Jason Ānanda Josephson Storm https://orcid.org/0000-0002-4607-8060 Williams College
Keywords: hylosemiotics; semiotics; more-than-human culture; agency.
Abstract: This article develops a metamodern framework for the study of more-than-human cultures by integrating a materialist semiotics (hylosemiotics), with a processual social ontology of agency. Against anthropocentric accounts that restrict culture to language and symbolic thought, I argue that culture emerges wherever social animals coordinate through materialized signs. Drawing on cases from corvid tool use to multispecies infrastructures, the article reconceptualizes meaning as situated inference rather than transmission. Agency is asymmetrically distributed across socio-material systems, not lodged in individual actors or flattened into universal actancy. The result is a framework that preserves normativity without either anthropocentrism or anthropomorphism and naturalizes meaning without reducing it to matter. By tracing how coordination, agency, and responsibility are routed through materialized signs and social kinds, metamodern theory offers an analytic for studying multispecies cultures as well as ethical accountability under conditions of planetary scale transformation.
At the edge of the forest above Wabao village in New Caledonia, a young crow watches from its perch on a pandanus leaf. Below, among the garden plots and scrubby paths, its parent grips a branch with the steady confidence of long practice. What follows is what Jennifer Holzhaider and company call a “well-coordinated sequence of cutting and ripping actions,” but that clinical phrasing hardly captures it (Holzhaider et al, 2010: 210). The adult crow is sculpting. Cutting here, stripping there, working the branch with deliberate precision until it has been reshaped: now broad at the base, tapered to a barbed tip. The finished tool is purpose-built for one task—teasing insects from deep inside rotting wood. Here’s what makes it remarkable: this isn’t instinct in motion. This type of tool making seems to be socially transmitted. The crow learned by watching. The technique varies from region to region. And with each generation, the birds refine it. By some accounts, this is a prime example of non-human “material culture” (Hunt and Gray, 2003).
While attributing culture to a species of corvids may not surprise readers of this particular journal, claims of this sort were, until relatively recently, nearly unspeakable. Despite its etymological roots, the term “culture” has long been held captive to an anthropocentric frame. For much of anthropology’s history, in particular, culture was presumed to be the exclusive domain of Homo sapiens, a mark of our supposed exceptionalism. Anchored in language and symbolic thought, culture was not merely what humans had, but what made us human. It was taken to mean the symbolic mediation of human meanings by human minds. Likewise, culture and nature were presumed to be a binary opposition. Within this tradition, the phrase “more-than-human culture” would have sounded like a category error. All that is to say, dominant anthropological frameworks doubled back on themselves—defining the human by culture and culture by the human—and foreclosing more expansive theorizing that might include other species. And yet, as the crows of New Caledonia suggest, other species remember, imitate, refine. Their tools carry knowledge. The circular definition did not preserve a truth. It prevented inquiry into what humans share with other social animals.
Thankfully, given the recent special issue of this journal (see introduction by Herrmann-Pillath 2022), I need not rehash the now-familiar, albeit still incomplete, shifts of the past two decades—the revival of Jakob von Uexküll and the multispecies anthropology catalyzed by Eduardo Kohn and others. This journal’s refocus, along with a broader shift toward multispecies and more-than-human approaches, suggests that the value of such work is finally being recognized. Even so, much of the existing theorization remains conceptually under-specified and hence, in this respect, I believe I have something to add. In Metamodernism: The Future of Theory (2021), I also argued for the ethical and ontological stakes of multispecies flourishing and I developed frameworks for semiotics and social ontology that may offer a useful complement to the study of more-than-human cultures. In particular, accounts of more-than-human culture often operate with a flattened semiosis that blur crucial distinctions between agency, causation, sign-production, and interpretation. What I want to propose here is a recalibration that sharpens those distinctions while preserving the gains of multispecies scholarship.
Hylosemiotics: Toward a Naturalized Semiotics of Culture
The “hylo” in hylosemiotics derives from the Greek ὕλη (hylē), originally meaning “wood” by extension “forest,” and then later coming to mean “matter” in Aristotelian philosophy. As the term suggests, my brother (Seth Josephson) and I coined “hylosemiotics” to name a semiotics grounded in materiality. It starts from two claims. Meaning is not confined to language. All signs are materialized.
