamentally a dynamic relation of dominance and subordination, associated with and resulting from the establishment of a metabolic gradient or gradients (Reference Child1915, 88).

An organism’s “axis” (e.g., analogous to Earth’s axis – the imaginary pole running from “top” to “bottom”) represents direction of a gradient from higher to lower rates of metabolic reactions, such that, if I’m reading Child correctly, those rates decrease in the direction of that axis (1915, 20). Child was dedicated to finding evidence for dynamic boundaries of metabolic individuality.

Taken together, individuality in immunology, ecology, and metabolomics (metabology?) is a varied and interconnected landscape relevant to life and health. While much philosophical attention has directed toward individuals in evolutionary biology, these are solid foundations for branching to other non-evolutionary biological sciences.Footnote ³¹

Up to this point, biological individuality was approached from multiple vantage points including evolutionary individuality (in broader and more specific senses concerning individuals in selection), and different types of non-evolutionary biological individuals. Next, some conclusions are drawn before turning toward critical discussion of biological individuality in the production of scientific knowledge.

1. A Critique: Heterogeneity & Pluralism

First, recall how Reference Godfrey-SmithGodfrey-Smith (2009) argued that individuals in selection reproduce by yielding reproductive bottlenecks, exhibiting reproductive division of labor, and behaving as functionally integrated units. He recognizes degrees of individuality – entities score high on these parameters as paradigm individuals versus marginal individuals that score low on all three. Reference BoothBooth (2014, 671) views the account as pluralistic because Godfrey-Smith identifies two types of biological entities – metabolic organisms and Darwinian individuals. This is a plurality of biological individuals, but not of individuals in selection specifically. Godfrey-Smith’s definition of individuals in natural selection is constrained by a single set of parameters where entities fall along a gradient. Different entities, for example, viruses and organisms, can become or cease to be Darwinian in the evolutionary sense.

Second, recall Reference HunemanHuneman’s (2014a, b) proposal that strength of interaction marks boundaries of individuals. Evolutionary individuals exhibit strongest levels of interaction delineating interactions of entities with one another versus with entities outside of that unit (Reference 88Huneman2014b, 377). Increasing levels of interaction produce individuality pluralism because nested individualities comprise levels of biological hierarchy (374). Different entities across scales of interaction, in an ecological sense, also become or cease to be evolutionary individuals depending on the strength of interaction among them. In other words, evolutionary individuality as one type of biological individuality is distinguished from ecological individuality (i.e., another type of biological individuality) by way of interaction strength.

To clarify, Godfrey-Smith and Huneman use ‘pluralism’ to describe their views concerning many types of biological individuals. There are at least three ways to consider pluralism and biological individuality:

1. (1) Many types of biological individuals that are roughly domain-specific (e.g., evolutionary, metabolic, ecological).

2. (2) Many types of evolutionary individuals, of ecological individuals, of metabolic individuals, and so on (i.e., each type has a plurality of subtypes).

3. (3) Both (1) and (2).

In (1), biological individuality is a heterogenous category of all types of biological individuals. Those types are defined by different concepts, concepts which range over complexes of processes and patterns relevant for (and across) subdisciplines of the life sciences. Several classificatory concepts characterizing biological individuality differently (evolutionary, ecological, immunological, and so on) were previously discussed.

However, when unpacking the prospect of (2) (and thereby (3)), consider how several classificatory concepts characterize, for example, evolutionary individuality differently.Footnote ³² That is, concepts that organize the class or category of all individuals in selection, what I call the ‘EI category’.Footnote ³³ Let me explain.

Taking Lewontin’s recipe to define the class of all individuals in selection, our EI category, then at minimum all characterizations of EIs must exhibit inherited variation in fitness. However, there are several views on how individuals satisfy Lewontin’s heredity constraint: through formal senses of reproduction to transmit traits (Reference Godfrey-SmithGodfrey-Smith 2009, Reference Godfrey-Smith, Bouchard and Huneman2013, Reference Godfrey-Smith2015), of reconstruction including non-vertical forms of transmission of adaptive traits (Reference Ereshefsky and PedrosoEreshefsky and Pedroso 2013, Reference Ereshefsky and Pedroso2015; Reference 85ClarkeClarke 2016; Reference PedrosoPedroso 2017; Reference McConwellMcConwell 2017a, b), and of transmitted interactional patterns in a functional sense with differential persistence (Reference Doolittle and BoothDoolittle and Booth 2017). If heredity is satisfied in different ways under Lewontin’s view (in the sense of Reference McConwellMcConwell 2017a, b), and each way generates a different EI concept, then candidates for the EI category are organized according to how heredity is characterized.Footnote ³⁴ Consider Figure 6.

Figure 6 EI category vs. EI concepts. The EI category is defined by Lewontin’s view of evolution by selection. EI concepts organize candidates for evolutionary individuality in different ways. To be a pluralist about EI concepts is to reject that there is one single correct way to define the EI category

The conceptual typology of EI is organized further according to functional and material approaches. For example, if the typology is defined functionally, then heredity for Lewontin might just be one constraint on the functional or theoretical role of evolutionary individuals realized multiple ways. That aligns well with Reference ClarkeClarke’s 2013 view: evolutionary individuality is a theoretical role realized by varieties of (reproductive) mechanisms. However, if the mechanical details of inheritance matter for pluralism, a plurality of evolutionary individuality concepts can be distinguished on the material basis of how trait transmission occurs. This pluralism is all about perspective.

There would be dispute over the EI category, the class of all evolutionary individuality types, if there were dispute over how evolution by selection works. So far, there seem to be variations consistent with the original Lewontin formulation; slightly different accounts of the “recipe,” such as Reference OkashaOkasha (2006), Reference BrandonBrandon (1999), and Reference BourattBouratt (2015). Hull’s replicator-interactor account may be just an extension of the original Lewontin formulation. The Price Equation might be as well. Reference OkashaOkasha (2006) explores the link between Price’s equation and Lewontin’s account of evolution by selection. Price’s equation tracks combined effects of two or more levels of selection on evolutionary change (Reference Okasha2006, 18ff). The point is that a different recipe would characterize the EI category differently and as such shift the basis of present discussion.

While a synchronic typology of biological individuality is ontically oriented, disciplinary dimensions of Table 1 further muddy the picture. I return to this in a moment. For now, disambiguating pluralism(s) in part depends on what sort of pluralism one has in mind.

2. A Positive Proposal: Synchronic & Diachronic Plurality in Life’s History

How do new individuality types emerge over time? I propose shifting toward biological individuality’s broader context: the history of life. The following considers how new types of individuality occur over time, a diachronic view.

There is a strong connection between individuality and major transitions in evolution. Major transitions mark pivotal turning points in life’s history (Maynard-Reference Smith and SzathmarySmith and Szathmary 1997). Some evolutionary events are deemed major, like transitions from unicelled to multicelled life, because of momentous shifts in organization. Life’s history and hierarchy contextualize individuality pluralism against timescales. How things come together or “associate” matters for that story.

Symbiotic associations occur (1) between two or more organisms from different species and (2) are constituted by varieties of mechanisms. In endosymbiosis, physical merging events occur with one partner subsumed within boundaries of another. In ectosymbiosis, the partner remains outside, but on the surface. In mutual associations, the relationship is beneficial to all parties. Commensalism occurs when only one partner benefits with no positive or negative benefits to the other. Parasitism occurs when one partner receives positive benefits and the other negative benefits.

Symbiosis, particularly endosymbiosis, contributes to the emergence of new individuals through a union. For example, a well-known origin story of eukaryotes outlines how one type of cell, an archaean, engulfed a proteobacterium over 1.5 billion years ago. That proteobacterium evolved into what is now known as mitochondria (Reference O’MalleyO’Malley 2014, 29). This case is a transition: symbiosis generates new individualities. That’s unsurprising to someone like Lynn Reference MargulisMargulis (1998) who argued in favor of symbiogenesis theory. Symbiotic relationships are sources of innovation. New individualities are formed by means of incorporation (rather than competition).

Another way associations occur is through aggregations of cells. At what point is an aggregate one individual rather than mere assembly? Reference Hammerschmidt, Rose, Kerr and RaineyHammerschmidt et al. (2014) discuss the role of cooperation, cheater suppression mechanisms, and controlled generation of cheater phenotypes using Pseudomona fluorescens, their model organism. Reference Rainey and KerrRainey and Kerr (2010) approach the same case study but in reference to life cycles, collective reproduction, and again cheater control. Others study how collectives of Saccharomyces cerevisiae form a multicellular collective in response to environmental stress that includes evidence of reproductive division of labour (e.g., Reference Libby, Ratcliff, Travisano and KerrLibby et al. 2014). And finally, additional model organisms are used to investigate transitions to multicellularity, including unicellular and multicellular phases in life cycles, such as Dictyostelium discoideum (slime molds) or other social amoeba (Reference Queller and StrassmannQueller and Strassmann, 2012). Studying transitions to multicellularity, these researchers explore how new types of (often evolutionary) individuals emerge with features across levels of organization comprised by entities previously units of their own. The microbial world changes how evolutionary processes are viewed (e.g., see Reference SappSapp 2009).

Transitions from mere assemblages to individuals sometimes occur among closely related entities, like daughter cells glued together in bacteria and multicellular yeast; P. fluorescens and S. cerevisiae. Other times transitions occur among entities not closely related, such as multispecies biofilms, or our endosymbiotic origin story of eukaryotic mitochondria mentioned above. Reference QuellerQueller (2000) identifies the former “closely related” transitions as fraternal and the latter as egalitarian where allegiances are forged such that “higher-level” units emerge.

While major transitions in the history of life involve more than transitions in individuality (e.g., oxygenation of the earth, see Reference O’MalleyO’Malley 2014), Evolutionary Transitions in Individuality, ETIs, mark some major events (Reference Michod and RozeMichod and Roze 2000). ETIs are caused by evolutionary processes, and their occurrence affects evolutionary processes. Individuality’s role in Leo Buss’s The Evolution of Individuality (Reference Buss1987), as the title suggests, is integral to the evolution of hierarchical organization. He views life’s history as a “history of transitions between different units of selection (Reference Buss1987, 171). Put differently, individuality is the answer to the “why” when Buss asks, “why is life hierarchical?” (183). Individuality transitions are diachronically productive of new types. But can individuality ever be destructive?

While much work focuses on transitions to multi-celled life as examples of ETIs, consider the prospect of a failed transition. Godfrey-Smith discusses the process of de-Darwinization: when lower-level competition suppression, cheater controls, and so on fail such that selection drops a level, often to the detriment of the higher-level unit. Cancer is case and point: cancers as diseases of multicelled life include out-of-control cellular proliferation in a variety of forms (see Reference PlutynskiPlutynski 2018). Is cancer the emergence of new (or old?) individuals? Or is it the breakdown of multicelled individuality? As a breakdown, cancer could mark a major transition failure. Transitions, such as transitions in individuality, are often considered complete and in the past, but some transitions are still in progress and success is not guaranteed. Whether the emergence of new individualities necessarily destroys old types is beyond the scope of present discussion.

The key issue is that individuality according to historical timescales shifts the vantage point. The emergence of new individualities supports a picture of evolutionary individuality pluralism over time (Reference McConwellMcConwell 2017a, b).Footnote ³⁵ How much change is necessary to distinguish a new type of individuality, or whether forms of individuality in some sense “speciate”, will depend on:

1. (1) how we think about evolution, and

2. (2) the constraints of our typology (i.e., how types are defined in a system of analysis).

That is, evolutionary individuality as an evolutionary product takes a diachronic plurality of forms: different types of individualities likely emerge, evolve, and disappear over the course of evolutionary time (see Reference McConwellMcConwell 2017b, 131ff for a view on diachronicity and synchronicity of types, and Reference CurrieCurrie 2019b, 38ff on the emergence of different kinds through historical processes). Diachronicity requires some flexibility in how we think of evolution itself: is evolution a stable process? Or do evolution’s causal processes evolve? Some have looked for evidence of, for example, the evolution of heredity mechanisms themselves (as per Reference DarlingtonDarlington 1958). Others have considered the conditions under which a dynamic view of evolution might be supported (see Reference Calcott and SterelnyCalcott and Sterelny 2011). Within wider contexts of both historical timescales and shifts in organizational hierarchies a diachronic pluralism of individuality is more than wild speculation.

However, if evolution is itself a stable process, then presumably the role of individuality remains static, which constrains our typology. It’s reasonable to take a theoretical snapshot too. Reference ClarkeClarke (2010, Reference Clarke2013), for example, identifies a functional role for individuality realized through multiple policing and demarcation mechanisms, and is also working within a particular evolutionary framework.

Clarke does venture into broader contexts in “Origins of Evolutionary Transitions” (Reference Clarke2014, 314) working with multi-level selection models to track when “solitary organisms become subsumed within a new higher-level organism, which participates in a higher-level selection process.” However, when discussing origin, rather than evolution, of individualities there exists a difficult metaphysical gap to bridge.Footnote ³⁶

The major transitions literature develops origin narratives not necessarily presupposing prior existences of particular individuals. That is, when there’s interest in a particular form of individuality, it is often asked: how did that individuality originate? Analyses are conducted in contexts where expressions of individualities already exist, but the question remains of how that sort of organization began.

Concepts of conflict and cooperation are used to bridge both directions of the “one and many.” Two origin stories are tracked:

1. (1) One making many (e.g., such as fraternal allegiance among daughter cells that stick together),

2. (2) Many coming together to form one (e.g., such as archaeon and protobacterium coming together to form a eukaryotic cell).

If an ontology is desired beginning with one, you’ll look at things one way. If an ontology is desired beginning with many, you’ll look at things another way. For most of the practical problems concerning transitions in individuality (e.g., in labs, field experiments, etc.) many expressions of individuality already exist and so origins (and the evolutions) of individualities become local empirical questions to solve.

3. Closing Remarks on Disciplinarity and Classification

In closing, there are some key questions of individuality pluralism that remain open for future investigation. How do we best understand a plurality of types at a time (i.e., a synchronic plurality) or over time (i.e., a diachronic plurality)? What sorts of local empirical cases can better inform accounts of origins and evolutions of biological individualities?

I maintain that many answers to such questions are shaped by, and conceptually contained within, the “edges” or boundaries of disciplinary domains and subspecialties. Others exist at the interfaces of those disciplines as they change over time. In all cases, there must be a robust and recalibrating pluralism with the capability to respect its own range and limits within and among its domains. That is what it means to conduct domain-driven analyses of the sort advocated for in this section.

Table 1 began this discussion. It represents an idealized ontic landscape because the chosen port of entry is somewhat arbitrary. If Table 1 was historically constructed, that would complicate the typology of subjects. Ditto for arranging primarily by discipline or profession. So, how are we to understand pluralism and biological individuality? Inspired by Reference WimsattWimsatt’s (1972, 68) “Complexity and Organization” consider the following:

While an array of individualities was each analyzed in isolation to broach the topic, it is mistaken to approach biological individuality as merely an aggregate concept glossing over the complexities exemplified by the table’s disciplinary entry points. As Wimsatt directs us (albeit on a different matter, but I take the same messaging to apply), complexity is in part located at the interfaces of those domains.

