Varieties of Organicism: A Critical Analysis

Abstract

In earlier work I wrestled with the question of the “ontological status” of organisms. It proved difficult to come to a clear decision, because there are many candidates for what such a status is or would be and of course many definitions of what organisms are. But what happens when we turn to theoretical projects “about” organisms that fall under the heading “organicist”? I first suggest that organicist projects have a problem: a combination of invoking Kant, or at least a Kantian “regulative ideal,” usually presented as the epistemological component (or alternately, the complete overall vision) of a vision of organism – as instantiating natural purposes, as a type of “whole” distinct from a merely mechanistically specifiable set of parts, etc. – and a more ontological statement about the inherent or essential features of organisms, typically presented according to a combination of a “list of heroes” or “laundry list” of properties of organisms. This amounts to a category mistake. Other problems concern the too-strict oppositions between mechanism and organi(ci)sm, and symmetrical tendencies to “ontologize” (thus objectifying) properties of organisms and to “subjectify” them (turning them into philosophies of subjectivity). I don’t mean to suggest that no one should be an organicist or that Kant is a name that should be banished from civilized society. Rather, to borrow awkwardly from Sade, “organicists, one more effort!” if one wants a naturalistic, non-foundationalist concept of organicism, which is indeed quite active in recent theoretical biology, and which arguably was already alive in the organismic and even vitalist theories of thinkers like Goldstein and Canguilhem.

In an earlier paper (Wolfe, 2010), I wrestled with the question of the “ontological status” of organisms. It proved difficult to come to a decision on this matter, because there are many candidates for what such a status is or would be and of course many definitions of what organisms are (see also Pepper and Herron (2008), Pradeu (2010a), and Pradeu (2016), Bouchard & Huneman, 2013) – the “fungus problem,” as it were. But I did not focus in any detail on actual contemporary projects that rely on, or even assert, a certain concept of organism – namely, antireductionist, organicist projects (I’m fully aware that I haven’t defined organicism yet). And I wish to show that these projects have an “issue,” a sort of conceptual confusion, at least in part. My concern is thus not to argue for or against organicism, or to reconstruct a variant of organicism that is immune to some recent objections (a common enough practice), but rather, echoing two very different authors, to say “One more effort to achieve an organicism worth wanting.”¹

3.1 Organicism as Ontology, as Epistemology or as a Blurry Mix

A curious consensus without examination of presuppositions reigns today in the various subfields concerned with the properties (organizational, systemic, etc.) of living beings, at least those which define themselves as more or less “organicist.” There is no strictly defined group of organicists, but attempts are regularly made to group various antireductionist trends in biology under this heading (e.g., Gilbert & Sarkar, 2000; Nicholson, 2014). Descriptions of how “the whole is greater than the sum of the parts” or, slightly less generically, how specific “system-level” or “higher-level” properties need a set of explanations (and perhaps an ontology) appropriate to them abound and have been repeated almost identically since the early twentieth century, with predecessors such as Aristotle and Kant invoked almost as often. At times, the concept of organism at work in these discussions is mostly a placeholder, such as when Lewontin, and developmental systems theory more generally, stresses that the focus is on organism-environment relations (“Just as there is no organism without an environment, so there is no environment without an organism”²). But more current organicism included various recent proclaimed “returns” of the organism concept³ and does not just treat the concept as a placeholder, a functional or a heuristic term. They move toward what I’ll call an ontologization of the notion: at that point, organicism has an ontology, whether it is clearly stated as such or not – it is then an objectified account of the nature of organisms.