The first claim aligns with biosemiotics and with several post-Peircean approaches, though my own argument diverges in emphasis and detail. I argue that meaning arises wherever a sentient being makes inferences about its world. By “sentient,” I mean capable of interpreting signs. The category reaches beyond animals to plants, fungi, possibly certain information-processing machines. It may even make sense to conceive it on a spectrum, not a hierarchy, but a range of distinct kinds of sentience (see Godfrey-Smith 2016). Still, not everything that reacts to its environment is thereby interpreting signs. A rock that heats under sunlight responds causally by absorbing energy and increasing in temperature, just like other features of its surroundings. It does not infer. Interpretation requires that an entity treat some material trace as carrying information relevant to its situation. Nor does it possess agency in any non-vacuous sense. When a New Materialist claims that a rock “has agency not only in its capacity to excite the human mind, but also in the energy of its atoms that support a mountain overhead,” (Tagnani 2019, 218; also discussed in Storm 2021: 161), this amounts to little more than the observation that rocks exist materially and can be perceived or described by humans. Here, the concept of agency does no explanatory work. The attribution trades on an equivocation between material efficacy and interpretability. Material interaction is ubiquitous. Interpretation is not. Where that threshold lies—between brute physical causation and genuine semiosis—is worth contesting. But the distinction itself is not optional (I’ll return to and recoup a more generative account of agency below).[1]
When a bird scans the horizon for a shadow, when a child reads the sky for snowfall, both are interpreting signs. Meaning is not limited to acts of communication, nor is it an imposition on a mute world. It emerges from the practical exigencies of life, from what matters to an organism in a given context (Storm 2021: 151–2).
Language evolved by exploiting perceptual systems already in place. Long before humans coined words, other animals navigated their surroundings through tracking patterns of similarity, spatial affordances, and other correlations that bore on survival. The shape of a paw print. The shift in a scent. The sudden stillness in a forest. These were signs for animals because meaning could be extracted from them. In this respect, the perceptual capacities used to interpret the movements, calls, and gestures of conspecifics were not categorically distinct from those used to read the broader environment. Communication did not create interpretation; it focused it. Hence, the central claim follows: interpretation does not require intentionality on the part of the sign producer.
In this, I am indeed drawing on Uexküll’s insight that organisms inhabit worlds (Umwelt) structured by functional significance, but I depart from him in a crucial way. Hylosemiotics foregrounds an inherent asymmetry between sign production and sign consumption. Sentient beings consume voluntary signs, those deliberately emitted, and involuntary ones, unintended but no less meaningful. A snake’s hiss may be deliberate. A broken twig is not. Both can be meaningful signs, depending on the interpreter and the context.
To be clear, “voluntary’’ and “involuntary’’ name a continuum, not a binary.[2] Consider a mundane example. When I give a talk on “Competing Epistemologies in the History of Japanese Medicine,” I produce signs I intend an audience to interpret: claims about Kampō lineages, subtle ki 気 anatomies, diagnostic regimes, and the politics of nosology. That is the voluntary, sender-directed layer of meaning. But the same performance radiates involuntary signs. My North American accent. The cadence of my enthusiasm. The slight quickening when I reach material I particularly enjoy, perhaps even the fatigue from the red-eye I foolishly took to get there. None of these are meanings I actively set out to convey. Indeed, they occupy a graded field of “voluntariness”: I can attempt to suppress my accent or mask exhaustion, but only with limited success, and never entirely. Yet these cues may be among the most salient meanings inferred by particular listeners, especially those bored by the talk’s main topic.
This is why I argue that meaning is inevitably to some extent interpreter-relative. There are inferences sign producers intend to communicate—otherwise they would not bother to produce voluntary signs at all. But the same signs, identical in material form, can yield divergent meanings depending on who is doing the inferring and to what end. Hylosemiotics treats this divergence not as a problem in need of a solution but as evidence of what meaning is: an act of situated inference operating across a spectrum of intended and unintended cues.