Those interfaces cannot be understood without considering plurality as a thicket in the Wimsattian sense: when navigating the weeds, it’s about how you strike that machete. Any entry point to our landscape is frayed around the edges; disciplinary machetes do not have sharp blades. For example, Child was writing before the modern synthesis “hardened” when some disciplines were ignored or simply left behind. Thus, one can draw pre-synthesis insight from Child who drew from an array of specialties concerning metabolism, immune function, and growth occurring in developmental cycles. While analyses are domain-driven, one type of biological individuality is not exclusive to some subdisciplinary area.Footnote ³⁷

Biological individuality is held accountable to disciplinary ambiguity, changing disciplinary boundaries, and ways in which classificatory systems are affected by said ambiguities and frankly evolutions of disciplines themselves. And so, Table 1’s taxonomy of individuality types was arranged according to domain; evolutionary, immunological, ecological, metabolic, developmental, and so on, which remains somewhat misleading. If domain-driven analyses of biological individuality are derived from the life sciences, the resulting pluralism must be considered against the thickets of disciplinary backgrounds and historical timeframes in which they are conceived.Footnote ³⁸

1. Pillars of the Past: Colonial Logic & Objectification

Reference SinclairSinclair (2020, 67) discusses how biological individuals are taken for granted on the ground level of many (philosophies of) sciences. There is critical overlap between focus on individuality and problematic enlightenment metaphors, concepts, and values. The tendency to “see the world as composed of essential, clearly delineated individuals is part of a wider settler-colonial philosophical project that justifies the subjugation and attempted erasure of Indigenous relational science and knowledge” (Reference Sinclair2020, 67). By referencing the enlightenment, Sinclair notes a pillar of the past, but what’s the problem?Footnote ⁴² Consider the following to help illustrate the point.

Sinclair discusses how indigenous logics emphasize relationships and nonbinary natures. That is in stark contrast to asking the following: Are puzzle cases for individuality, such as ecological communities, individuals or not? A binary logic frames the expected answer as only one disjunct of an exclusive “or”. That binary expectation has social and political consequences, especially in conservation efforts that assume only singular, unified organizations or “individual agencies” can be objects of recognition and protection. This is part of why I take Aldo Leopold to have sought an organismality (or as Millstein has argued an “individuality”) designation for land communities. In that sense, individuality as a designation translates to recognizing those rights and protection.

In other words, colonial logic has normative consequences for ontological designations of individuality. For example, Millstein’s work on Leopold’s land ethic suggests that the individuality status of land communities (i.e., both biotic and abiotic components) is a prerequisite for moral consideration in conservation efforts. Individualism as an organizing principle in colonial frameworks protects ecological “agencies” by reducing restraints on freedom and by recognizing eco-interactions and natural systems as entitled to special privileges. That is, if forced to work within the binary logic of a colonial past, then designating land communities as individuals will secure both freedom from external disturbances that impede functioning (e.g., like polluting, commercial use, etc.) and protection in the form of persistent security. As one singular, unified individual (i.e., the land community in this case), it has needs capable of recognition, such as to persist on its own terms without disruption, and protection of those needs.

Similarly, the status of individuals as objects of recognition and protection (or conversely, of denial and extermination) applies to cases of conservation through management of the environment. In colonial frameworks, that management is often viewed as a one-way relationship of control, rather than a reciprocal and embedded process of humans alongside and within nature. The former separates the manager (i.e., humans) as external entities imposing their will often for use and exploitation of the land and by that action objectifies nature. This suggests that determining the individuality status of some ecosystem, ecocommunity, or Leopold’s land community, and so on for conservation and protection efforts is itself a historical product of nature’s colonial objectification. That objectification is driven by binary assumptions about how individuality as an organizing principle is applied precisely for (human) management and control of nature.

2. Individuating Characters & Their Phenomenal Qualities

Undoubtably, individuality is a scientific tool to objectify the natural world: what are biological individuals if not objects sorted and typed by systems of classification. Those systems organize where distinguishable organic units belong under those schemas. Promoting epistemic traction on nature’s complexity are qualities that make organic objects readily identifiable. For example, consider the slender body shape of hornets and wasps compared to bees, the varying texture of mountain pine needles compared to red pines, the red marks as “ears” on red-eared slider turtles as a distinguishing feature compared to painted turtles, and so on. Such phenomenal qualities – individuating characteristics from the physical and behavioral markers of bone shapes to the pitch of howls signaling aggression – help naturalists identify, track, and sort according to the systems of organization existing in whatever field of biology.

Philosophers already discussed individuating characters and the discerning work they do in phylogenetics, and the nature of characters as basic units of biological analysis. However, the epistemic value of individuating characters is not only how they historically trace degrees of relatedness in descent, but also how they provide material for certain adaptive functions (Reference GriffithsGriffiths 2006). It is “confusing that ‘character’ can mean either any measurable property of an organism or only a property recognized as biologically significant in some theory of the organism” (Reference GriffithsGriffiths 2006, 12). The ambiguity of ‘character’ becomes more complex when not just of single-species organisms, but also of multi-species collectives spanning organizational scales. However, Griffiths clarifies that philosophers are familiar with the same ambiguity in ‘property’. Whether measurable or theoretically significant or both, ‘character’ is used in a highly specialized sense – they are the phenomenal qualities put to use, measured, and manipulated. In that sense, they are tracking tools for the ways in which individuals are phenomenally objectified, rather than considered active subjects.

In contrast to prior work on character individuation, phenomenal qualities associated with biological individuality shift slightly. Phenomenal qualities for discerning biological individuals range from part–whole relationships to boundaries based on both material borders and/or functional limits.

For example, while a mitochondria’s functional capacity can be limited by the bounds of its cellular container (e.g., its mobile capacity limited by that material boundary to move between cells under certain contexts), the functional interactions, rather than material boundaries per se, of keystone species with other members of an ecological community identify the bounds of a functional group.Footnote ⁴³

Other phenomenal qualities, such as those detected through immediate experience or with the help of a microscope or other tool, are tougher to get a handle on, and so philosophers ask whether the one-to-many symbiotic relationship between a Bobtail squid and their illuminating bacteria is consistent enough to consider that consortium as one, continuous individual. Or, concerning matters of identity, they ask whether the continuous turnover of our cellular parts challenges both the oneness and sameness of who we are.

And so, the phenomenal qualities just discussed – part–whole relationships, boundary closure (both material and functional), stability, continuity, and identity – can be contrasted against how character individuation decomposes objects. The listed qualities are distinctly philosophical. As analytic categories – the lenses of analysis – they harken back to the hallmarks of concepts like identity and discernibility that shape philosophy’s past as a metaphysical enterprise. Past pillars indeed.

Philosophy, even in the practice turn, runs biological objects through the same ringer, through the same categories of analysis Reference SinclairSinclair (2020) points to when emphasizing the enlightenment’s long shadow. I am critical of the assumption that determinate properties of individuals are (and will) manifest in their concreteness: that an individual’s individuate-ability is something rather distinct from how they are conceived. Pre-theoretical concepts of phenomenal qualities do not always harmonize with the shifting, processual nature of biology, nor in how those qualities function in a binary colonial logic. So, individuate-abilities must be operationally fluid.

If rejecting positivist standards (e.g., standards of necessary and sufficient conditions, uncovering universally absolute truths from scientific reasoning and data), then exactly what standards govern conceptual analyses of biological individuality? Products of our analytic categories are at times familiar, and sometimes surprising. Reference Bueno, Chen and FaganBueno et al. (2018) argue that assumptions shaping categories of analysis like mereology and boundary closure are challenged by the way biological practice is conducted. But one consequence is that a puzzle’s status as a puzzle of individuality is in part determined by our analytic categories and in part determined by the way scientists conduct empirical work.

In conclusion, phenomenal outputs are characterization-laden: they refer to abstract ideas, and are learned, taught, and housed in the kind of disciplinary content that many would like to see deconstructed and even dissolved in philosophical analysis, such as colonial and positivist frameworks. Certainly, disciplinary practices change. Yet isn’t one promise implicit in the so-called “practice turn” that we’re turning away from such pillars of the past? If so, there may be reason to receive that promise with some caution.

Turn-talk punctuates history into before the turn and after: it divides practices, mechanisms, processes, concepts, and theories into those two select categories, for example, practice-based vs. not (see Reference GriesemerGriesemer 1996a, Reference Griesemer1996b). The risk is, of course, homogenizing stages as if some kind of categorical shift occurred. But history never really stacks so neatly. “Turn-talk” is branding.Footnote ⁴⁴ It marks resistances to what came before. But it must be asked, what exactly is resisted and where to next? What pillars of the past should be left behind when analyzing biological individuality?

Analytic categories concerning absolute natures, boundaries, and mereology in philosophy are infused with past folly and foible.Footnote ⁴⁵ If those categories continue to shape the way individuation activities across the sciences are reconstructed and epistemically evaluated, then the “turn” will continue its circular direction around the conceptual axis Sinclair is worried about. Epistemology of biological individuality is not conducted in a history-free vacuum.

1. Looking Back

Those specializing in science and technology studies, including philosophers, have already drawn conclusions about individuation and identity based on technologies produced from applied sciences to navigate and observe nature, while also trouble shooting medical needs. As Donna Haraway states in Cyborg Manifesto (Reference Haraway and Haraway1991, 163, 180), intervention efforts focus on control strategies concerning interfaces:

Haraway tells us that the agent has an active role in organizing boundaries of (or in?) their worlds. Taking seriously the deconstructive element in Haraway’s work – pulling apart preoccupations with stable and sharp boundaries – then agents are responsible for where those boundaries rest at any local point of time. The heavy philosophical lesson, as I take it, concerns the relational status of boundaries inspired by both techno science and science fiction: foundational distinctions are contingent at best. That is due to how biology and politics are intertwined, Haraway’s biopolitics, through feminist reflections on social stratifications (i.e., race, gender, sexuality, class, etc.) in science and biotechnology. A main lesson from Section 1 of this Element concerned pluralities of individualities as domain-driven, that is, as defined according to some disciplinary area of the life sciences and held accountable over time to changing and ambiguous disciplinary boundaries. My intention in returning to Haraway is to emphasize how practitioners have an active role in organizing material and ideological aspects of those boundaries. Consider the following:

In other words, the agent is an active contributor to boundaries imposed by technology. Additionally, though, as emphasized in Companion Species Manifesto, that role is entrenched in the history of companionship with other species – be it dogs or microbes. Our participation in these relationships is shaped by both nature and socio-cultural ideology.Footnote ⁴⁷

Haraway draws conclusions that undermine characterizations of biological individuals apart from their social contexts. And that is exactly what piques public interest now: considering individuality at the complex juncture of bio-social spaces. There are many popular science articles that demonstrate this channel of interest begging for technical philosophical input. Here are three examples.

Reference Wallace-WellsWallace-Wells (2015) published “Adventures in the science of the superorganism” using examples to motivate why we should care about biological individuality, such as one twin ingesting the embryo of the other twin in utero, the trillions of gut bacteria that house themselves in/on us, the viruses that colonize our DNA, and the ever-so-strange fetus-in-fetu cases. What is at stake, according to the author of this public-facing op-ed, is reconciling our own personal individuality with the notion that humans are super-organisms with mosaic identities. However, Wallace-Wells is concerned about a richer sense of individuality – its social sense – as if that hinges on biological boundaries. This line of bio-social reasoning is taken even further in a different article published the same year.

Reference SalisburySalisbury (2015) in Science Daily asks whether the pronoun “I” is obsolete due to recent microbiological research. The proposal is that holobionts and hologenomes – associations between macro and microorganisms are the units of biological organization subject to selection, drift, and mutation. And while there has been philosophical work examining various forms of scientific pluralism using holobionts as a case study (Reference ŞencanŞencan 2019), Salisbury presses the social consequences of a mosaic self. Searches on Aeon and Quanta Magazine will produce similar results concerning the bio-social interest that shapes the way individuality translates to both cross-disciplinary and public realms. Enter technology into the equation and the complexity increases though.

Reference DonahueDonahue (2017) published a National Geographic article titled “How a color-blind artist became the world’s first cyborg.” The article covers the case of Neil Harbisson, a color-blind artist, who uses an antenna-like implant that allows him to “hear” color via vibrations in his skull. While certainly, as the article states, humans have used technology to alter physical and mental capabilities, the claim of the article is that we’re already cyborgs in Haraway’s sense. Harbisson views the antenna as just another part of his body: we alter both ourselves and our environments. So, when Haraway says we are responsible for boundaries – that we are the boundaries – this science fiction reality is what I take Haraway to have in mind.

Popular op-ed articles are watery signatures of philosophical concepts suspended in time: they take bits of philosophy that look interesting from the outside and press on questions that sometimes philosophers no longer want to touch. In other words, one might object that popular science articles about social questions concerning individuality in biotechnology may not be motivating for philosophers. To that end, Reference Kendig, Bartley and BurstenKendig and Bartley’s (2019) work on synthetic kinds could have been further developed as an analogous way to approach synthetic individuality. Or controversies surrounding human gene editing and the birth of the first three parent human individual in 2016 could have been considered.Footnote ⁴⁸

In response, I maintain that how popular science articles frame social questions about biopolitics and biotechnology should be investigated, especially if philosophers of science have a responsibility to contribute to public understanding of science beyond what science journalism provides (see Reference Cartieri and PotochnikCartieri and Potochnik 2014). Philosophy’s social relevance includes not just educational and institutional reform, but also management of science’s social position, which involves its public expressions by both journalists and scientists.Footnote ⁴⁹ Popular biotechnology articles are one main avenue through which publics will access questions about biological individuality. Thus, social questions drawn from those sorts of articles exemplify how science fiction, synthetic biology, and politics merge for the publics in bio-social engineering of individuality.

While the relationship between biology and society is informed by knowledge systems that define those spheres (i.e., biology and society, science journalism, fiction, and public education), that juncture is approached by philosophers of science and biology in other areas, such as evolution’s meaning and science’s sociality. Biological individuality is just another space in which public-facing work can be fruitfully repurposed beyond traditional questions of metaphysics and epistemology. Looking forward there is much opportunity for a “science and society” approach to biological individuality.

2. Looking Forward

The normative consequences of individuality and identity are significant in contexts of biotechnology from organoids and chimeras to CRISPR-Cas9 gene therapy in biomedical research contexts.Footnote ⁵⁰ Organoids are biological structures created from stems cells in vitro, which partially mimic the function of organs, while cerebral organoids are “stem-derived 3D biological structures, which are self-organizing in morphological units resembling a developing brain” (Reference Hostiuc, Rusu and NegoiHostiuc et al. 2019, 119). How do these cases fit into systems of values? Cerebral organoids are of special interest because they derive from human embryonic or adult cells but are subject to the use of techniques that can potentially alter human constitution to include genetic material from other species resulting in chimeras. As Reference Hostiuc, Rusu and NegoiHostiuc et al. (2019) state, human-animal chimeric organoids have already been developed and chimeric cerebral organoids could be a next step. They cite Karpowicz et al. who argue that these neural chimeras should,

1. (1) Use a minimum number of human stem cells possible, and

2. (2) Make sure that the host animal is not too morphologically or functionally similar to humans to mitigate the risk of developing human-like neurological networks (2019, 120).

These criteria are followed by attempts to define some threshold of considerability unique to humans to reduce the risk of causing harm, or at least to mitigate the risk of harm in unknown contexts (120ff).

Ethicists will no doubt groan at the threshold-unique-to-humans line and even the most scientifically inclined metaphysicians will recognize the tale as old as time: identity and (re)constitution. Replace the Ship of Theseus with gene editing technologies and Petri dishes and a philosopher will argue that we’re still in that same boat. What’s missing in Theseus’s old ship parable, however, is who replaces the planks: as philosophers we tend to focus on the end result, the published planks, without paying much attention to who is doing the constructive work. So perhaps Theseus’s ship needs an updated framework, one which aligns with Haraway’s insights; namely, about who is reconstructing and deconstructing the boundaries that shape biological identity, alongside their motivations and methods.

Such a focus on scientific work – that is, on who individuates and how those individuation practices have philosophical consequences for the products of scientific knowledge – is not enough to inform the normative concerns cited above. More is needed from the science and values and STS networks to develop individuality’s epistemic dimensions in conjunction with its social and political features across biotech contexts. Approaching a similar theme, Reference LonginoLongino (2002, 59) has described the lesson I’m now trying to capture: “ … nature does not, cannot, act alone in the laboratory … but needs a whole social and institutional structure in order to be deployed.” The point stands in both experimental and bioengineering contexts: synthetic individuality draws attention because of the immediacy of human intervention, but it is not necessarily unique in either (1) its dependency on disciplines or domains as was discussed in Section 1 of this Element or (2) its relationship with social and political values, for example, like assumptions of human “specialness” as will be further discussed in Section 3’s historical setting.

With the ability to shift the constitution of a biological entity’s individuality through technology, there is promise for productive work ahead. Philosophers of science engage biotechnology and biopolitics in a way that both departs from, yet complements, the current work of bioethicists. Philosophy of science, even that which focuses on values and public policy, varies in nuanced ways from that of analytic and applied ethicist interventions. Looking forward, philosophical possibilities emerge from synthetic biological individuality, that is, not what biological individuality is or how it’s used, but what it could be and how it should be designed for society.

1. Two Contemporary Takes on Agency in Biology’s History

At first glance, approaching biological individuality through an agency lens risks coloring outside the lines of biology. Connecting the biological to the social is controversial and treacherous ground.Footnote ⁵¹ For individuality, that means connecting basic boundaries of biological objects – whether distinguished by us or discovered “out there” – to social concepts like self-identity, intention, and meaning. Socializing the agency of biological individuality risks a vestige of biology’s past. So, how might agential dimensions be considered without venturing too far down that kind of path?

Around the time that Reference WalshWalsh’s 2015 book Organisms, Agency, and Evolution was published, it seemed in stark contrast to current debates over evolutionary individuality discussed in Section 1. The latter was an alternative research context focused on how inheritance occurs for individuals in selection, which did not include organisms as active participants in evolution per se. Walsh resists notions like intentionality and consciousness, especially in a recently coauthored exploration of the agency perspective (see Reference Sultan, Moczek and WalshSultan et al. 2021). Working at the edge of the social, while at the same time resisting that dimension, marks an intriguing foray into organismality from an agency point-of-view.

Reference WalshWalsh’s (2015, Reference Walsh, Nicholson and Dupre2018) view embraces a traditional dimension of organisms; an active dimension that became less popular after the modern synthetic theory of the mid-twentieth century.Footnote ⁵² A return to active organismality marks an extended evolutionary point of view, but what exactly is the relationship between active organismality and an extended evolutionary view?

The processes of development are not just outcomes of some genetic script, but an outcome of processes happening within the organism on all sorts of (environmental and other) information (Reference Sultan, Uller and LalandSultan 2019). In discussion, evolutionary ecologist Sonia Sultan explained that causes of phenotypic variation must be distinguished from their record as DNA sequences. Many evolutionary experiments are designed to surpass environmental components, which naturally inflates genetic components (see Reference Sultan, Moczek and WalshSultan et al. 2021). And so, on an extended evolutionary view the organism matters as the locus of interaction dependent on environmental contexts.

Reference Sultan, Moczek and WalshSultan et al. (2021) illustrate how an agency perspective draws from a different conceptual framework. Mechanistic explanations address components of systems to explain how they work (e.g., the genetics), which is successful if a system works the same to produce the same outcomes across contexts. However, if that system is context-dependent, then a mechanism-based explanation will be incomplete producing gaps in understanding.Footnote ⁵³ An agency perspective is not in competition, but rather complementary to mechanism-centered views: organisms integrate both genetic and environmental influences to produce a specific phenotype, and so how a phenotype works (or doesn’t) for the organism is what needs to be explained.

This research program encourages the redesign of experiments to reveal complex evolutionary and developmental interactions considering organisms as active participants in that process. It is in that sense that organismal individuality is agential. One main focus of Reference WalshWalsh (2015) is why organisms are alienated from evolutionary biology; why they were eclipsed by preoccupation with mechanisms, genes, and population dynamics. Evolution as an ecological phenomenon means that it “happens to a population (or lineage) as a consequence of individual organism’s purposive engagement with their conditions existence in the struggle for life” (Reference Walsh2015, 208). Of greatest theoretical significance is not the gene or the organism per se, but rather the “organism situated in a system of affordances” (209). In that sense, organisms are agents as both subjects and objects of evolutionary change. Reference WalshWalsh (2015), Reference Sultan, Uller and LalandSultan (2019), and Reference Sultan, Moczek and WalshSultan et al. (2021) are all explicit about their inspirations from Lewontin who said,

Thus far, a different conceptual stance was discussed: a Lewontin-inspired call to rethink experimental design according to Sultan et al. That stance designates active roles to organisms in evolution through their development (i.e., from an “evo-devo” point of view) as proper subjects, rather than passive objects, which was conceptually eclipsed at some point. Consider two intriguing narratives on how that eclipse occurred.

On the one hand, Walsh tells us that organisms in their ecological contexts (i.e., what they do in pursuit of their ways of life) is a cardinal lesson from Darwin’s Origin (2015, x). This was obscured by marginalization of organisms under Synthetic theory, which disadvantages evolutionary understanding. That organisms as purposive agents contribute to – or enact – evolution is a feature lost to history nearly a century after Darwin (xii). Walsh states, “In Darwin’s evolutionary thinking, organisms surrender some, but by no means all, of their theoretical significance” (44). On this reading of Walsh’s work, centering agency is a calculated return to Darwin.

On the other hand, historian Jessica Riskin tells us that Darwin wrestled with agency. After Darwin’s study of Paley’s argument from design at Cambridge, he adopted certain principles, that is, “the notion of mechanical adaptation or ‘fitness’ of parts, and the related requirement that parts be passive, that a properly scientific account of living phenomena ascribe no agency to the phenomena themselves” (2016, 215). Darwin was deeply torn between the “mechanist dictate to banish agency from nature and the organicist impulse to naturalize agency, to make agency synonymous with life” (2016, 215). This reveals a tension in the history of evolutionary thought, which complicates the return to Darwin. But why?

One explanation is that agency’s relationship with purposive action is not unlike “occult qualities” from scholastic science. Scholastic science posited innate tendencies, for example, such as an object’s innate tendency to fall once dropped because of the object’s drive toward Earth’s center. In Scientific Revolutions, Kuhn contrasted scholastic science with early Newtonians seeking corpuscle-mechanical explanations of gravity, rather than posit innate qualities. So, accepting agency risks something similar: a scholastic absence of mechanical explanation in favor of occult natures.Footnote ⁵⁴

In fairness, agency is empirically established by those working within evo-devo: feedback processes in development, and between organisms and their environments, actively contribute to shaping evolutionary change. Reference Sultan, Moczek and WalshSultan et al. (2021, 4) provide an experimental example. When parental anemonefishes are exposed to high concentrations of carbon dioxide in their development, their offspring then develop normally in elevated CO_2, and offspring of unexposed parents demonstrate sharply reduced growth. They summarize agency as “the capacity of a system to participate in its own persistence, maintenance, and function by regulating its structures and activities in response to the conditions it encounters” (Reference Sultan, Moczek and Walsh2021, 4). While not purely mechanical, agency does not defy scientific explanation.

However, naturalizing agency still attracts some stigma due to what agency historically represents. Consider when Riskin says,

In other words, Darwin rejected two sources of agency because internal striving harkened back to the scholastic occult qualities of concern, while external divinity brought in supernatural explanation. In a post-Newtonian era, any successful view about the cosmos’ mechanical nature ought to commit neither offense. Riskin continues about Darwin,

Riskin identifies two kinds of inner agency in Darwin’s work: inherent tendency of the parts (e.g., Nisus formativus or the tendency to multiply) and behavior of the overall organism. She clarifies (Reference Riskin2018, 227), “Darwin was abidingly ambivalent of the first sort of agency as he was unwavering with regard to the second,” which means Darwin was ambiguous on the subject of innate tendencies, such as the tendency to vary (the causes of variation), as well as inner strivings to “to complexify, to progress, to revert” (233).

And so, one risk of naturalizing agency is rediscovering relics of innate tendencies and design. Modern contexts of agency should remain historically informed due to agency’s historical relationships with occult and supernatural qualities.

2. A Tale of Caution: Agency & Biological Individuality Writ Large

Discussion thus far centered on organismal agency in evolution and why it remains controversial. In contrast, our cautionary tale is about the agency – inner striving, direction, and purpose – of biological individuality in evolution writ large, which surely seemed innocuous during its inception (but arguably was nocuous later).

For Reference HuxleyJulian Huxley (1912, Reference Huxley1923), the evolution of agency was significant in his theoretical work. As we’ll see, Huxley tied agency with a progress-loaded view: evolution aimed toward “Perfect Individuality” through three different grades. Those grades are discussed later. For Huxley, humans were at the pinnacle of that progress. Individuals in the Animal Kingdom (Reference Huxley1912) is a bizarre read under the scope of agency in evolution, but downright alarming when Huxley’s later role in the history of eugenics is considered. He, like many of his contemporaries, aimed to scientifically manage the evolution of individuality. As discussed later, at times he writes metaphorically about evolution becoming conscious of itself or “evolution’s agency” with individuality marking the emergence of an agential capacity to control the directions toward which evolution tends. For Huxley, it was a moral imperative for humans as products of that agential capacity to do just that – to scientifically manage and perfect evolution of human individuality.

Next, the scope of that cautionary tale about biological individuality is developed. Huxley was a significant contributor to the Modern Synthesis coining its name. His early view of an agent’s active role, specifically human agents, in controlling their own evolution meant that individuality reconstructs the conditions in which populations evolve. Coupled with the evolution of agency – agency as perfecting individuality – was an ideology of progress. Was that progress toward perfection theological?

While there is not a single mention of God in the 1912 book, as we’ll see, Huxley later wrote a spirited introduction to Pierre Teilhard de Chardin’s Phenomenon of Man (Reference Teilhard de Chardin1958). Huxley declares his independently anticipated claims in Teilhard’s book (Reference Teilhard de Chardin1958, 12), such as the idea that evolutionary processes should only be described in terms of their direction (rather than origin), the tendency of increasing complexity, and the scientific study and management of human evolution in that context. Huxley maintains that while scientists like himself may find it tough to follow Teilhard’s reconciliation of evolution with God, the merging of individual human variety into a unity with the emergent Divinity in “no way detracts from the positive value of [Teilhard’s] naturalistic approach” (Reference Teilhard de Chardin1958, 19). And so, naturalized agency and Christianity, in the cases discussed next, follow two sides of the same fraught coin: ideologies of “progress” infuse biological individuality, such as the “merging” or ridding of deviant variation that undermines its perfection. Our cautionary tale indeed.

1. Biological Compounds: Origins vs. Growth

In the Origin, Darwin considered hybrid animals as “two different structures or constitutions having been blended into one” (4th Ed. Origin, Reference Darwin1866, 317). However, it wasn’t until the 6th edition (Reference Darwin1872, 441) that Darwin adds a glossary of scientific terms, which included an explicit definition of “zooid” that aligns with Thomas Huxley’s view on individuality. Darwin says,

Thomas Huxley was not happy with Darwin’s problem of compound individuality (Reference ElwickElwick, 2007, 128). As Elwick describes, Huxley defined biological individuals not by their independence but “as the entire product of a single sexually fertilized ovum” (Reference Elwick2007, 133). Independence as a criterion led to difficult questions: “were the detached and free-swimming sexual parts of marine invertebrates entire individuals, or ‘mere organs?” (128). And so, “Huxley privately began calling each false individual a ‘zooid’” and announced in 1851 that the distinction between zooid and individual was founded on a zoological basis and a fact of development” (133). Huxley sought to close off the possibility of plants and animals as compound animals by making the problem of compound individuality a pseudo-issue (Reference ElwickElwick 2007, 149).

To Huxley, the aphid, for example, was a single individual composed of zooids that all budded after an initial act of sexual fertilization (Reference ElwickElwick 2007, 133). The term “zooid” was to disambiguate:

1. (1) Individuals as components of species starting from “true ova” as single cells undergoing development, rather than

2. (2) mere reproductive bodies as aggregations of cells.

Huxley’s temporal definition concerned the developmental cycle starting with a single cell as one individual. He recognized trouble with his temporal definition – to someone interested in morphology the emphasis on origins was an issue and the morphological and anatomical distinctions between a zooid and individual were not always easy to determine (Reference ElwickElwick 2007, 136). How do we distinguish the production of something new from mere growth of the same? Morphologically the answer is difficult to establish. But Huxley thought that origin made a qualitative difference: despite the similarity of shape, for example, such as Nereis worms produced by budding versus those emerging from sexually fertilized ova. Distinguishing growth from production of a new individual was the developmental cycle as suggested in a later lecture:

This excerpt characterizes Huxley’s view about the cyclical containment of individuality where parts share the same fate in the reproductive process only to begin anew indicated by reproductive “bottlenecks” from one cell to many and back again. Huxley used developmental life cycles – what he designated the Sisyphean process – to distinguish growth of the same individual from the origin point of a new one. And to emphasize individuality’s incessantly repetitive and recursive nature.

I draw on Elwick’s discussion of Thomas Huxley’s compound animals, individuality, and zooids as it situates biological individuality historically. Through Elwick’s archival work, biological individuality was a factor in the professionalization of the life sciences; a terminological factor carving disciplinary boundaries. Yet, individuality also shaped Huxley’s views concerning the importance of origins: a technical departure, as Elwick (138) states, from competing views, such as Richard Owen’s who argued that morphological similarity was more important than origin.

2. Contextualizing Life Cycles & Compounds

Thomas Huxley’s choice to contextualize the repetition of individuality’s temporal character as Sisyphaean is striking to philosophical ears: the Evolution and Ethics lecture is Huxley’s warning against drawing social meaning from biology. Huxley meant to convey the injustice, and even immorality, of nature through a Buddhist framework of meaning (Reference HimmelfarbHimmelfarb 2014). However, read through the lens of individuality reveals him grappling with the individual’s place in it all: “individual existences are mere temporary associations of phenomena circling round a center, like a dog tied to a post” (Reference HuxleyHuxley 1893). He muses at the cyclical going forth and returning to the starting point as like ascent and descent, which he viewed as the great leveler: all life forms “from very low forms up to the highest … the process of life presents the same appearance of cyclical evolution” or the incessant return to origin points (Reference Huxley1893, also see footnote 27 of this Element). The beanstalk parable with which the lecture begins indicates that science gets us to high places but has nothing to say about what it all means. One can infer that for Huxley, biological individuality marked a pervasive cycle with no greater social meaning. Biological individuality was the great, repetitive leveler of monotony and evolution’s recursive nature was more of a cycle than directional per se: changes leading nowhere in particular as depicted in Figure 8.Footnote ⁵⁵

Figure 8 Depiction of Thomas Huxley’s cyclical characterization of individuality through time contrasted against Julian Huxley’s view that individuality progresses along a vector directed toward Perfect Individuality. While T. Huxley resisted applications of biology to society, J. Huxley’s concept of Perfect Individuality concerns that very application

However, Thomas Huxley’s grandson, Julian, did not share the same outlook.

A young Julian Huxley cited his grandfather’s work as an epigraph in the first essay of Essays of a Biologist (Reference Huxley1923, 3). Compounds, for Julian were not pseudo-problems, but rather emergent and indicative of directional progress (Reference Huxley1923, 242).

Huxley (heretofore Julian – the grandson) argued that new combinations and properties arise all the time; that compounds, whether chemical or organic, were not merely reducible to their parts. Citing his own first publication Individuals in the Animal Kingdom (Reference Huxley1912), Huxley identifies compound animals, whether physically bound like Hydra or a Portuguese Man o’ war, or “mentally bound” like eusocial insects, as cases exemplifying intermediate steps of progress (Reference Huxley1923, 85). In other words, compound-ing is a mode of progress. He states,

There is a complex dimension to how young Huxley identifies individuality with progress in his Reference Huxley1912 book published a decade or so prior. Individuals in the Animal Kingdom identifies three grades of individuality with compound wholes as essential building blocks toward the second and third grades (see Reference Huxley1912, 158). Let’s consider all three to unravel Huxley’s view of how evolutionary progress occurs.

3. Individuality as a Driver of Progress in Evolution

The first grade of individuality consists of individuals that, while often homogenous, are “marked off” as a closed system (50). Compound individuals without (e.g., colonial algea like Gonium) and with (e.g., Volvox) division of labor are contrasted against full individuals (e.g., a protozoan or fertilized ovum) of the first grade.

The second grade includes parts of compounds losing their own independence and is “in essence a progress towards greater complexity” through internal differentiation and specialization (136). Compound individuals without (e.g., some sponges) and with (e.g., hydroid colonies) division of labor are contrasted against full individuals in this second grade, namely, humans “regarded singly,” and Hydra.

The third grade, in Huxley’s appendix, identifies full individuals with examples like ant communities, human society, and lichen. The third – and as such highest – grade involves subordination of lower individualities into higher with symbiotic interdependence as a (sometimes loose) binding factor. Compared to the second level where human organisms are the paradigmatic example, the third level of eusocial colonies and human societies is viewed as immature or rudimentary in its current development overall. But his idea was that we can manage that progress of human society’s evolution by constraining and directing its individual identity. Huxley identifies the highest common measure of progress as individuality, which establishes “a direction in which its movement is tending, and from that deduce the properties of the Perfect Individual” (Reference Huxley1912, 2). This directionality of life is riddled with “limitations of her own physical basis” where imperfect materials only carry so far, but the emergence of the human brain saves the day:

Human consciousness, or agency, transcends substance and thus humanity is meant to further individuality’s perfection by manipulating the very processes that gave its existence.Footnote ⁵⁶ The relationship of this view with the social peril of perfection in eugenics will be discussed in more detail later.

Given that Perfect Individuality peaks with humanity’s emergence and management of evolution, recall that Huxley’s introduction to Pierre Teilhard de Chardin’s The Phenomenon of Man (1958) was one of praise about their converging ideologies. Citing his own past work concerning the uniqueness of humanity and human evolution and defining evolutionary progress, Huxley (1958, 27) describes himself as independently coming up with two similar ideas: First, that evolution was finally becoming conscious of itself and second, that the evolutionary process led to higher degrees of organization toward becoming One.

For Teilhard, perfection was a complicated concept.Footnote ⁵⁷ In a rather favorable review, Reference SimpsonSimpson (1960, 207) defined Teilhard’s view of perfection as “the consciousness of the universe, which will have evolved through man, [and] will become eternally concentrated at the ‘Omega point’ … [t]he whole process is intended; it is the purpose of evolution, planned by the God Who is also the Omega into which consciousness is finally to be concentrated.” Citing Teilhard’s views on an irreversible perfection (i.e., indicating a vector of progress along a line of time), Simpson notes that book is submitted as a scientific treatise, yet wholly devoted to a thesis that is not demonstrable scientifically. Teilhard’s premises were predominantly religious, and the work is not a derivation of religious conclusions from scientific premises (Reference SimpsonSimpson 1960, 207). In contrast, one might object that Huxley’s view of progress toward Perfect Individuality was derived from scientific premises by way of his empirical work. However, how Huxley describes the premises of his own work is more nuanced than that.

In the 1912 preface he states, “I have tried to show in what ways Individuality, as thus defined by me, manifests itself in the Animal Kingdom” (viii, original italics). Thereafter, Chapter 1 begins with a quote from Nietzsche’s Zarathustra concerning the perfection of individuality, to which Huxley proclaims that his own individuality, and individuality in general, is now within the purview of the Zoologist, rather than only the philosopher. But why would the Zoologist take interest in understanding individuality in nature and curating its ideal?

One of Huxley’s close colleagues, J.B.S. Haldane, also later characterized the evolutionary process as “passing from the stage of unconsciousness to that of consciousness” in an article outlining the possibilities for radical improvement of humanity’s evolutionary future (Reference Haldane1947, 51).Footnote ⁵⁸ Haldane identifies biologists as not just mere fact collectors, but as life’s tailors. Generally, in the post–civil war era, there was intensifying concern of humanity’s degeneration after the end of slavery, the arrival of immigrants, and criminality, as well as birth control methods to prevent the “wrong” people from proliferating in human populations. In distinguishing his eugenics (i.e., as the application of biology to society) from that of Hitler’s, Haldane argued that when tailors perform measurements and advise adjustments (e.g., for clothes), it is not without disregard for humanity’s qualities beyond mere lumps of matter (Reference HaldaneHaldane 1947, 45). While he recognized the need for diversity to shape the resilience of evolutionary populations, that diversity was of a very particular sort. Reference HaldaneHaldane (1947, 51) identified birth control and artificial insemination by “great men” as liberal methodologies to reform society’s illnesses and systemic poverty. Huxley too wanted to “improve” the human stock achieved through contraception, and artificial insemination by “highly gifted men” (Reference Huxley1936, 199). The relationship between Reference HuxleyHuxley’s 1912 book and his later lectures on eugenics in the 1930s and 1960s is discussed later. For now, the driving point is the following: Huxley’s stated premise in his Reference Huxley1912 preface concerning individuality’s perfection as now within the purview of the Zoologist is not benign.

Huxley resonating with Teilhard’s Phenomenon of Man is telling from the perspective of John Reference SlatterySlattery (2017). Slattery controversially explores Teilhard’s connection to eugenics. His archival work reveals Teilhard’s views on the inequality of races, the acceptability of violence, and eugenics as a means to perfect humanity. The Omega Point, in Teilhard’s own words, depends on the inequality of races and methods to counteract “unprogressive ethnical groups,” a view held not just before WWI, but into the last decade of his life after the horrors of concentration camps became known (Reference Slattery2017, 75). Teilhard’s line of reasoning also included notes on groups of inferior value compared to whites, which he contended were not due to religious belief, but an inferiority with natural foundation (74). Teilhard’s Omega Point concerned the biological maturing of the human type toward that of more value – it was motivated by the aim to gain control over evolution toward biological purification and perfection (79).

So, while Teilhard saw evolutionary perfection through a religious lens, Huxley’s compounding complexity – his grades of individuality – mark the irreversible and progressive path toward Perfect Individuality. Huxley’s very clear about his independent arrival to some of Teilhard’s conclusions about progress in evolution (see Reference Teilhard de Chardin1958, 12). He reports that it was in Essays of a Biologist (Reference Huxley1923) where he began defining evolutionary progress. And in those essays, he argues that compound animals were not pseudo-problems, but emergent phenomena signaling progress (Reference Huxley1923, 242). There he cites his own 1912 book Individual in the Animal Kingdom as identifying grades (of individuality) that exemplify the immediate steps of progress (Reference Huxley1923, 85), but that progress was not socially neutral:

In Huxley’s own words then, his prior Reference Huxley1912 work on individuality’s perfection and progress was the foundation for the next task of biology: to direct evolution’s progress along the vector of perfection through means of social (not just environmental) reform.

In summary, together agency and increasing complexity shape Julian Huxley’s grades of individuality as placeholders for directional progress toward perfection. That’s in sharp contrast to his grandfather’s views concerning life’s recursive and repetitive cycling. For Julian Huxley biological individuality drove progress of a socially charged variety. The consequences of that view are unpacked next.

1. Julian Towing the Line: Individuality as Progress in the History of Life

There is a tension concerning the role of individuality in the history of life. The preceding essay of this Element outlined Julian Huxley’s views on individuality, that is, the three grades, in contrast to his grandfather’s Sisyphean conception of individuality as representing a continuous, nondirectional cycling. Specifically, the focus was on Julian’s notion of Perfect Individuality as one of progress. However, individuality as a progress-marker in the history of life is not an idea that originated with Julian Huxley – he was towing nearly a century-old line. Asa Gray, in his Reference Gray1861 review of the Origin, tells his readers that,

In other words, “being” as an active quality of organization in nature is individuated by structure, which for Gray was a gradual process often manifest in nested form. He continues,

According to Gray, individuality is the goal that nature struggles to achieve only (and finally) reaching its highest pinnacle in humanity. In other words, for Gray individuality drove ideals of progress in evolution.

As the title of Gray’s review suggests, he was arguing that Darwin’s theory of natural selection is not inconsistent with theology. That organic life strives for individual perfection is a comment Gray makes while trying to convince his readers that Creator origins do not preclude natural order, and that the record of order implies design. Gray states in reference to biblical creation: “the pristine individuals were corporeally constituted like existing individuals, produced through natural agencies” (28, my emphasis).

In other words, to concede that existing individuals were created “after their [pristine] originals” says nothing about what the original types are of, or their mode, or anything else about them. Gray is likely referring to prior views about the independent creation of species; kinds of species were akin to pristine Platonic forms, which are now ruins, mere imperfect caricatures, in their particular instantiations. But what is Gray suggesting about individuality as striving for its perfection?

Gray’s statement, “individuals as created after their kinds” refers to organisms created after their species, but the species are the unchanging and independent categories of nature decreed by God. That is the familiar Great Chain of Being exemplifying an ideological form of progress upward to humans, to angels, to God. Yet Gray is defending Darwin’s work that natural connections among species are material and genealogical. Progress is repurposed: the patterns of striving for individual perfection are rediscovered through ascending gradations; distinguishing lower from higher according to the degree parts are subordinated to a common purpose of being, as he so defined it. And so, for Gray what would achieving Perfect Individuality be like? By ascending nearer to God. Humanity was created through processes of natural selection in His image after all. And since individuality is achieved naturally according to pre-ordained design and marked by the subjugation or dominion over constituents, it marks progress of a very religious (and patriarchal) sort.

Julian Huxley’s view of perfection was not far from Gray’s. In a cringe-worthy passage from Individuality in the Animal Kingdom (Reference Huxley1912, 4 my italics) he states, “In this making of Nature his own, civilized man has an individuality vastly fuller, more perfect, than the savage. Both in resisting adverse forces and in harnessing the indifferent to his will, he is far superior … The gradual increase of independence up from the Protozoa to the highest animals … ” is due to increased independence, complexity, and adaptability. That progress for Huxley is, to modern eyes, eugenically packaged, though without God as the driver despite uncanny similarity to Asa Gray’s rendition of why individuality matters in the history of life.

2. Eugenics & Individuality: A Tension

Julian Huxley’s role as a leading voice of eugenics is complex though: Reference WeindlingWeindling (2012) identifies him as a crucial bridging figure from the old eugenics (i.e., under state control) to the new eugenics attempting to sustain the scientific management of human evolution as socially progressive, rather than fascist offerings of “solutions” to poverty and disease. For Huxley, individuality’s organization was progressively improved through natural selection reaching the pinnacle of a human creature, yet with deficiencies that demonstrated that “unfinished” type (Reference HuxleyHuxley 1962, 123). He states: “the evolutionary biologist can point out to the social scientist and the politician that this importance of the exceptional individual for psychosocial advance is merely an enhancement of a long-established evolutionary trend” (Reference Huxley1962, 129). The 1936 and 1962 Galton Lectures speak for themselves: If Individuality in the Animal Kingdom (Reference Huxley1912) is the theory of progress through perfection of individuality, those Galton lectures are the how-to guides with concrete suggestions for disincentivizing reproduction of “lower quality” individuals and management of both genetics and the environment as a moral (and even religiously-endowed – see Reference HuxleyHuxley 1936, 11) imperative.Footnote ⁵⁹

The foregoing develops the view that individuality was considered an ideological marker of progress in evolution, even if it’s not the kind of “progress” that is appropriate: the ideological undertones are dark. Reference GouldStephen Jay Gould (1989) was concerned with how evolutionary iconography portrayed ideological assumptions exactly like this. Who would have thought that biological individuality had such a significant social role to play? Individuality’s role in ideologically charged views of progress takes a different, but no less surprising turn around the Modern Synthesis of the mid-twentieth century. Individuality in life’s history becomes aligned with the liberal and democratic values at the time. Consider the following contrast.

George G. Simpson (1902–1984), eminent paleontologist and evolutionist, published a paper titled “The role of the individual in evolution” (1941). He criticized analogies drawn between (1) social groups and individuals and (2) individuals and their organ parts, which both converge on totalitarian ideologies (1941, 15).Footnote ⁶⁰ The view of individuals as primarily subordinate units is viewed as dangerous. He expressed grave concern toward entomologists arguing that social insects do not behave for individual satisfaction. Society as a super-organism or “epi-organism” evolving toward greater integration sets the biologist face-to-face with the totalitarian ideal (16). He chastised reckless biologists elevating the super-organism metaphor to the social realm (18). And for scientifically inaccurate representations of individuals as “pawns of fate” designed toward subservience, rather than their inheritance and environment as of their own making (8). Despite critiquing Fascism and National Socialism as ideologies of “progress” through further subjugation of individuals, Simpson praises Julian Huxley’s view as advocating for organisms as more developed than their constituent cells, and that the human society is less individuated than its individual units (19). Individuality in evolution represented progress for Simpson through the furthering of individual satisfaction and breaking free from external control.

And so, in the history of evolutionary thought individuality was viewed as a marker of progress by some serious players – Gray, J. Huxley, and Simpson – who together build a narrative about individuality and evolution’s meaning.

Herein lies a tension, however.

On the one hand, becoming more “individuated,” was considered a marker of liberal progress in contrast to fascist and communist ideologies. Becoming more individuated meant increasing autonomy and control over one’s environment and the overall direction of their life. That was Simpson’s angle.

On the other hand, the aim for individual perfection was tied to the scientific management of human evolution in the form of eugenics practices, which of course manifested as conservative control over bodily autonomy, environmental restrictions, and directing evolution toward a standard of excellence identified only by the elite class under the guise of social reform.

I have no resolution for this tension. Instead, I close with a warning. The historical analyses in this final section are not meant to be whiggish condemnations of past work based on current standards. However, it would be naive and categorically dangerous to assume that social and political motivations did not affect the inner logic of views by prominent figures in the history of biology, even that of Julian Huxley. Taking seriously the role of values in technical scientific reasoning and the inner processes of scientific practice, contemporary presentations of Huxley’s scientific work (or of any other figure in the history of biology) must not be read in a value-free vacuum. Individuality has a dark side, but what is to be done about that? This question warrants future interrogation of the tendency to insulate work of biologists from their broader social meanings and motivations. Biological individuality is a case study setting the foundation for exactly that sort of project. The social and political contexts of biological individuality are embedded within the history of eugenics, contexts which reveal how agency, progress, and managing the direction of human evolution together constitute biological individuality’s dark side. Lest we forget that dark past: history may not strictly repeat itself, but it can, and sometime does, rhyme.