Treatments of organisms as more than just placeholders in a theoretical biological system, indeed, as entities with objective properties that should be the object of an ontology do not pose a problem as such – except perhaps to the instrumentalist, for whom it is a mistake to attribute any kind of innate meaning or definition to biological entities, as these are merely instrumental (Wolfe, 2010). What seems more curious to me is the way these treatments variously blend epistemological and ontological versions of organicism, sometimes in awkward ways. Generally, these accounts involve a combination of:

• (a) an invocation of Kant, or at least a Kantian “regulative ideal,” usually presented as the epistemological component (e.g., organicism as a theory about what constitutes “knowledge of life”) or alternately, as the complete account of a vision of organism – as instantiating natural purposes, as a type of “whole” distinct from a merely mechanistically specifiable set of parts, interdependency, self-organization, and so on (Weber & Varela, 2002; Longo et al., 2012).

• (b) a more ontological statement about the inherent or essential features of organisms (organicism as a theory about the actual properties of living systems) – essentially a “laundry list” of core features, e.g., “reproduction, life-cycles, genetics, sex, developmental bottlenecks, germ-soma separation, policing mechanisms, spatial boundaries or contiguity, immune response, fitness maximization, cooperation and/or conflict, codispersal, adaptations, metabolic autonomy, and functional integration” (Clarke, 2013, 415), itself usually tied to a kind of special tradition of thinkers who had a “feeling for the organism”: typically, Claude Bernard,⁴ Sherrington and Waddington, and more recently Maturana and Varela with the notion of autopoiesis, Rosen’s (M,R) systems, Gánti’s chemoton, or Luisi’s minimal cell, all of which at one time or another are presented as the defining “characteristics of life,” or “life criteria” (Griesemer & Szathmáry, 2008). These authors, or perhaps their theoretical constructs, are “heroes of homeostasis,” as it were, or “masters of metabolism,” given the latter concept’s close association with definitions of life, e.g., when it is defined as “the collection of chemical reactions that define a living organism and allow it to make its components and obtain the energy required for staying alive” (Cardenas & Cornish-Bowden, 2011). To be precise, Bernard and others named above do not all perfectly match Clarke’s “laundry list” of core features of life, e.g., fitness maximization. I used her list because it is one of the more expansive yet precise such lists I’ve seen.

The combination of (a) and (b) poses several problems. Note that I am not suggesting that a theory of organism has to be either epistemological or ontological. It can be both, or neither (what one might term constructivism, i.e., the view that organisms are constructs, and furthermore inasmuch as they are “sense-making” entities, they need to be understood via acts of cognitive construction).⁵ What seems problematic to me is to combine epistemological and ontological claims in an “ungrounded” way, not always clearly presented as such (but, conversely, presented as straightforwardly empirical).

Perhaps the most obvious difficulty created by the unreflexive combination of (a) and (b) is something like a category mistake, in this case treating an epistemological property as an ontological property;⁶ it is not quite right to invoke the authority of the Kantian “projective” approach to organisms (Huneman, 2007) in order to assert a set of ontological specificities about organisms. Because this is precisely what the Kantian regulative ideal concept was designed to avoid, in explicit contrast to what he would have called “rational metaphysics.” That is to say, to provide an empirical set of criteria for why living beings are special and to claim that this supports or is supported by a Kantian framework is not a good idea given that the latter framework explicitly rejects the idea of giving empirical definitions of organism, inasmuch as Kant’s organism concept is explicitly built around his notion of regulative ideal.⁷ For Kant, organism is a “reflective” construct rather than a “constitutive” feature of reality, and reflective judgments are “incapable of justifying any objective assertions.”⁸ The purposive character of organisms, on this view, imposes itself upon on us by empirical experience, forcing us to (cognitively) recognize such entities as “organisms,” rather than as machines or mere bundles of matter. Note that this more “constructivist” approach is by no means ruled out in biology, where such regulative concepts can serve as heuristics, e.g., regulative teleology;⁹ the problem lies more in claiming to have an “empirical Kantian” foundation for organicism, because of the inherent tension therein – as when, e.g., Weber and Varela state that Kantian teleology is somehow “true” of organisms; “Our immodest conclusion is that Kant, although foreseeing the impossibility of a purely mechanical, Newtonian account of life, nonetheless was wrong in denying the possibility of a coherent explanation of the organism” (Weber & Varela, 2002, 120). As noted earlier, I am not suggesting that all organicists are Kantians or (a narrower claim) that all organicists make a category mistake with respect to Kantian notions; indeed, some contemporary organicists call for a return to Aristotle, while earlier organicists could also place themselves under the patronage of Hegel (thereby, by the way, eliminating the category mistake in one stroke, as that is a full-blown ontological realism about organisms) (for a rare recent Hegelian move in organicism, see Gambarotto and Illetterati (2020)).

The second difficulty in certain organicist arguments lies in their empirical stance: it is not clear in any case why it counts as an argument against “mechanism” or “reductionism” to present a list of key features (such as the “generative” circular causality in organisms: Mossio, 2020, 55, 57), especially from a standpoint that curiously combines Kantianism, or at least Kantian elements, and the ontologization of organisms. I called this the “laundry list” problem above;¹⁰ notice that the problem with this form of organicism – a kind of literal, empirical claim listing key definitory features of organisms – is very close to that found in defenses of “definitions of life,” particularly “criterial” definitions of life (Malaterre, 2010; see Bich and Green (2017) for a defense of definitions of life).

A different reason why the laundry list of the organic is not the argument-stopper some treat it as is the flexibility of mechanism: mechanism is not reducible to the strictures of say early modern mechanistic ontology (in which what is real is what is specifiable in terms of size, shape, and motion) or to a “toy model” ontology in which what is real in a system is its decomposition into parts (indeed, decomposition is typically presented as an explanatory strategy in contemporary mechanism, e.g., in Bechtel & Richardson, 1993, 30; here I primarily mean decomposition into parts). Organicism in this sense neglects, or almost forces itself to neglect, the reality of “expanded mechanism,” that is, the existence of nondogmatic, or non-foundationalist, or ontologically open mechanistic programs. In this sense organicism can be overly polarized (this is my disagreement with Nicholson, specifically Nicholson (2013); for a more open-ended view, reflecting its historical focus, see Toepfer, this volume). By extension, organicism can at times run the risk of verging on antinaturalism, although antimechanism is obviously not synonymous with antinaturalism (I’ll return to this below). Either organicism neglects the possibility of the reducibility of the laundry list (i.e., that it is potentially reducible to more basic features), or conversely, of the existence of expanded mechanism: the “remit,” the “span” of mechanism is broader than we are accustomed to think when we repeat the mechanism-vitalism or mechanism-teleology opposition. One could adduce the sheer historical complexity of mechanism and its various “expansions,” or the concept of teleomechanism – not in its historically “restricted” sense as in the Lenoir Thesis (Lenoir, 1982), but in a broader sense (Wolfe, 2014c), which stresses that even paradigm cases of “mechanism” or “the mechanical philosophy” (from Descartes and iatromechanism onward) are filled with functional language, including “function,” “use,” and the “office” of an organ. And they are concerned with properties of the organism such as health and survival, that is, not just with microstructure or size, shape, and motion, if indeed, “medicine is the most useful of the sciences” for Descartes (Discourse on method¹¹).

3.2 A Current Organicist Consensus?

Of course, the theoretical biologist interested in articulating a concept of organizational closure¹² might say: none of this matters, because what these “positions” are, are simply theoretical constructs, bricolages intended to facilitate the articulation of an ongoing research project. This can be seen in an example from the earlier part of the twentieth century, Kurt Goldstein’s The Structure of the Organism [Goldstein, 1939/1995], a work which combines a “bottom-up” study of brain-damaged soldiers from World War I with a “top-down” set of ontological, not just tacit commitments but explicit claims about the “whole”: if we treat Goldstein’s references to Goethe and Naturphilosophie seriously, we can charge him with “romanticism,” with being “antimodern science,” and so on. But if we treat these references as simply a typical case of an educated German scientist of the early twentieth century displaying his “humanistic” breadth, then we can study the overall argument in much more naturalistic, or naturalism-friendly terms.

But the goal of the present remarks is to clarify the possible or actual commitments of these organicist projects in current research, to ask if they hang together successfully or not, and to distinguish between them in a way that has not previously been attempted. This is justified also by the fact that many of the actors involved go out of their way to claim a Kantian (or in some cases Aristotelian¹³) pedigree for their experimental, modeling, or otherwise “empirical” projects.

Now, it is not easy to produce a typology of current organicist views, partly because the actors involved waver between positions that we might consider clearly demarcated (notably, epistemological vs ontological positions on organisms and their “reality,” as in the (a) and (b) positions above). Varela, for one, sometimes stresses an “epistemological” standpoint but sometimes also speaks the language of the key features of organisms. Rosen (1991) says he is speaking about “models” but then calls his book Life Itself and indeed complains that mechanistic science has not grasped “life itself.” He also holds that mechanistic laws are the special cases, the outliers, whereas biology gives us a truer picture of nature; but that is not our concern here.

3.3 Organicism Strong and Weak, and the Ghost of Vitalism

Interestingly, as regards the “autonomy and organizational closure,” researchers never seem to make this particular move (different from what I called a category mistake above) of reifying “life” or “organization” in a foundationalist way. The concept that emerges here which seems to overcome the epistemological/ontological divide is organization. Because, as can be seen with the case of Bechtel, both “organicists” and “mechanists” can help themselves to the concept of organization; differently put, organization can be a key concept for both of these approaches, and it can end up being more a matter of perspective than of some purported deep ontology that one specifies this concept in mechanistic terms (especially, expanded-mechanist terms) or in organicist terms (especially, non- or weakly ontologized organicist terms: Wolfe, 2010) – the same can be said of the notion of “structure.” I use the term “perspective” because the choice of studying a system as mechanistic and/or as organizational is also a choice of perspective, or “standpoint” as Needham put it: “all things are organisms and all things are atomic systems also. You choose your standpoint” (Needham, 1930, 85, longer citation in Wolfe, 2014a). However, it is nevertheless possible that any robust notion of organization, unlike older notions of systems à la von Bertalanffy (1933), does rely on a concept of life (and thus is not tantamount to mechanism).¹⁴

Further, the Kantian/epistemological view can also be transfigured into something “ontological,” not as a claim about the world (i.e., entities such as livers and hearts or microbes or finches or coral reefs) but as a claim about the unique nature of the subject (interiority, subjectivity, first-person knowledge, etc.,¹⁵ in which these are distinguished from the physical world as a whole). This is not so surprising if we consider Kant’s famous proclamation that “there will never be a Newton of a blade of grass”: a point which is both about our cognitive capacities and – at least the ambiguity is unresolved – about the uniqueness of organic entities in comparison with the entities studied by physics and the physicomechanical sciences more broadly. In the view of some notable interpreters, “the third Critique essentially proposed the reduction of life science to a kind of pre-scientific descriptivism, doomed never to attain authentic scientificity, never to have its ‘Newton of the blade of grass’” (Zammito, 2006, 755). Actually, Newtonianism was very fruitful as a methodology and an analogy in eighteenth-century life science, both in obvious cases like Albrecht von Haller’s physiology— the first sentence of his textbook on physiology famously proclaims that “the fibre is to the physiologist what the line is to the geometrician” – and in less obvious cases like Montpellier vitalism, which is also filled with Newtonian methodological invocations (Wolfe, 2014b).

This slippage from the epistemological to the ontological is exemplified in earlier twentieth-century theoretical biology in work like Uexküll’s, in which the “epistemic” (the states of knowledge of the world of given biological entities such as the tick: Uexküll, 1934/2010) becomes a feature of the organic as such. Similarly, Goldstein’s theory of organism (Goldstein, 1939/1995), which is a really a theory of the organismic (actually, holistic) capacities of meaning-making and world-making by agents such as humans; it is really more about holism and personhood than about biological organisms per se, but in any case Goldstein’s theory actively moves between these two positions, but that should not prevent us from trying to articulate the difference for the sake of clarification. When Grene discusses phenomenological and philosophico-anthropological approaches to life, including Goldstein’s (esp. Grene, 1966), like Goldstein himself, she elides the difference between states and processes of “the knower” (the subject) and objective properties of “the known” (the object). While most phenomenologists would claim to only be describing the former, Grene shifts the ground to describing the latter: matters of “ontological import” (Grene, 1968, 239).

When organicism is a strong ontology (e.g., Grene, 1966, 1968), which I have also discussed under the heading of “ontologization,” is it so different from its yet-stronger cousin, vitalism? That is to say, organicism recurrently tries to present itself as a “safe word,” the reasonable version of antireductionism (Gilbert & Sarkar, 2000: vitalism is the weird version of our view, so let’s call it organicism), while some biologists will playfully invoke “molecular vitalism” in mainly provocative ways (Kirschner et al., 2000). Indeed, this attempt to present organicism as, typically, the less metaphysical version of vitalism – today one would tend to say the “naturalized” version is an old, perhaps defining feature of this theory (or family of theories, since early twentieth-century organicism is not exactly cohesive in those terms). For instance, Needham wrote that organicism combined “the insistence of vitalism on the real complexity of life with the heuristic virtues of the mechanistic practical attack.”¹⁶

Now, it would be unfortunate if organicism operated with as caricatural a presentation of vitalism as the Cricks and Monods do, for vitalism itself lives in serious tension between different meanings, which indifferent history of science or philosophy often neglects. An eighteenth-century Montpellier vitalist, Bichat, Bernard (both of whom combine denunciations of earlier forms of vitalism with insistence on the existence of uniquely vital properties, Wolfe (2022)), or Hans Driesch – not to mention Bergson or Canguilhem, moving into the twentieth century – are defending very different and at times incompatible programs (Wolfe, 2015a). Indeed, vitalism itself exhibits a tension or plurality between more metaphysical and less metaphysical views and, contrary to a view still found, does not always maintain the existence in an ontologically specific sense of a vital principle, force, entelechy, élan vital, etc. In that sense, when organicism states “there is legitimate non-reductionist investigation of the core properties of life/organization/organism, and there is illegitimate, metaphysical speculation about or hypostatization of these properties,” it is just replaying the standard exclusionary game like mechanism in earlier times; the organicist who treats vitalism as the absolute cardinal error, in a self-congratulatory act of exclusion, should remember the warning “Don’t throw stones in glass houses” (see also Oyama (2010), and for a different version of this point on how organicism may have difficulty demarcating itself from vitalism, especially the less metaphysical variants of vitalism, see Chen (2019)).

In a standard case of organicism defining itself over and against a “bad version of itself” called vitalism, Cardenas and Cornish-Bowden insist that the recognition of metabolism is tantamount to a decline in vitalism, which they see in the nineteenth century: “Although a living organism is identified and recognized by its physical appearance, and hence by its structure, its status as living is defined by its chemistry, and thus by its metabolism” (Cardenas & Cornish-Bowden, 2011, 1016). This approach to metabolism emphasizes that the processes occurring in a living organism are fundamentally chemical reactions and implicitly echo nineteenth-century affirmations (Bernard, Wöhler) according to which chemistry rules out vitalism.

One need only contrast Jonas (see Jonas, 1966/2011) and Cornish-Bowden and Cardenas on metabolism to see the way organicism can be more or less “ontologized,” more or less subjectivist. That is to say, metabolism and homeostasis are good examples of important physiological properties or processes which allow of a kind of philosophical “overdetermination” or “overinvestment” (as happens with Jonas, when a notion like metabolism cannot just mean what it means to a biologist but has to mean more). Notably, metabolism is taken to raise the problem of “subjectification,” i.e., when an organic property becomes construed as requiring a “self” or “subject.” Thus, Jonas presents his work as “a new reading of the biological record [to] recover the inner dimension” and describes metabolism as constituting the organization of organisms for “inwardness, for internal identity, for individuality” while also turning the organism outward “toward the world in a peculiar relatedness of dependence and possibility”; life is “self-centered individuality, being for itself and in contraposition to all the rest of the world, with an essential boundary dividing ‘inside’ and ‘outside’ –notwithstanding, nay, on the very basis of the actual exchange.”¹⁷

One imagines that most physiologists from Bernard and Sherrington onward would find it quite strange to have metabolism equated with “inwardness”; doubtlessly, Jonas is (i) stressing that metabolic entities have an inner/outer relation and (ii) hypostasizing this relation as a philosophical justification of interiority. Organicism is not, then, automatically vaccinated against the excesses of metaphysical vitalism (whether it is Jonas on metabolism, Merleau-Ponty on the flesh, writing “Just as the sacrament not only symbolizes … an operation of Grace, but is also the real presence of God … in the same way the sensible has not only a motor and vital significance but is nothing other than a way of being in the world that our body takes over […] sensation is literally a communion” (Merleau-Ponty, 1945, 245),¹⁸ or Evan Thompson’s reworking of Merleau-Ponty in an even more explicitly dualist direction: “Life is not physical in the standard materialist sense of purely external structure and function. Life realizes a kind of interiority, the interiority of selfhood and sense-making. We accordingly need an expanded notion of the physical to account for the organism or living being” (Thompson, 2007, 238)¹⁹), and because its reasonable variants themselves vary on issues like holism (itself just a word, to be sure).

Of course, organicism (or at least some forms of organicism) seems to steer clear of considerations concerning “subjectivity,” “the transcendental,” “embodiment,” and so on, due to its resolute focus on organization (for a different perspective, see Van de Vijver & Haeck, this volume). Granted, the latter notion is not always strictly defined, but even if it is related to “closure,” it does not seem to require a “controller,” a “hegemon,” as the notion of constraint indicates. However, it allows of more “abstract” and more “literally real” renditions.²⁰ But organicism comes in different strengths, different degrees of ontologization. Some contributors to the history of organicism in the twentieth century, e.g., Jonas, seem to present organisms (which in this case are never nematodes or slime molds, but typically humans or creatures resembling them) as almost outside the realm studied by natural science. But even the “weak organismic view” will sometimes rest its claims on the existence of certain irreducible, specific, key features of organisms (à la Claude Bernard) – what I called the “laundry list” approach above.

Now, this approach remains “biochauvinist,” to use Ezequiel Di Paolo’s suggestive term:²¹ it is essentially a naturalized form of (substance) vitalism, that is, claims about the “uniqueness” of biological entities. Indeed, besides seeking to define itself in opposition to the purportedly metaphysical version, vitalism, organicism can also subdivide into warring groupuscules: to cite one example among many, Needham distinguishes between “dogmatic” and “legitimate” organicism (regarding the concept of organization).²² One could also ask whether all forms of organicism are naturalist; how naturalist can one be here?²³ This is a delicate matter, because most or all contemporary organicists either accept naturalism in a broad sense or at least would be wary of endorsing some antinaturalist position. But at the same time, when they invoke, inter alia, Merleau-Ponty or Jonas, they are explicitly using conceptual accounts of, e.g., life and embodiment, which insist that “life,” “mind,” “selfhood,” “subjectivity,” and “interiority” are somehow not in physical space, outside of physical space, and require almost a separate ontology: on most accounts, that is antinaturalism. If naturalism is a desideratum – if it is desirable for the position argued for in the context of discourses on biology and organisms to be naturalistic or at least broadly compatible with naturalism – then organicism has two different ways of achieving it: it can opt for the epistemological view, as Goldstein does, stressing that organisms are “knowers” who construct their worlds, or it can opt for naturalizing, e.g., declaring teleology to be integrated in a naturalistic framework (e.g., Mossio & Bich, 2017).

3.4 Ontology and Ontologization

One organicist response to my raising the issue of ontologization could be that in fact, they have moved on, and the issue is really explanations: that is, what is unique about biological entities is the type of explanations they require (an idea that might remind us of the epistemological position). So, for instance, Arno Wouters writes that “Functional explanations are, nevertheless, crucial to understand life, because they show us how the characteristics of an organism fit into the requirements for being alive.”²⁴ Of course, such specific explanations may in turn be “generated” or “induced” by features that are claimed to be ontologically real (explanatory autonomy is not an argument-stopper). Indeed Wouters does not seem content to dwell at the level of explanations, as the dimension of organization he emphasizes is ontological: “Organization can be defined in terms of dependence on the composition, arrangement and timing of the relevant system, its parts and their activities.”²⁵ When Dan Nicholson writes that “Organicists, in contrast, are committed to the belief that the integrated and inherently purposive nature of organisms reflects the very essence of what they actually are in reality,”²⁶ I think this is right about a number of forms of organicism (and this leads to some of the problems surrounding “ontologization” I have discussed above), but it does not apply, for instance, to the organicism of Kurt Goldstein (together with another important figure who I have not discussed here, Georges Canguilhem²⁷), because the latter form of organicism seeks to integrate – or better, it builds on – the irreducible existence of a feature of organisms one might call existential or perspectival: the fact that an organism always has an existential attitude, a perspective on other organisms, and that, thereby, part of its vital activity involves acts of (cognitive) construction.

It is as if the Kantian constructive concept of organisms had been ontologized without being subsumed under a potentially refutable or reducible list of vital features. Because, as Canguilhem argues powerfully in his essay “Aspects of vitalism,” vitalism (and here we could replace this term with “organicism,” if this term is not construed in the strictly empirical, ontologized, criterial, “laundry list” sense) is not refutable like geocentrism or the phlogiston theory are. It is a different kind of theory. If organicism were like geocentrism or the phlogiston theory, it would be because it was a strictly empirical (and thus refutable) set of claims. But, on the constructivist view, according to which organisms both require acts of cognitive construction and perform such acts, organicism is not reducible to an empirical set of claims (a.k.a. definitions of life). Goldstein and Canguilhem do not give a set of empirical features of organisms (purposiveness/teleology, integration, reflexivity, etc.) and then state “Here are organisms and here is why organicism is the theory that does justice to them.” They argue less in empirical, “realist” terms and more in existential, constructivist terms.

Now, reflecting on this existential or perspectival dimension (which will be familiar to students of Uexküll, of course) also helps us with our rough typology of organicisms, since it leads us to nuance, not so much the question of ontologization and weak or strong organicism, but at the other end of the spectrum, the question of the “subject.” That is to say, if earlier I tried to diagnose a problem of “ontologization,” in that sense, objectification (of properties of organisms), there is a kind of symmetrical problem, the other horn of the dilemma, which is subjectification. By this I mean the kind of “romantic” vision of organism as deeply opposed to machines, or inert matter, because it is a version of subjectivity, because it has a “self” (this is explicit in Hegel’s philosophy of nature, Schelling’s and I think Jonas’s – as defended again in Michelini et al. (2018); it is – debatably and controversially – also present in the later Varela (Varela (1991), Varela (1996), Varela & Shear, (1999)), and definitely in Evan Thompson). Arguably, the organizational perspective on organisms allows the organicist to escape the danger of subjectification, but so does the Goldstein-Canguilhem approach, i.e., the “constructivist” vision of organism.

3.5 Conclusion

To sum up, my point is not to assert or deny the reality or pertinence of organisms or organismic theories in biology. I notice that no one seems to bother to do this anymore: Bickle (2003) argues aggressively for reductionism (of cognitive neuroscience to cellular-level neuroscience), but otherwise it has become more than rare – irrelevant – to argue “reductionism” against “organicism” (contrast Nagel in the 1960s). Present-day neomechanists need not worry, if we consider Bechtel’s own “organizational” commitments. For Bechtel, mechanism and organization stand in a complex and fruitful interaction, and mechanism can provide an adequate account of organization by “placing as much emphasis on understanding the particular ways in which biological mechanisms are organized as it has on discovering the component parts of the mechanisms and their operations”; “It is the fact that the system is organized (and the type of organization it has) that makes it amenable to mechanistic description and analysis” (Bechtel 2007, 270; Levy & Bechtel 2013, 244). My point is not to defend “neomechanism” in extenso (it has received a number of quite different, and significant criticisms) but to observe that one of its most useful features in the present context is to not just oppose mechanistic and organizational approaches. As Moreno puts it, “holism and mechanistic decomposition can be combined for the purposes of biological explanation,” increased complexity creates “selective functional constraints,” giving rise to “levels of organization in which a mechanistic decompositional strategy might be locally applicable.”²⁸

Similarly, my point is not to simply put forth a “safe,” reductionist critique of all or any such programs. On the contrary, I view it as extremely useful to track such articulations, whether or not we fully commit to them (in this I feel kinship with the “epistemology of Life” in Canguilhem; see Canguilhem (2008a) and Wolfe (2017)). To clarify one potential ambiguity, my point here is deflationary (about organicism, i.e., not reductionist in the sense of being committed to a more minimal ontology, but deflationary inasmuch as I challenge some higher-level concepts in favor of stripped-down versions thereof) but does not take a stand on organisms (their ontological status); I’ve taken several different stances on organism in earlier papers, e.g., in my 2010 “Do organisms have an ontological status?” arguing for a weakly ontological status – indeed rather consonant with what I’m saying here about organicism – and in my 2014 “The organism as ontological go-between” emphasizing rather the “nomadic concept” aspect, the hybridity in its fertility (Wolfe, 2010, 2014a). Organisms as conceptual hybrids are not so directly opposed to mechanism(s) as in the usual organicist party line. They are also so not ontologically solid or essentialist (that’s what the word hybrid indicates).

I have tried to suggest (a) that organicism is not necessarily so far removed from the more naturalistic forms of vitalism (and this, to be clear, is not a bad thing); but (b) that when organicism relaunches itself as a fully respectable scientific theory – no one’s mistress, it might say to Haldane – it creates for itself several potential conceptual problems. One is the unstable combinations of the Kantian epistemological perspective with deliberately empirical, ontologized claims – as if organicists wanted to play both sides in an unprincipled way (sometimes claiming features must really be part of organisms, other times claiming they are only regulative ideals).²⁹ The other is, in a sense, another kind of instability resulting from the reduction of organicism to a “merely empirical,” criterial set of definitions (the “laundry list”). This instability is, ironically, the more Kantian component of the present remarks: faced with the confidence of thinkers asserting the “the soul and the world exist,” Kant suggested an approach that would not be founded on directly refutable empirical claims. Founding claims for the irreducible nature of organisms on directly refutable empirical definitions create this instability.³⁰ Not all contemporary organicists make this Kantian move, or simply the (ontologized) empirical criteria move; I noted in addition (c) that symmetrically to this danger of the excess, the addiction to ontologization, there is the potential to reintroduce into organicism a kind of Romantic love affair with subjectivity, ineffability (as we saw with Merleau-Ponty and Canguilhem’s sensitive commentary) and the first-person perspective. It would be a good idea to try and achieve some conceptual clarity so that the observer can have an idea of what is involved in the current smorgasbord of mildly or strongly organismic projects.