This inferential model of meaning also corrects several dominant accounts of meaning and information. It challenges prevailing linguistic theories within the humanities by relocating meaning away from communicative acts and the internal architecture of linguistic structures. It likewise rejects communication-theory models in which “information’’ is identified with statistical unpredictability or complexity. Meaning, in this view, is emphatically not equivalent to Shannon-style entropy. A string of unpredictable nonsense syllables may score maximal “information” in the formal computational sense, yet remain semantically inert. The phrase “Frumpy go pick Michael hats now tickle money gravity” ranks high in statistical novelty or entropy, but achieves negligible interpretive yield. Conversely, the phrase “No, I am your father” uttered by Darth Vader—statistically predictable to the point of cliché—precipitates a dense inferential cascade for an attuned interpreter such as Luke Skywalker. Meaning, on this account, is not a measure of information-qua-unpredictability. It is a function of situated inference. Meaning crystallizes in reception. The sender may propose a meaning, but the receiver disposes.
Meaning emerges from a sentient being’s interactions with and interpretations of its environment. But against some accounts of “natural information,” the world does not come preloaded with fixed, interpretation-free packets of information waiting to be decoded. Significance is not built into things. It arises in context-sensitive inference. Take categorization. Identity, or more precisely similarity, is never absolute. It is fundamentally context and purpose dependent (Storm 2021: 79).
Take the term “berry.” Botanically, it includes cucumbers, grapes, and chili peppers and excludes strawberries. In common usage, the reverse holds. Neither classification is inherently wrong. Each picks out different clusters of properties in light of a distinct task: one scientific and biologically developmental, tracking the properties shared by fruits derived from the ovary of a single flower; the other culinary, tracking sweet, small, edible items that behave similarly in recipes. But a category that lumps together strawberries, cucumbers, unicorns, and a particular shade of yellow would be analytically, void of purchase. It would be useless to anyone because it would fail to track any coherent cluster of properties relevant to any actual task. Inference is powerful, but not all interpretive moves are conceptually or practically valuable.
All that is to say, hylosemiotics distinguishes receiver meaning (what the interpreter infers) from sender meaning (what the emitter hopes will be inferred) (Storm 2021: 172–4). These may align or misfire by degrees. Either way, inference, not intention or statistical novelty, drives meaning. To be clear, hylosemiotics does not offer a universal metric of meaning, nor a replacement for ethnographic or ethological inquiry. It is a framework for specifying how meaning is enacted under particular material and ecological constraints.
The second key conception of hylosemiotics is that all signs are materialized (Storm 2021: 163–168). This is not a metaphor. It names a basic ontological and epistemic constraint: before any sign can be interpreted, it must take material form. If I speak a word, your eardrum must vibrate; without that vibration—without air pressure changes in a medium—there is no auditory sign to interpret. When you read this sentence, you are not receiving some disembodied meaning; you are engaging with patterns of light emitted as pixels on a screen.
Recast, signs exist only as physical traces—gestures, odors, prosodic modulations, bodily cues, inscriptions, footprints, pheromone trails, digital pulses. Meaning is always embodied in some substrate and therefore susceptible to decay, distortion, drift. There is no “pure” sign behind its materialization; the materialization is what makes the sign a sign.
For readers steeped in New Materialism or multispecies studies, the distinction here matters. Hylosemiotics does not claim that matter itself is meaningful, nor that any material interaction automatically counts as semiosis. That move, common in strands of New Materialism, collapses causation into communication and risks attributing “agency” or “meaning” wherever there is physical change. Hylosemiotics draws a harder line.
Materialization is a precondition for semiosis, not its completion. A falling rock may leave an indentation in the soil. That indentation becomes a sign only if some sentient being can draw an inference from it. Matter furnishes the trace. Meaning arises when an interpreter takes that trace as carrying meaning for them. Signification therefore requires two things simultaneously: a material form in the world and a situated inferential act by a sentient being.
This framework moves beyond merely cataloguing potential signs toward an account of why some patterns come to matter for specific organisms. Signs are mind dependent in the sense that they require interpretation, but they are not arbitrary. Their salience is shaped by physical, biological, evolutionary, and historical forces that carve the world into affordance laden patterns, clusters of similarity and difference more likely to be noticed, tracked, or acted upon by organisms capable of inference. For example, scent marks on a trail are not inherently meaningful: they become signs for a wolf because its morphology, olfactory capacities, and evolutionary history render certain molecular traces salient as indicators of territory, kinship, or potential prey. Semiosis, on this view, is not a gloss imposed on a neutral world. It is a way of navigating fields of potentiality structured by both material constraint and interpretive capacity.
What hylosemiotics offers, than, is a rigorously naturalized semiotics that honors materiality without conflating it with meaning. The sign is always in the world; the meaning is always in the inference. This makes signs neither sheer constructions nor rigid reference to natural kinds, but processual affordances that permit potential actions for the interpreter.
That said, hylosemiotics takes this insight to a maximal extension by denying a rigid bifurcation between a mental inside and a physical outside. We are not disembodied intellects. The mind is not a sealed chamber but a participant in a semiotic ecology that extends across bodies, artifacts, and environments. Knowledge emerges not from abstract contemplation, but via the exploratory manipulation of the physical environment of an embodied being (Storm 2021: 198-199). This position resonates loosely with work in distributed and extended cognition, but it differs from those accounts by treating materialization not merely as a cognitive aid but as the very condition under which signs become available for interpretation and thus cognition becomes possible in the first place.
Cognition is not only embodied. It is also offloaded. Matter and energy are used to store information and alter cognitive complexity (Storm 2021: 199-200). We learn to count on our fingers. To jot a number on a napkin while calculating a tip is to reduce the cognitive complexity of a task by externalizing thought into the world. The napkin becomes a cognitive artifact. Thought extends into matter. We let things do some of our thinking for us. Writing is only the most familiar instance of this. Long before written language, shepherds used tallies to track their flocks. More recently, many of us memorized the phone numbers we dialed most often. Today that information has been stored in our devices and, in many cases, lost to conscious recall.
The externalization of memory cuts both ways. As Bernard Stiegler (2009) warned, externalizing memory enables vast accumulation of knowledge but risks alienating us from it. We gain access but lose possession. This tension sharpens in the era of generative AI, where machines simulate understanding even as our own grasp of underlying concepts atrophies. The outcome of this phase shift remains uncertain. Our prosthetic minds grow stronger. Our capacity to wield them may not.
Crucially this suggest that: culture is not primarily lodged in individual cognition but in shared ecologies of signs. Put differently, another key implication of a hylosemiotic account is its emphasis on the public dimension of semiosis and the forms of coordination this makes possible. In humans, as in other social animals, sign production and uptake function as mechanisms of alignment, ways of synchronizing attention, expectation, action across individuals. This coordination is never perfect. Members of a group do not share identical representations. They approximate, negotiate, adjust. What matters is not uniformity but partial alignment. There are also unintended manifestations. Still, from a hylosemiotic perspective, cultures are composed not of shared mental contents but of individuals whose concepts are oriented around overlapping constellations of materialized signs. This shift allows for more precise analysis of culture, no longer framed in terms of diffuse collective representations but as the outcome of concrete interactions mediated by specific material vectors.
Public representations do more than transmit information. Material signs actively structure social relations. They mobilize affect, as in propaganda. They cultivate desire, as in advertising. They enforce power, as with weapons. They stabilize hierarchies, as in regimes of accumulation. These dynamics are not exclusive to humans. Many nonhuman social worlds likewise depend on material mediators to coordinate action and dominance. Dominance displays. Scent markings. Even in paradigmatically human cases, collectives cannot be reduced to aggregates of persons alone. A hospital, for instance, is constituted not only by doctors and patients but by diagnostic machines, pharmaceutical supply chains, architectural wards, insurance protocols, and the constant flow of blood, data, and sterile equipment.
In sum, our social world is constructed not of immaterial discourse, but in materialized forms. Culture emerges from the practices by which social animals coordinate via materializing signs. Not only humans. Crows, elephants, octopuses, bees. All participate in culturally mediated forms of coordination that exceed individual learning while remaining embedded in specific ecological and bodily constraints.
To head off a misunderstanding, I am not denying that human cultures and languages exhibit distinctive traits. As Susan Pearson and Mary Weismantel (2010: 17) aptly warn, anyone studying animal social life must steer between “the Scylla of anthropomorphism and the Charybdis of anthropocentrism.” Human language has species-specific affordances. Hierarchically recursive syntax chief among them. No known nonhuman culture matches the scale and cumulative complexity of the current human world-system. Yet, what is often treated as uniquely human “culture” is already richly multispecies. Humans transform plants and animals. They in turn transform us. For example, the coevolution of humans and domesticated grains reshaped human dentition, metabolism, settlement patterns, entangling biological change with cultural practice. The contrast, then, is not between culture and its absence but between different regimes of semiotic coordination. Put plainly, different sentient beings coordinate through different kinds of signs and senses, at different biological tempos, with very different consequences for bodies and lifeways.
Social Kinds: A Social Ontology for Agency in Complex Systems
I want to introduce one further plank of the model I developed in 2021 and connect it explicitly to questions of agency. That work pushed back against the methodological individualism that still organizes much social and cultural theory. Rather than treating the social as either reducible to individual actors or dissolved into vague structural forces, I argued that the social world is composed of process social kinds, best understood as temporary zones of stability in unfolding social transmission processes, sustained through their materialization and through the patterned interactions that give them durability over time. The social world is built not only out of social animals but out of the practices and materialized signs through which coordination becomes possible.
At least provisionally, I defined social kinds as “socially constructed dynamic clusters of powers, which are demarcated by the causal processes that anchor the relevant clusters.” (Storm 2021: 111) The details need not detain us here. What matters is what this definition allows us to recover conceptually.
Approached differently, humans and other social animals continually bring new entities into being. Calendars and currencies, certainly. But also territories marked by scent. Tool traditions in corvids. Ant colonies. Dominance hierarchies within flocks. Migration routes stabilized across generations. Domesticated species. Kinship roles. Traffic flows. Spears. iPhones. Bureaucratic offices. Waste streams that reorganize ecosystems. Some of these are artifacts. Many are not. Some are explicitly institutional. Others are emergent patterns of coordination sustained by repetition, perception, material constraint. I refer to these second-order entities as “social kinds.” But that language should not lead one to presume unintended oppositions. Social is not here intended as an antonym to physical or biological. Physical properties are part of what give social kinds their social powers, and physical objects are brought into being by social forces. Looping effects often make the biological and the social difficult to disentangle (as in the case of domesticated species). Many animal species produce social kinds. So nature-culture, nature-social binaries are unwarranted. All that is to say, many social kinds interweave what could be called social, physical, and even biological properties.
This framework supports a distinctive account of “the systematic embedding of agency” (to echo the call for the special issue)—one that rejects both voluntarism and ontological flattening. Agency does not “shift” between humans and nonhumans, nor does it dissolve into an undifferentiated field of actants. It is distributed asymmetrically across systems. Humans remain sites of interpretation, deliberation, and accountability, but we are never the sole drivers of action. What matters analytically is not simply who acts, but how action is enabled or constrained by the social kinds through which it passes. This preserves responsibility without anthropocentrism—precisely the conceptual leverage required for thinking about culture and accountability under the planetary-scale transformations this special issue confronts.
Teleology is not inherent in all things, but it is common in social kinds. Social kinds often exhibit structures that reinforce their own propagation toward particular goals. This means there can be a rationalization toward function in a Weberian sense. Armies, for example, tend to be rationalized such that they become better at fighting. Traits that make them more successful at that are likely to be copied or reproduced. They beat other armies.
Similarly, ant colonies evolve toward more efficient resource extraction and colony survival. Their organizational features, division of labor, pheromone-based communication, adaptive foraging strategies, are shaped by selection pressures that reward success in maintaining the collective. The structure stabilizes around this teleological vector, even if no individual ant “intends” the outcome.
Yet in human societies, the stated goals of social kinds and the ends toward which they are actually rationalized often diverge.[3] Take the academy. Its avowed aim is the production of knowledge. What is in fact reproduced, however, are often the practices that generate career success. In other words, those most likely to be published, rewarded, and advanced. I’m not denying the possibility of progress in academic fields, especially insofar as rewarded systems are well-calibrated, but such progress always occurs under professional selective pressures. This produces shifting local equilibria: waves of scholarly fashion emerge because certain approaches are temporarily easier to publish and more legible as merit. Once widely adopted, they lose that advantage, and the cycle turns. To be clear, I’m not denigrating the academy, which is under threat across the globe at the moment, but merely reminding us of the importance of careful attention to those instances when stated aims and professional incentives diverge.
This has direct implications for planetary ethics and governance. Many of the systems reshaping the Earth—energy regimes, food systems, logistics networks, research infrastructures—are social kinds with powerful teleological momentum. They optimize for throughput, growth, or efficiency even when those trajectories undermine collective flourishing. