THE SOCIAL ORIGIN OF MULTICELLULARITY

Aggregative Prehistory and the Clonal Transition in the Animal Lineage

Keywords: multicellularity, aggregation, major transitions, clonal bottleneck, epigenetic differentiation, Choanoeca flexa, Capsaspora owczarzaki,scaffold hypothesis, Holozoa, animal origins

Abstract

Animal multicellularity has a social prehistory. The standard account of the major transition treats aggregative and clonal pathways as competing alternatives, crediting clonal development as the exclusive route to animal complexity. We argue this dichotomy is empirically inadequate for the animal lineage. Molecular evidence from unicellular holozoans — the closest living relatives of animals — shows that the developmental toolkit of metazoans was assembled in an aggregative social behavioral context before the clonal transition occurred, and that the regulatory transition at the metazoan origin was not the invention of new toolkit genes but the addition of enhancer-based, position-specific gene expression control onto socially assembled raw material. We call this the Social Scaffold Hypothesis, and distinguish two analytically separate contributions: aggregative behavior assembled the toolkit; the clonal bottleneck resolved the conflict problem that prevented the toolkit from scaling. Epigenetic specialization within clonal bodies instantiates the same organizational principle — identity constituted by relational position — using chromatin machinery that predates multicellularity. Oncogenesis demonstrates that the conflict problem the bottleneck deferred resurfaces as somatic mutation regenerates genetic heterogeneity, and that organisms evolve convergent solutions to a problem the clonal transition only postponed. The paper draws implications for major transitions theory, biological individuality, and the relationship between the two pathways, which this account reconstructs as temporal succession rather than competing alternatives.

1. Introduction

The origin of complex multicellular life from unicellular ancestors is one of the best-studied major transitions in evolutionary history. Two broad mechanisms have been proposed. In clonal development, daughter cells fail to separate after division, forming genetically identical collectives. In aggregative development, independently living cells come together voluntarily, forming collectives of genetically heterogeneous individuals. The dominant view holds that complex multicellularity has arisen exclusively through the clonal route. Aggregative collectives, on this account, have remained facultative, transient, and organizationally simple across half a billion years of evolution, and the standard inference is that genetic heterogeneity sets a hard ceiling on social complexity (Grosberg and Strathmann, 2007; Rokas, 2008). Brunet and King (2017) provide the most detailed current articulation of this framework for the animal lineage, proposing that animal multicellularity arose through modification of pre-existing ECM synthesis and cytokinesis regulation in clonal ancestors, with choanoflagellate rosette colonies as the appropriate model for the early animal condition.

This paper challenges that inference, though not the comparative pattern on which it rests. The pattern is real: no aggregative lineage has achieved animal-grade complexity. But the question of what aggregation was doing in the pre-animal lineage, and what molecular legacy it left behind, has not been adequately addressed in major transitions theory. Recent evidence from unicellular holozoans, the group of organisms most closely related to animals, complicates the picture in ways that matter for how the transition is understood.

Molecular profiling of Capsaspora owczarzaki shows that its aggregative life stage activates gene families whose animal orthologs govern tissue architecture, adhesion, and cell-cell signaling (Sebé-Pedrós et al., 2013). The regulatory genome of Capsaspora encodes these genes under a simpler, promoter-proximal architecture lacking the enhancer elements that permit positionally specific expression in animals (Sebé-Pedrós et al., 2016). Work on Choanoeca flexa documents multicellular sheet formation through aggregative, clonal, or combined mechanisms, with aggregation preceding clonal expansion under ecological stress conditions, and with kin discrimination between genetically distinct strains (Ros-Rocher et al., 2026). A fourth holozoan lineage, Chromosphaera perkinsii, independently forms spatially organized multicellular colonies through cleavage divisions, including symmetry breaking and co-existing distinct cell types (Olivetta et al., 2024), strengthening the inference that social and developmental behaviors were broadly distributed in the unicellular ancestor of animals.

We propose the Social Scaffold Hypothesis to integrate these findings. Its claim is that aggregative social organization in the pre-metazoan ancestor was the evolutionary context in which the molecular toolkit of animal development was assembled, and that the regulatory transition at the metazoan origin was not the invention of new toolkit genes but the addition of a new regulatory architecture onto socially assembled raw material. Crucially, the hypothesis separates two functions that the standard account conflates: aggregative behavior assembled the toolkit, while the clonal bottleneck resolved the conflict problem that prevented the toolkit from scaling. These are analytically distinct contributions from sequential stages of the same lineage, and recognizing their distinctness reframes the relationship between the two pathways from competition to succession. The clonal bottleneck did not create the toolkit. It removed the one constraint preventing the toolkit from being exploited at greater complexity.

The paper is organized as follows. Section 2 reviews the two pathways and the aggregative complexity ceiling. Section 3 develops the molecular evidence for the scaffold relationship. Section 4 examines epigenetic specialization in clonal bodies as an internal instance of social dynamics. Section 5 addresses oncogenesis as the re-emergence of the conflict problem within clonal organisms. Section 6 draws implications for major transitions theory and biological individuality. Section 7 addresses objections.

2. Two Pathways and a Persistent Ceiling

2.1 The Standard Dichotomy

Grosberg and Strathmann's (2007) influential review argued that the single-cell bottleneck was the key innovation enabling complex multicellularity. By resetting genetic composition at every generation, the bottleneck maximizes within-group relatedness, aligns cellular and organismal fitness, and suppresses the within-collective conflict that would otherwise favor selfish lineages over collective function. The comparative support for this argument is strong. All five eukaryotic lineages that independently achieved complex multicellularity, including animals, land plants, fungi, red algae, and brown algae, use clonal development as their primary organizational mode.

The aggregative lineages follow a different trajectory. Dictyostelium discoideum and its relatives in the Amoebozoa have practiced aggregative multicellularity for at least 600 million years. Under starvation stress, thousands of independent amoebae aggregate, differentiate into stalk cells that die without reproducing and spore cells that disperse, producing a coordinated fruiting body from genetically heterogeneous individuals. Myxobacteria show functionally analogous organization. Neither lineage, across half a billion years, has approached animal-grade complexity.

2.2 The Conflict Ceiling

The standard explanation for the aggregative ceiling is the cheater problem. Because aggregating cells are genetically non-identical, any lineage that preferentially occupies reproductive roles has higher fitness than cooperating neighbors. This produces persistent evolutionary pressure toward defection that clonal collectives do not face. Dictyostelium has evolved kin discrimination to manage this pressure. Two transmembrane proteins, TgrB1 and TgrC1, function as a molecular recognition system enabling cells to identify kin and preferentially aggregate with them (Hirose et al., 2011). Kin recognition constrains cheating but does not eliminate it. Cheater strains that circumvent recognition are routinely found in natural populations, and the arms race between recognition and circumvention is ongoing.

The management ceiling that results is structural, not contingent. The more complex the collective organization, the more internal conflict it can generate, and the more resources must be allocated to conflict management rather than collective function. Boomsma (2009) drew an instructive parallel: the single ancestral mating event in the Hymenoptera maximized within-colony relatedness to r = 0.75 between full sisters in haplodiploid species, approaching but not achieving the full fitness alignment of clonal development through a single-cell bottleneck. The comparison clarifies what the bottleneck uniquely accomplishes. Monogamy reduces heterogeneity at the founding event but cannot eliminate it; the bottleneck eliminates it entirely by passing all genetic information through a single cell. It should be noted that the relationship between high relatedness and eusociality in the Hymenoptera has been contested following Nowak, Tarnita, and Wilson (2010), and the paper's use of this parallel is for the structural point about bottlenecks rather than as an endorsement of any position in that debate. What the aggregative lineages lack is not social organization but the structural reset of a clonal bottleneck, and no degree of elaboration of social conflict management substitutes for it. This is precisely what makes the scaffold hypothesis interesting: if the molecular toolkit was assembled in the aggregative context, the bottleneck's contribution was not to provide raw material but to remove the one constraint preventing that material from scaling.

The most sustained early argument that bounded individuality had to be evolutionarily won against persistent cellular conflict was made by Buss (1987), who showed that major features of animal developmental architecture — germ-cell sequestration, determinate cleavage, the early restriction of somatic cell fate — are best understood as adaptations evolved to suppress within-organism cell-lineage selection. On Buss’s account, the metazoan individual is not a starting condition but an evolutionary product, continuously re-manufactured by developmental systems whose function is as much conflict suppression as construction. The scaffold hypothesis extends this logic one stage earlier: if the bottleneck and its attendant developmental architecture were required to suppress conflict that had already been assembled, the pre-bottleneck lineage must have been one in which that conflict-generating molecular toolkit was elaborated. Buss identifies when the solution evolved; the scaffold hypothesis asks what the problem was built from.

2.3 The Puzzle the Standard Account Leaves Open

The standard account generates a clear prediction: complex multicellularity should arise from lineages with minimal social aggregative behavior, since genetic heterogeneity sets a ceiling on collective complexity. The more sophisticated a lineage's aggregative social organization, the more resources are consumed in conflict management and the less are available for collective functional elaboration. On this account, the pre-animal ancestor should have been primarily clonal in its multicellular tendencies, and aggregative behavior should either be absent from the holozoan clade or represent an independent, derived development within specific lineages after the animal transition.

The holozoan evidence does not fit this prediction. Aggregative behavior using the same molecular language as animal development is documented in at least five unicellular holozoan lineages, all of which are phylogenetically positioned as outgroups to animals and therefore represent ancestral conditions rather than derived ones. Capsaspora owczarzaki activates tyrosine kinase, integrin, and Hippo pathway genes during voluntary social aggregation. Choanoeca flexa forms multicellular sheets through aggregative processes with kin discrimination comparable to Dictyostelium's recognition systems. A third independent instance comes from Ministeria vibrans, a marine free-living filasterean that Li, Dharamshi et al. (2025, preprint) report forms stable homogeneous aggregates and deploys a broad set of animal multicellularity genes — cell adhesion, signaling, and transcriptional regulators — during the process, supporting the inference that the last unicellular ancestor of animals had the capacity to aggregate using the key genes of animal development. Two further lineages extend the distribution: Pigoraptor vietnamica and Pigoraptor chileana, novel predatory filastereans described by Hehenberger et al. (2017), form multicellular clusters as part of complex life histories and encode nearly complete integrin adhesomes including the IPP complex, scaffolding proteins, and receptor tyrosine kinases. Tikhonenkov et al. (2020) showed that Pigoraptor aggregations are driven by active chemical attraction and serve cooperative foraging on large eukaryotic prey — identifying predation pressure as a possible ecological driver of the behaviors in which the toolkit was assembled, one that complements intraspecific coordination as a selective context. The ecological selective pressures operating on the pre-metazoan ancestor were therefore multiple and convergent: cooperative foraging on large prey rewarded collective action; size-dependent predator evasion favored aggregation over solitary living; and environmental stress responses — including starvation and osmotic challenge — reliably induced the transition from dispersed to collective states. Each of these pressures would have selected for the molecular machinery of cell-cell adhesion, coordination, and collective responsiveness independently, and the holozoan toolkit accumulated at their intersection. What the scaffold hypothesis adds to this ecological picture is a molecular consequence: the genes elaborated for these collective functions are the same genes animal development would later co-opt, and their functional integration within the aggregative context — not merely their presence — is what made them useful raw material for the regulatory transition. The phylogenetic spread of five lineages within the holozoan clade — three filastereans, a choanoflagellate, and members of the newly characterized Pluriformea — is more consistent with an ancestral social condition than with independent derivation in each lineage. Deep-time corroboration comes from the Torridonian fossil Bicellum brasieri (approximately one billion years old), which displays two distinct cell types in spherical aggregates consistent with holozoan affinity and cell sorting topologically similar to cadherin-mediated adhesion (Strother et al., 2021), placing cell-type differentiation in the holozoan lineage more than 400 million years before animals. This is not a complication at the margins of the standard account — it is in direct tension with its core prediction about what the pre-animal ancestor should look like. The question the evidence forces is not how the aggregative behavior of holozoan relatives is compatible with the clonal transition but why the lineage destined for the clonal transition was, at the point of that transition, saturated with aggregative social behavior and the molecular tools it had assembled.

The scaffold hypothesis answers that question by separating two functions that the standard account conflates. Aggregative social behavior assembled the molecular toolkit. The clonal bottleneck resolved the conflict problem that prevented the toolkit from scaling. These are analytically distinct contributions, and recognizing their distinctness reframes the relationship between the two pathways: not competition between alternatives, but sequential stages of a single lineage, each doing different work toward the eventual outcome. What Section 3 must then show is whether the molecular evidence from unicellular holozoans actually supports the assembly claim — and whether the depth and integration of the toolkit overlap is sufficient to distinguish retention from convergent recruitment.

3. The Social Scaffold: Molecular Evidence

3.1 Capsaspora owczarzaki and the Pre-Assembled Toolkit

Capsaspora owczarzaki is a unicellular holozoan that lives as a symbiont in the haemolymph of the freshwater snail Biomphalaria glabrata. Its life cycle includes a unicellular filopodial stage, an aggregative stage in which cells form a colony with extracellular matrix, and a dormant cyst stage. Sebé-Pedrós et al. (2013) performed comparative transcriptomics across these life stages and found that the transition to aggregation is associated with significant upregulation of tyrosine kinase signaling components, integrin-mediated cell adhesion machinery, and elements of the Hippo pathway, the same gene families that coordinate tissue growth, cell adhesion, and multicellular patterning in animals. The authors note explicitly that the first multicellular animals could have recycled the genes controlling aggregation in their unicellular ancestors, a formulation that anticipates the scaffold hypothesis.

The scaffold hypothesis rests on a specific version of this claim, and the distinction matters. The argument is not simply that these gene families happen to be expressed during aggregation and happen to be useful in development — a coincidence that could reflect convergent recruitment of the most available molecular tools. The argument is that these genes were selected and functionally elaborated in the social aggregative context, and that the metazoan transition co-opted them with a new regulatory architecture rather than inventing equivalent tools from different raw material. The evidence that supports this stronger reading is the depth of the toolkit overlap. The shared gene families are not isolated adhesion molecules but functionally integrated signaling systems, including complete pathway architectures, whose animal versions retain the same interaction logic as their holozoan counterparts. Independent recruitment of entire integrated pathways for analogous functions is considerably less parsimonious than retention and regulatory repurposing of the same network.

That parsimony argument has been substantially strengthened by recent functional genetics. Phillips et al. (2022) developed CRISPR genome editing techniques for Capsaspora and used them to characterize the function of coYki, the Capsaspora ortholog of the Hippo pathway effector YAP/TAZ/Yorkie, a central regulator of tissue growth and a clinical oncogene in animals. The result was decisive: coYki loss-of-function mutants showed severe defects in the three-dimensional morphology and cytoskeletal organization of Capsaspora aggregates, with normal proliferation rates throughout. In animals, YAP/TAZ drives growth. In Capsaspora, it shapes multicellular aggregates and has no proliferative role. The ancestral function was aggregative morphology; the growth-regulatory role was acquired later, in the animal lineage. Phillips and Pan (2024) extended this analysis to the upstream Hippo kinase cascade: loss of the Hippo and Warts kinases produced identical cytoskeletal and aggregate-packing phenotypes and no proliferative effects, confirming that cytoskeletal regulation of aggregative architecture was the ancestral function of the entire pathway. This is functional evidence — not expression evidence — of a complete developmental signaling pathway being specifically elaborated for aggregative social function before the metazoan transition. It is the strongest available evidence that the toolkit genes were not merely present and available for convergent recruitment but were actively shaped by selection for aggregative behavior, and that their animal function represents derived co-option of an ancestrally social molecular system.

What would further confirm the stronger causal claim is evidence that Capsaspora's toolkit genes carry regulatory signatures of positive selection specifically in the aggregative context — not merely that they are expressed during aggregation, but that they were shaped by selection for aggregative function rather than retained under relaxed constraint. The Phillips findings provide strong functional support for this interpretation in the case of the Hippo pathway specifically, but the full toolkit encompasses integrin-ECM and tyrosine kinase networks whose ancestral functions in Capsaspora are not yet characterized at comparable genetic resolution. The hypothesis makes a distinct and testable prediction that extends beyond what current data adjudicate: a systematic demonstration that the interaction network topology of these pathways in Capsaspora mirrors the ancestral form of their animal counterparts, rather than a simpler or independently assembled configuration, would further distinguish retention from convergent recruitment. This gap should be understood as a research program the hypothesis generates, not as evidence against it — but it must be stated clearly.

The genome-level picture from Suga et al. (2013) supports the same interpretation. Capsaspora encodes a near-complete complement of animal multicellularity genes, organized in regulatory networks activated during the aggregative life stage. What the organism lacks is not the toolkit genes but the regulatory architecture that deploys them with the spatial precision animal development requires. Sebé-Pedrós et al. (2016) characterized this regulatory architecture directly. The Capsaspora genome encodes its developmental toolkit under small, promoter-proximal regulatory elements that respond to global environmental signals such as starvation and chemical inducers. The enhancers found in animals, which are distal regulatory elements capable of driving gene expression in specific cell types at specific developmental stages, are absent. The genome of Capsaspora is a toolkit without a positional operating system.

The regulatory gap between Capsaspora and animals is therefore larger than a simple gene count comparison suggests, but it is also more informative. It is not a gap in molecular raw material. It is a gap in the architecture that reads positional information and translates it into differential gene expression. This account locates the metazoan transition in the acquisition of that architecture, not in the prior assembly of the tools it reads.

3.2 Choanoeca flexa and the Living Intermediate

Choanoflagellates are the closest known unicellular relatives of animals. Among them, Choanoeca flexa is the most directly relevant. Brunet et al. (2019) showed that C. flexa forms multicellular cup-shaped sheets whose cells undergo coordinated inversion from concave to convex in response to darkness, driven by actomyosin contractility. The actomyosin machinery responsible is conserved with the morphogenetic machinery animals use for epithelial folding and tissue remodeling. This finding demonstrates that mechanical, tissue-level morphogenetic tools were present and functional in the choanoflagellate ancestor, extending the toolkit pre-assembly argument from cell adhesion and signaling genes to the physical machinery of morphogenesis.

More directly relevant to the argument here, Ros-Rocher et al. (2026) report that C. flexa multicellular sheets form through purely clonal processes, purely aggregative processes, or a combination of both depending on environmental conditions, specifically salinity levels in coastal splash pools that undergo repeated evaporation and rehydration cycles. Under rehydration, cells aggregate first and begin dividing second. The social assembly occurs under ecological stress; the clonal expansion consolidates the structure afterward. This sequence is what the scaffold hypothesis predicts in a lineage that has not yet committed fully to either mode. Sheets capture bacterial prey at twice the per-cell rate of dissociated single flagellates, providing a concrete selective rationale for why multicellularity is maintained in this organism.

The same study documents kin discrimination in C. flexa: cells from genetically distinct strains preferentially associate with their own strain when mixed. This places C. flexa in the same functional space as Dictyostelium's TgrB1/TgrC1 recognition system, and the co-occurrence of kin discrimination with retained clonal capacity identifies it as an organism navigating exactly this transition — one in which social conflict management and clonal structural solutions evolved in the same lineage simultaneously, neither yet displacing the other.

A fourth holozoan lineage provides a different kind of evidence. Chromosphaera perkinsii is an ichthyosporean, a group phylogenetically distinct from choanoflagellates and separated from them by substantial evolutionary distance within the holozoan clade. Olivetta et al. (2024) showed that Chromosphaera undergoes symmetry breaking and develops through cleavage divisions to form spatially organized colonies with distinct co-existing cell types. Chromosphaera is not aggregative — its relevance is not as another instance of the social pathway. Its relevance is as evidence against the convergence objection to the scaffold hypothesis. If the developmental toolkit genes shared between Capsaspora's aggregative stage and animal development were convergently recruited rather than homologously retained, one would not expect to find independent clonal multicellularity with genuine cell-type differentiation arising in a phylogenetically distant holozoan lineage using the same toolkit. The recurrence of differentiated multicellularity across the holozoan clade, from Capsaspora to Choanoeca to Chromosphaera, is more parsimoniously explained by an ancestral holozoan endowment of the toolkit than by repeated convergent recruitment. The Chromosphaera finding strengthens the toolkit distribution argument without adding another instance of the aggregative pathway.

3.3 The Regulatory Transition: A New Operating System on Old Hardware

The mechanistic question this account must answer is what specifically changed at the metazoan transition, given that the toolkit genes were already present and functional. Sebé-Pedrós et al. (2016) provide the most direct answer available. The transition from Capsaspora-type regulation to animal-type regulation involved the expansion of distal regulatory elements, enhancers, that can drive gene expression in a cell-type-specific and position-dependent manner. These elements read the combinatorial transcription factor environment of each cell, itself a product of the cell's position in the developing body, and translate it into a specific gene expression program. The same toolkit gene can be expressed in one spatial domain and silenced in an adjacent one, enabling the generation of distinct cell types from a uniform genomic substrate.

Complementary evidence comes from chromatin evolution. Grau-Bové et al. (2022) surveyed chromatin writer, eraser, and remodeler enzyme repertoires across eukaryotic diversity and found that the core machinery for placing and removing histone modifications is ancient, conserved across all eukaryotic lineages surveyed, and does not expand significantly at the multicellular transitions. What does expand markedly in lineages achieving complex multicellularity, including animals, streptophyte plants, and brown algae, is the repertoire of chromatin reader proteins: the domains that interpret histone modification states and translate them into transcriptional decisions. The cell's capacity to write contextual information onto its chromatin is phylogenetically old. What complex multicellularity added was an expanded vocabulary for reading that information and acting on it in cell-type-specific ways.

The picture that emerges from these two bodies of work is consistent. The pre-animal ancestor possessed the molecular toolkit, the chromatin writing machinery, and a basic capacity for contextual responsiveness. The metazoan transition added the regulatory architecture, enhancers on the DNA side and expanded reader domains on the chromatin interpretation side, that transformed a globally responsive system into a positionally specific one. Single-cell transcriptomic comparisons across early-branching animals confirm that this regulatory expansion scales with cell-type diversity: ctenophores, which have the most diverse cell types among the animals surveyed, show the strongest enrichment for distal regulatory elements relative to simpler animals whose regulation remains closer to the promoter-proximal mode of Capsaspora (Sebé-Pedrós et al., 2018) — a correlation consistent with the hypothesis that enhancer complexity enables cell-type diversity, though the causal direction is not definitively established by that data alone. Social responsiveness, the capacity to change gene expression in response to signals from neighboring cells, became developmental responsiveness, the capacity to acquire a stable, heritable cell-type identity from positional context within a growing collective. Coyle and King (2025) provide a comprehensive recent survey of regulatory transitions at the animal stem, confirming that developmental pathways in animals were built on core mechanisms inherited from protistan ancestors — a conclusion directly supportive of the scaffold hypothesis framing. They note, however, that whether distal regulatory elements truly emerged only in animals remains uncertain given incomplete sampling across unicellular holozoan diversity, a caution worth retaining: the current evidence from Capsaspora (Sebé-Pedrós et al., 2016), S. rosetta chromatin accessibility (Gahan et al., 2025), and three-dimensional genome architecture across multiple holozoan species (Kim et al., 2025) consistently finds no distal elements or chromatin loops, but the sample is not exhaustive.

Recent work at the level of three-dimensional genome architecture adds a second layer to the regulatory transition argument. It is not simply that Capsaspora lacks animal-type enhancers. It also lacks the physical genome folding mechanism that makes enhancer function possible: Kim et al. (2025) found that chromatin looping — the spatial contact between distal regulatory elements and their target promoters — is absent in unicellular holozoan relatives of animals and emerged specifically at the animal stem. The regulatory transition was therefore two steps, not one. The unicellular ancestor had the toolkit genes under simple promoter-proximal regulation and lacked both the distal regulatory elements and the genome architecture that would allow those elements to operate. Animal development required the assembly of both. This sharpens the account of what changed at the metazoan transition and what did not: the molecular tools were present; the regulatory reading apparatus was the innovation.

4. Epigenetic Specialization and the Internal Social Dynamic

4.1 Identity Without Genetic Difference

The preceding sections address how the developmental toolkit was assembled historically, at the phylogenetic scale, through aggregative social behavior in the pre-animal ancestor. A separate question operates at the organismal scale: once a complex clonal body exists, how does it generate hundreds of distinct, stable cell types from a single genome? This section argues that the mechanism, epigenetic differentiation, instantiates a structural logic similar to the one identified in the historical scaffold argument. The claim is not that epigenetic specialization is historically continuous with ancestral social dynamics — that is a stronger claim than the evidence supports — but that it operates by the same organizational principle: identity constituted by relational position rather than intrinsic properties. The molecular basis for treating this as more than structural analogy is that the chromatin machinery executing positional cell-fate determination is the same machinery that predates multicellularity and was already operating in social aggregative contexts in the unicellular ancestor — a shared molecular substrate, not merely a shared metaphor. The specific mechanism in each case differs enough to require that the connection be stated carefully, and the full treatment of the objection is deferred to Section 7.4.

The empirical foundation for this argument was laid in 1891, when Hans Driesch separated the first two blastomeres of a sea urchin embryo and found that each isolated cell developed into a complete, smaller larva rather than the predicted half-organism. The result directly falsified the prevailing mosaic model, in which each cell was presumed to carry only the developmental instructions for its own eventual fate. Driesch’s conclusion — that “the fate of a cell is a function of its position in the whole” — remains the foundational statement of the positional identity principle in developmental biology (Driesch, 1892). He called the embryo a “harmonious equipotential system”: every component has the full potential of the whole, and actual fate is determined not by intrinsic cellular properties but by location in the collective. Driesch eventually retreated into vitalism because he could not identify the mechanism by which position was read and fate specified. That mechanism is now understood: the chromatin machinery described below translates the signaling history accumulated from neighboring cells into stable, heritable gene expression states. What Driesch perceived as a mysterious positional force is the epigenetic state of the cell — its chromatin landscape shaped by the molecular signals it has received from its location in a growing body.

Every somatic cell in a mammalian body carries a complete copy of the genome. A hepatocyte encodes everything needed to be a neuron; a smooth muscle cell encodes everything needed to be an erythrocyte. Specialization is not achieved by distributing different genetic information to different cells. It is achieved by differentially silencing and activating the same information according to each cell's developmental history and position within tissue architecture. Histone modification, DNA methylation, chromatin remodeling, and non-coding RNA regulation together establish stable, mitotically heritable epigenetic states that determine which portions of the genome are accessible for transcription in a given cell.

These epigenetic states are written by signals from neighboring cells, by morphogen gradients encoding positional information, and by the signaling history the cell has accumulated through development. A cell's fate is a function of where it is and what its neighbors have told it. There is no intrinsic, context-independent property of the cell that determines whether it becomes a neuron or a glia. Cell identity is constituted relationally, by position in a social topology.

4.2 Ancient Machinery, New Function

The chromatin machinery underlying this positional constitution of identity is not an animal innovation. As Grau-Bové et al. (2022) document, the core writer-eraser-remodeler apparatus for histone modification is universal across eukaryotes, present and functional in unicellular lineages where it serves functions including stress response, cell cycle regulation, and transposable element suppression. Wang et al. (2021) showed that Dictyostelium, during transitions between its unicellular and aggregative multicellular states, undergoes significant changes in H3K4 methylation and H3K27 acetylation patterns at loci specifically active during the multicellular phase. Dictyostelium is an amoebozoan, phylogenetically distant from the holozoan lineage, and this evidence does not establish that the holozoan ancestor specifically used chromatin modification for social state-tracking. Holozoan-specific chromatin data is now available, however. Gahan et al. (2025) profiled chromatin modification states across the life cycle of Salpingoeca rosetta, the best-characterized unicellular choanoflagellate, and found that H3K27me3, deposited by Polycomb Repressive Complex 2, marks cell-type-specific genes in different life cycle states in a manner directly analogous to its silencing function in animal cell types. The same histone mark that in animals suppresses genes inappropriate for a given cell identity is already present and life-cycle-coupled in the closest unicellular relative of animals. Two qualifications constrain the scope of this finding. First, the chromatin accessibility profile of S. rosetta remains exclusively promoter-proximal, with no distal regulatory elements detected — independently corroborating, at the epigenetic level, the regulatory gap described in Section 3.3. Second, Capsaspora has secondarily lost both H3K9 and H3K27 methylation machinery, so the Gahan et al. (2025) data reflects the ancestral holozoan condition rather than the specific chromatin mechanism of Capsaspora itself, which likely operates through alternative modifications. The Dictyostelium evidence establishes that the general capacity for epigenetic responsiveness to collective transitions predates complex multicellularity — consistent with, though not proof of, an ancestral holozoan version of the same capacity — and the S. rosetta data now provides a holozoan-specific instance of that capacity operating at the level of histone modification.

On that qualified basis, the connection to the historical scaffold argument holds: the metazoan regulatory transition likely built on chromatin machinery that was already functionally coupled to collective behavioral states, whether or not the precise mechanism is recoverable from the Dictyostelium comparison. The expanded reader repertoire that Grau-Bové et al. (2022) identify as the animal-specific innovation made this contextual responsiveness positionally precise — the same chemical modifications driving different outcomes in cells occupying different positions. This reader expansion is the epigenetic correlate of the regulatory transition described in Section 3.3: just as the metazoan genome acquired distal regulatory elements and chromatin looping to translate positional transcription factor environments into cell-type-specific gene expression, it acquired the expanded chromatin reader vocabulary to translate histone modification states into stable, position-specific cell identities. The two transitions are two sides of the same innovation. The writers and erasers — the machinery for placing and removing histone marks — were already present in unicellular ancestors and were already coupled to collective behavioral states. What the animal stem added was not the capacity to write contextual information onto chromatin but the expanded vocabulary for reading and acting on that information in a position-dependent way. The operating system metaphor of Section 3.3 applies here directly: the pre-animal ancestor had the hardware and the data, but the animal-specific reader expansion was the software that could interpret them with positional precision. The internal social dynamic of the clonal body is built from ancient machinery redirected, not invented from scratch.

4.3 Specialization as Social Position

The practical consequence is that in a complex multicellular organism, cell fate is equivalent to social position. Two cells with identical genomes become different things because they occupy different locations in a network of signaling relationships. Disrupt the network by changing which cells are adjacent, altering a morphogen gradient, or blocking a juxtacrine signal, and cell fates change accordingly. The organism is not a collection of pre-specified cell types assembled in space. It is a relational topology of signaling relationships that generates cell types as emergent properties of position.

This is the central mechanism of developmental biology, not a peripheral observation about it. The major conceptual frameworks of the field, including Spemann's organizer concept, French flag positional information models, and lateral inhibition patterning, all describe how positional relational context generates distinct identities from a uniform genomic substrate. In this sense, the multicellular animal body is organized socially from within, by the same fundamental logic that organized the aggregative ancestor from without. The scaffold was replaced structurally but its organizational principle persisted.

The theoretical framework for this picture was sketched in advance of its molecular details by Waddington (1957), who proposed that differentiating cells move through a topographic landscape of branching developmental valleys — “chreodes” — whose stability is maintained against perturbation by the underlying regulatory network. A cell committed to a given fate occupies a deep basin in this landscape, and normal development is canalized: the same cell type emerges repeatedly despite variation in input signals. Kauffman (1993) formalized this intuition computationally, showing that gene regulatory networks of sufficient complexity spontaneously generate a small number of stable attractor states corresponding to distinct cell types, and that the number of attractors scales with genome size in a way that matches observed cell-type numbers across organisms. On this formalization, cell-type identity is an attractor in regulatory network dynamics, not the product of any single molecular component. The epigenetic machinery described in Section 4.2 — the chromatin writers, erasers, and readers, and the expanded reader repertoire that is the animal-specific innovation — is the mechanism that establishes and maintains these attractors: that deepens the developmental valleys and keeps the rolling marble from escaping. The stable, mitotically heritable cell identities that result are what Driesch called position-determined fates and Waddington called chreodes; they are what the molecular evidence shows to be chromatin states constituted by positional signal history. Section 5 will show what happens when the attractor fails.

5. Oncogenesis and the Return of Conflict

If the clonal bottleneck permanently resolved the conflict problem by establishing genetic identity among all collective members, social conflict dynamics should not recur within clonal bodies. Somatic mutation falsifies this expectation. Dividing cells accumulate mutations over an organism's lifetime, generating genetic divergence between daughter lineages. Cells whose mutation history includes loss of tumor suppressor function or activation of proto-oncogenes gain a proliferative advantage over their neighbors. In the structural terms of aggregative multicellularity, these are cheaters: cells that defect from the cooperative constraints of the collective to pursue a locally selfish replicative strategy, at the expense of collective function.

The organism's response to this re-emerging conflict is organized identically to the conflict management systems of aggregative collectives, implemented in different molecular substrate. TP53, a damage-response integrator that triggers apoptosis or senescence in cells experiencing DNA damage, oncogenic stress, or hypoxia, functions as a general gatekeeper against aberrant proliferative states; RB1 restricts cell cycle entry and suppresses progression past the G1 checkpoint, preventing cells from entering replication inappropriately. The functional parallel is to worker policing in eusocial insect colonies, where non-reproductive workers that attempt to lay eggs are attacked and their eggs destroyed by other workers.

The same structural logic operates at the level of immune surveillance and cell-contact signaling. The immune system surveils for cells displaying abnormal surface markers, identifying and destroying incipient tumor cells, as kin recognition in Dictyostelium excludes genetically foreign cells from the aggregate. Contact inhibition prevents cells from proliferating when surrounded by normal neighbors, as pheromonal suppression prevents reproductive development in worker insects. These are distinct mechanistic implementations of the same solution type: detection of deviation from cooperative constraint, followed by suppression or elimination of the deviating unit.

The parallel is not homology — tumor suppressor networks and insect worker policing share no common ancestral mechanism — nor is it mere analogy, which would make the comparison merely descriptive. It is structural convergence: when any collective of replicating units faces the same formal selection problem — maintaining cooperative integration against individually advantageous defection — a predictable solution space exists, and independent lineages find it. The claim is causal, not rhetorical: the same fitness logic produces similar regulatory architectures across unrelated lineages because the selective pressure is formally identical, not because the mechanisms share history. The consequence for the present argument is significant. The clonal bottleneck does not permanently solve the conflict problem. It defers it by eliminating the initial genetic heterogeneity that makes defection individually advantageous. Complexity at scale regenerates that heterogeneity through somatic mutation, and the conflict dynamics of the aggregative ancestor re-emerge within the body that had structurally escaped them. The organism then evolves, over evolutionary time, convergent solutions to a problem the bottleneck only postponed. Molecular evidence supports this characterization directly. Trigos et al. (2017) analyzed transcriptomes from 3,473 tumor samples across seven solid cancer types and found that tumors universally show systematic upregulation of genes shared with unicellular organisms alongside downregulation of genes that emerged specifically in metazoans. The transcriptomic age index of cancer cells shifts toward the unicellular ancestral condition. This is not a structural parallel between cancer and aggregative dynamics but direct evidence that oncogenesis involves the reactivation of unicellular gene expression programs as multicellular regulation fails. The atavism is molecular: the same gene networks that unicellular holozoan ancestors expressed during their free-living state are preferentially upregulated in cells that have lost the conflict-management architecture of the clonal body.

The phenomenon Trigos et al. document is atavism in the classical sense — not Haeckelian recapitulation, by which development adds new terminal stages that replay evolutionary history, but the de-repression of an ancestral regulatory program when the derived controls overlying it are removed (Gould, 1977). This reading connects the scaffold hypothesis to a distinct theoretical tradition in cancer biology. Davies and Lineweaver (2011) proposed that cancer tumors represent a reversion to ancestral unicellular or early multicellular gene expression programs, which they termed the “atavistic model” of cancer, and predicted that systematic transcriptomic comparisons would reveal phylogenetically ancient gene signatures preferentially active in tumors. Trigos et al. (2017) provide precisely that confirmation. The scaffold hypothesis supplies what the atavistic model requires but does not provide: a mechanistic historical account of what those ancestral programs were doing before the metazoan transition, and why they remain molecularly accessible in cells that have shed their multicellular regulatory controls. The atavism is not metaphorical but stratigraphic — a literal expression of evolutionary history layered into the genome, with the ancestral program persisting beneath the derived one. Gould distinguished recapitulation (adding ancestral stages to the end of development) from paedomorphosis (retaining ancestral juvenile features in derived adults); the Trigos finding is the regulatory analog of neither, but of a third possibility: the reactivation of an evolutionary precursor’s gene expression program when the multicellular regulatory architecture that suppressed it is dismantled by mutation. The tumor cell is not recapitulating by adding unicellular stages to its developmental trajectory. It is reverting — as the derived regulatory landscape is degraded — to an ancestral attractor state in gene regulatory network dynamics, the basin from which multicellularity lifted the lineage and to which cells return when the mechanisms sustaining the derived state fail. This framing clarifies why the Trigos finding is not merely an analogy between cancer and aggregative dynamics but a structural consequence of the scaffold relationship: the ancestral unicellular program was not erased by the metazoan transition but overwritten, and it resurfaces when the overwriting wears thin. The palimpsest metaphor of Section 6.3 is here made molecular.

This recurrence also illuminates the evolutionary arms race between oncogenes and tumor suppressor networks. Standard accounts frame this as a conflict between levels of selection, with cell-level selection favoring proliferation against organism-level selection favoring restraint. This framing adds a temporal dimension. This conflict is the re-emergence, within the body, of the same cheater-policing dynamics that governed the aggregative ancestor. The body contains, as a permanent structural tension, the same organizational challenge the clonal bottleneck was adopted to resolve.

6. Discussion

6.1 Revising the Major Transitions Framework

Maynard Smith and Szathmáry (1995) treated the clonal and aggregative routes to multicellularity as alternatives, crediting the clonal route with the successful transition; the specific mechanism enabling that success, the single-cell bottleneck, was most fully articulated by Grosberg and Strathmann (2007). This account does not dispute the outcome but offers a different account of how the successful route was reached. If the molecular toolkit of animal development was assembled during social aggregation in the pre-metazoan ancestor, the two routes are not independent alternatives that competed. They are stages in a temporal sequence, and understanding animal development requires understanding its aggregative prehistory.

The practical implication is that the most informative comparison for understanding animal origins is not animal versus non-animal but aggregative versus developmental expression of the same toolkit genes in unicellular holozoans. The field has begun moving in this direction, particularly following the Capsaspora and C. flexa studies, but the theoretical framing of what that comparison is looking for matters. If this account is correct, what the comparison should reveal is not independent recruitment of conserved genes but homologous expression of the same regulatory network in the ancestral social context and the derived developmental context.

A further implication follows from Queller’s (1997) distinction between fraternal and egalitarian major transitions. Fraternal transitions occur between genetically similar units — clonal kin whose shared identity aligns fitness and suppresses conflict; egalitarian transitions occur between unlike units whose cooperation pays through functional complementarity rather than kinship. Animal multicellularity is the paradigm case of a fraternal transition: the clonal bottleneck maximizes within-group relatedness, and developmental architecture suppresses the cell-lineage selection that would otherwise undermine it. The aggregative scaffolding stage described here fits neither category cleanly. The unicellular holozoan relatives of animals aggregate with conspecifics — the interaction is fraternal in kind — but with genetically heterogeneous partners, so the fitness alignment of true fraternal transitions is absent and the conflict ceiling persists. Major transitions theory currently lacks a well-developed category for this intermediate: cooperative social behavior among genetically non-identical conspecifics, deploying shared molecular tools in service of collective function, but without the structural reset that would make that function scale indefinitely. The scaffold hypothesis names what this intermediate contributed: not a completed major transition, but the assembly of the molecular toolkit that the completed transition would later deploy. The clonal bottleneck did not create a new toolkit; it fraternalized an existing one.

6.2 The Derived Status of Biological Individuality

A persistent assumption in evolutionary biology treats sociality as an anomaly requiring explanation against a background of competitive, asocial individuals. The Hamilton-Trivers tradition frames cooperation as a problem: given that natural selection favors individual fitness, how does cooperative behavior arise and persist? The scaffold hypothesis inverts the question for the animal lineage. Social aggregation is ancestral. The molecular toolkit for cooperation, adhesion, and collective coordination was assembled before clonal development, in organisms practicing voluntary social aggregation as a primary ecological strategy. The bounded, clonal individual is the derived state, produced by a specific structural solution to a specific conflict problem. What requires explanation is not the emergence of social behavior from asocial individuals but the emergence of stable individuality from a prior social substrate.

This inversion has antecedents that predate the molecular evidence by more than a century. Virchow (1858), whose cellular pathology established the cell as the fundamental unit of both life and disease, described the healthy organism as a “cellular democracy” — a republic of cells in which each member retains equal viability, and in which disease arises not as a curse on the whole but as a disruption of the cooperative civic order among cellular participants. Virchow’s political framing was explicitly inverted from the reigning atomism: the body is not an independent individual containing subordinate parts, but a collective whose coherence is achieved by the ongoing civic behavior of its components. The present account vindicates this framing mechanistically. The bounded animal individual is not the starting condition from which social behavior must be explained but the end product of a molecular history in which the social organization came first and the clonal individual was constructed on top of it. Buss (1987) made the same point from the developmental direction: the architecture of animal development — germ-cell sequestration, cleavage restrictions, the early silencing of alternative cell fates — is an evolved system for managing cellular conflict, not a neutral scaffold for constructing tissues. The individual animal body is, on both accounts, less an atom that entered into social relations than a society that achieved the appearance of individuality through structural conflict management. The molecular evidence from holozoan unicellular relatives of animals now provides the specific history of how that achievement was staged.

This inversion does not dissolve the questions the Hamilton-Trivers tradition asks. Kin selection, reciprocity, and multi-level selection remain valid frameworks for analyzing cooperation among pre-formed individuals. The inversion applies specifically to the question of origins: for the animal lineage, the tools of collective organization are phylogenetically prior to the bounded individual, and major transitions theory should account for that priority.

6.3 Biological Individuality and the Temporal Dimension

Godfrey-Smith (2009) proposed that individuality is a graded property defined by the degree to which a collective suppresses within-collective competition and becomes the primary unit of selection. Queller and Strassmann (2009) operationalized this with a cooperation-conflict two-dimensional framework in which organisms are defined by high cooperation and low conflict among components. Both frameworks are synchronic: they assess the current distribution of cooperation and conflict to determine the current degree of individuality.

This account adds a diachronic dimension to both frameworks. The individuality of a complex animal is not simply the current outcome of conflict suppression by the clonal bottleneck. It is the outcome of a layered sequence: social toolkit assembly in the aggregative ancestor, regulatory transition at the metazoan origin, bottleneck-enforced conflict suppression at the founding event, and secondary re-emergence of conflict through somatic mutation with corresponding evolution of policing systems. The individual animal is a palimpsest — a document written over repeatedly, where earlier inscriptions are not erased but persist beneath the surface and occasionally show through. Oncogenesis makes the conflict dynamics of the aggregative ancestor visible within the body. Developmental positional signaling makes the social constitution of identity visible within the growing embryo. The aggregative behavior of C. flexa makes the earliest inscription visible in a living relative. Individuality, on this account, is not a stable achieved state but an ongoing, historically layered process of conflict management.

7. Objections

7.1 Convergence Rather than Homology

The most direct objection is that aggregative behavior in Capsaspora and choanoflagellates is independently derived and does not reflect an ancestral holozoan condition. On this view, the overlap between aggregative toolkit deployment in these organisms and animal developmental genes reflects convergent recruitment of the most available molecular tools for cell-cell interaction, not the retention of a socially-assembled toolkit by the animal lineage.

Convergence cannot be definitively ruled out with currently available data. Several features of the evidence make it a less parsimonious account, however. The toolkit overlap extends across functionally integrated signaling systems, including the Hippo pathway, integrin-ECM interactions, and tyrosine kinase networks, rather than a few isolated cell-adhesion genes. Convergent recruitment of entire integrated pathways is considerably less probable than homologous retention. Critically, the Phillips et al. (2022, 2024) functional studies demonstrate that the Hippo pathway in Capsaspora is not merely present and expressed during aggregation but is genuinely required for normal aggregative morphology — the ancestral function was aggregative, and the growth-regulatory animal function was derived. This is the kind of evidence that distinguishes retention from convergent recruitment: a pathway shaped by selection for aggregative function is a pathway that was elaborated in that context, not independently recruited into it. The phylogenetic distribution of aggregative behavior deploying animal toolkit genes, now documented in at least five independent holozoan lineages — Capsaspora owczarzaki (Sebé-Pedrós et al., 2013), Choanoeca flexa (Ros-Rocher et al., 2026), Ministeria vibrans (Li et al., 2025, preprint), and two Pigoraptor species encoding near-complete integrin adhesomes (Hehenberger et al., 2017) — is more consistent with an ancestral social condition than with repeated convergent recruitment. Additionally, independent clonal multicellularity with cell-type differentiation in Chromosphaera (Olivetta et al., 2024), a lineage that is not aggregative, is more parsimoniously explained by ancestral toolkit distribution than by independent convergent recruitment of entire integrated signaling pathways in multiple holozoan lineages. Preliminary preprint evidence of kin discrimination in C. flexa, now confirmed in the published Ros-Rocher et al. (2026) study, adds a fourth feature more consistent with ancestral social dynamics than with independent convergent recruitment.

7.2 The Ceiling Still Requires Its Own Explanation

A second objection holds that even if this account correctly describes the molecular origins of the animal toolkit, the aggregative complexity ceiling requires a separate explanation. The fact that social aggregation provided the raw material for animal development does not explain why aggregative lineages themselves never achieved animal-grade complexity. The two facts could be independent.

This account does not replace the conflict explanation for the aggregative ceiling, and the two are compatible. Aggregative behavior assembled the toolkit; the conflict inherent in genetically heterogeneous collectives prevented that toolkit from scaling without the structural fix of the clonal bottleneck. This framing makes the ceiling more interpretively significant, not less. The ceiling is why the social route was superseded. The clonal bottleneck was not an invention that made new things possible from scratch but a structural solution to a specific problem, and the scaffold is what made that solution consequential. Without the pre-assembled toolkit, removing the conflict ceiling would have generated a more complex but still limited organism. The two facts are causally connected rather than independent. One complication deserves careful acknowledgment. Capsaspora lives as a symbiont in the haemolymph of Biomphalaria glabrata, and Ros-Rocher et al. (2023) showed that its aggregation is induced by host-derived lipid cues — lipoproteins and free phosphatidylcholine — rather than by intraspecific social signals. The transition is regulated and reversible, demonstrating that the aggregative program is a genuine biological response to specific triggers rather than passive cell stickiness, which is consistent with the toolkit genes having been shaped by selection for aggregative function. However, the host-derived trigger means that the relevant selective pressure on some gene families may have involved host-symbiont dynamics rather than intraspecific social coordination. This complication does not undermine the scaffold argument, since the hypothesis requires only that the aggregative state was the functional context in which toolkit genes were elaborated — a claim supported by the transcriptomic profile of that state regardless of what initiates it — but it prevents confident attribution of all toolkit elaboration specifically to intraspecific collective organization. Ministeria vibrans, a free-living filasterean without a symbiotic host, provides a cleaner case: Li et al. (2025) document aggregation and toolkit gene deployment in an organism with no known host-derived induction, strengthening the inference that aggregative toolkit assembly is not an artifact of the Capsaspora symbiosis.

7.3 Scope: Is This Specific to Animals?

Animal multicellularity is one of five independent clonal transitions. Do plants, fungi, red algae, and brown algae show evidence of an aggregative social precursor? The hypothesis, as currently evidenced, is specific to the animal lineage and should not be presented as a general revision of major transitions theory without qualification.

We accept this limitation. The evidence for a scaffold relationship is currently specific to the holozoan lineage and its unicellular relatives. Whether other clonal transitions involved similar social precursors is an open empirical question. The claim advanced here is that the standard dichotomy between aggregative and clonal pathways is empirically false for the animal lineage, and that this revision warrants updating the theoretical framework for at least this case. Given that the animal lineage accounts for the majority of complex multicellular organization on Earth, the revision is not narrow in its implications even if it remains lineage-specific in its current evidence.

7.4 The Epigenetic Parallel as Metaphor

A fourth objection targets Section 4 directly. The claim that epigenetic cell-type determination constitutes an 'internal social dynamic' conflates causal mechanism with metaphor. Cells are not voluntary agents; they do not choose to aggregate, negotiate roles, or defect. Positional context in a developing body is a physical-chemical fact, not a social relationship. On this view, framing epigenetic specialization as 'social' imports language from one domain into another without explanatory gain, and the paper's unity claim — that the same organizational logic operates at the historical and organismal scales — is a rhetorical move rather than a scientific one.

This objection has real force and Section 4 should be read in light of it. The paper does not claim that cells are social in the sense that organisms are social — the Section 4 opening explicitly states that the epigenetic argument is a structural analogy, not a claim of historical continuity. The substantive claim is narrower: first, that cell identity in a clonal body is constituted entirely by relational position rather than by intrinsic genomic properties, which is a precise empirical statement about the mechanism of differentiation; and second, that the chromatin machinery executing this positional constitution is shared with the machinery that tracked social context in aggregative ancestors, which is a molecular homology claim, not a metaphor. Whether these two facts warrant the phrase 'internal social dynamic' is a question of terminology. The underlying claims are separable from the terminology and stand independently. Readers who find the framing too liberal are free to read Section 4 as making two distinct points — one about developmental mechanism, one about molecular ancestry — that are unified by shared substrate rather than by any claim about agency or intentionality in cells.

8. Conclusion

The standard account of animal origins tells a story of structure defeating society. The clonal bottleneck suppresses the conflict that aggregative collectives cannot resolve, and developmental complexity follows from the resulting genetic uniformity. This account is accurate about outcomes but incomplete about sequence.

The molecular evidence from the closest unicellular relatives of animals places aggregative social organization before clonal development in the holozoan lineage, not as a competing strategy that was abandoned but as the context in which the developmental toolkit was assembled. Capsaspora activates animal developmental gene networks during voluntary social aggregation, and functional genetic studies of the Hippo pathway in Capsaspora show that the ancestral function of this signaling system was regulation of aggregate morphology, not proliferation — the growth-regulatory role was derived in the animal lineage (Phillips et al., 2022, 2024). Choanoeca flexa forms multicellular sheets through both aggregative and clonal modes, aggregating first under ecological stress and expanding clonally afterward (Ros-Rocher et al., 2026). The regulatory transition at the metazoan origin was the addition of enhancer-based, position-specific gene expression control, and of expanded chromatin reader repertoires, onto molecular raw material that aggregative ancestors had already assembled and deployed.

Within clonal bodies, the same ancestral chromatin machinery that — if the Dictyostelium evidence reflects a broader holozoan capacity — tracked social context in aggregative ancestors now constitutes cell identity from positional context: a shared molecular substrate underlying two instances of the same organizational principle, identity determined by relational position rather than intrinsic properties. This is not a claim of historical continuity between social and developmental dynamics, but of shared molecular ancestry and structural convergence on the same logic. The individual animal is in this sense the palimpsest described in Section 6.3: the conflict problem the bottleneck deferred re-emerges as somatic mutation regenerates genetic heterogeneity, and the organism responds with the convergent solutions the aggregative ancestor would have recognized.

Social organization was not what the animal lineage overcame on the way to complexity. It was the ground from which animal complexity grew, the scaffold that was overwritten but not erased, and the dynamic that resurfaces within the resulting organism whenever the structural solution to the conflict problem begins to fail. The animal lineage did not escape its social prehistory. It built a body around it.

References

Boomsma, J. J. (2009). Lifetime monogamy and the evolution of eusociality. Philosophical Transactions of the Royal Society B, 364(1533), 3191-3207.

Buss, L. W. (1987). The Evolution of Individuality. Princeton, NJ: Princeton University Press.

Davies, P. C. W., & Lineweaver, C. H. (2011). Cancer tumors as Metazoa 1.0: Tapping genes from ancient ancestors. Physical Biology, 8(1), 015001. DOI: 10.1088/1478-3975/8/1/015001.

Driesch, H. (1892). The potency of the first two cleavage cells in echinoderm development: Experimental production of partial and double formations. Translated in B. H. Willier & J. M. Oppenheimer (Eds.), Foundations of Experimental Embryology (1964). Englewood Cliffs, NJ: Prentice-Hall.

Gould, S. J. (1977). Ontogeny and Phylogeny. Cambridge, MA: Belknap Press of Harvard University Press.

Kauffman, S. A. (1993). The Origins of Order: Self-Organization and Selection in Evolution. Oxford: Oxford University Press.

Queller, D. C. (1997). Cooperators since life began. Quarterly Review of Biology, 72(2), 184–188.

Virchow, R. L. K. (1858). Cellular Pathology as Based upon Physiological and Pathological Histology [Trans. F. Chance, 1860]. London: John Churchill.

Waddington, C. H. (1957). The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology. London: George Allen and Unwin.

Boomsma, J. J. (2009). Lifetime monogamy and the evolution of eusociality. Philosophical Transactions of the Royal Society B, 364(1533), 3191-3207.

Brunet, T., & King, N. (2017). The origin of animal multicellularity and cell differentiation. Developmental Cell, 43(2), 124-140. DOI: 10.1016/j.devcel.2017.09.016. Brunet, T., Larson, B. T., Linden, T. A., Vermeij, M. J. A., McDonald, K., & King, N. (2019). Light-regulated collective contractility in a multicellular choanoflagellate. Science, 366(6463), 326-334.

Godfrey-Smith, P. (2009). Darwinian Populations and Natural Selection. Oxford University Press.

Grau-Bové, X., Navarrete, C., Chiva, C., Pribasnig, T., Antó, M., Torruella, G., Galindo, L. J., Lang, B. F., Moreira, D., López-Garcia, P., Ruiz-Trillo, I., Schleper, C., Sabidó, E., & Sebé-Pedrós, A. (2022). A phylogenetic and proteomic reconstruction of eukaryotic chromatin evolution. Nature Ecology and Evolution, 6(7), 1007-1023. DOI: 10.1038/s41559-022-01771-6.

Olivetta, M., Bhickta, C., Chiaruttini, N., Burns, J., & Dudin, O. (2024). A multicellular developmental program in a close animal relative. Nature, 635(8038), 382-389. DOI: 10.1038/s41586-024-08115-3.

Kim, I. V., Navarrete, C., Grau-Bové, X., Iglesias, M., Elek, A., Zolotarov, G., Bykov, N. S., Montgomery, S. A., Ksiezopolska, E., Cañas-Armenteros, D., Soto-Angel, J. J., Leys, S. P., Burkhardt, P., Suga, H., de Mendoza, A., Marti-Renom, M. A., & Sebé-Pedrós, A. (2025). Chromatin loops are an ancestral hallmark of the animal regulatory genome. Nature, 642(8069), 1097-1105. DOI: 10.1038/s41586-025-08960-w.

Grosberg, R. K., & Strathmann, R. R. (2007). The evolution of multicellularity: a minor major transition? Annual Review of Ecology, Evolution, and Systematics, 38, 621-654.

Hirose, S., Benabentos, R., Ho, H. I., Kuspa, A., & Shaulsky, G. (2011). Self-recognition in social amoebae is mediated by allelic pairs of tiger genes. Science, 333(6041), 467-470.

Maynard Smith, J., & Szathmáry, E. (1995). The Major Transitions in Evolution. Oxford University Press.

Nowak, M. A., Tarnita, C. E., & Wilson, E. O. (2010). The evolution of eusociality. Nature, 466(7310), 1057-1062.

Queller, D. C., & Strassmann, J. E. (2009). Beyond society: the evolution of organismality. Philosophical Transactions of the Royal Society B, 364(1533), 3143-3155.

Rokas, A. (2008). The origins of multicellularity and the early history of the genetic toolkit for animal development. Annual Review of Genetics, 42, 235-251.

Coyle, M. C., & King, N. (2025). The evolutionary foundations of transcriptional regulation in animals. Nature Reviews Genetics, 26(7), 447-464. DOI: 10.1038/s41576-025-00819-4. Gahan, J. M., Traylor-Knowles, N., & Bhatt, S. (2025). Chromatin profiling identifies putative dual roles for H3K27me3 in regulating cell type-specific genes and transposable elements in choanoflagellates. Nature Communications, 16(1), 9549. DOI: 10.1038/s41467-025-55133-6. Hehenberger, E., Tikhonenkov, D. V., Kolisko, M., del Campo, J., Esaulov, A. S., Mylnikov, A. P., & Keeling, P. J. (2017). Novel predators reshape holozoan phylogeny and reveal the presence of a two-component signaling system in the ancestor of animals. Current Biology, 27(13), 2043-2050. DOI: 10.1016/j.cub.2017.06.006. Li, R., Dharamshi, J. E., Kwok, K., Ruiz-Trillo, I., & Gerdt, J. P. (2025). A close unicellular relative reveals aggregative multicellularity was key to the evolution of animals. bioRxiv 2025.05.14.654023. PREPRINT.

Phillips, J. E., Santos, M., Konchwala, M., Xing, C., & Pan, D. (2022). Genome editing in the unicellular holozoan Capsaspora owczarzaki suggests a premetazoan role for the Hippo pathway in multicellular morphogenesis. eLife, 11, e77598. DOI: 10.7554/eLife.77598.

Phillips, J. E., & Pan, D. (2024). The Hippo kinase cascade regulates a contractile cell behavior and cell density in a close unicellular relative of animals. eLife, 12, RP90818. DOI: 10.7554/eLife.90818.

Ros-Rocher, N., Kidner, R. Q., Gerdt, C., Davidson, W. S., Ruiz-Trillo, I., & Gerdt, J. P. (2023). Chemical factors induce aggregative multicellularity in a close unicellular relative of animals. Proceedings of the National Academy of Sciences, 120(18), e2216668120. DOI: 10.1073/pnas.2216668120.

Ros-Rocher, N., Reyes-Rivera, J., Horo, U., Combredet, C., Foroughijabbari, Y., Larson, B. T., Coyle, M. C., Houtepen, E. A. T., Vermeij, M. J. A., Steenwyk, J. L., & Brunet, T. (2026). Clonal-aggregative multicellularity tuned by salinity in a choanoflagellate. Nature. DOI: 10.1038/s41586-026-10137-y. [Published online ahead of print, 2026.]

Sebé-Pedrós, A., Irimia, M., del Campo, J., Parra-Acero, H., Russ, C., Nusbaum, C., Blencowe, B. J., & Ruiz-Trillo, I. (2013). Regulated aggregative multicellularity in a close unicellular relative of metazoa. eLife, 2, e01287.

Sebé-Pedrós, A., Chomsky, E., Pang, K., Lara-Astiaso, D., Gaiti, F., Mukamel, Z., Amit, I., Tanay, A., et al. (2018). Early metazoan cell type diversity and the evolution of multicellular gene regulation. Nature Ecology and Evolution, 2, 1176-1188.

Sebé-Pedrós, A., Ballaré, C., Parra-Acero, H., Chiva, C., Tena, J. J., Sabidó, E., Gómez-Skarmeta, J. L., Di Croce, L., & Ruiz-Trillo, I. (2016). The dynamic regulatory genome of Capsaspora and the origin of animal multicellularity. Cell, 165(5), 1224-1237.

Strother, P. K., Brasier, M. D., Wacey, D., Timpe, L., Saunders, M., & Wellman, C. H. (2021). A possible billion-year-old holozoan with differentiated multicellularity. Current Biology, 31(12), 2658-2665. DOI: 10.1016/j.cub.2021.03.051. Suga, H., Chen, Z., de Mendoza, A., Sebé-Pedrós, A., Brown, M. W., Kramer, E., Carr, M., Kerner, P., Vervoort, M., Sánchez-Pons, N., Torruella, G., Derelle, R., Manning, G., Lang, B. F., Russ, C., Haas, B. J., Roger, A. J., Nusbaum, C., & Ruiz-Trillo, I. (2013). The Capsaspora genome reveals a complex unicellular prehistory of animals. Nature Communications, 4, 2325. Tikhonenkov, D. V., Hehenberger, E., Esaulov, A. S., Belyakova, O. I., Mazei, Y. A., Mylnikov, A. P., & Keeling, P. J. (2020). Insights into the origin of metazoan multicellularity from predatory unicellular relatives of animals. BMC Biology, 18(1), 39. DOI: 10.1186/s12915-020-0762-1. Trigos, A. S., Pearson, R. B., Papenfuss, A. T., & Goode, D. L. (2017). Altered interactions between unicellular and multicellular genes drive hallmarks of transformation in a diverse range of solid tumors. Proceedings of the National Academy of Sciences, 114(24), 6406-6411. DOI: 10.1073/pnas.1617743114.

Wang, S. Y., Pollina, E. A., Wang, I.-H., Pino, L. K., Bushnell, H. L., Takashima, K., Fritsche, C., Sabin, G., Garcia, B. A., Greer, P. L., & Greer, E. L. (2021). Role of epigenetics in unicellular to multicellular transition in Dictyostelium. Genome Biology, 22(1), 134. DOI: 10.1186/s13059-021-02360-9.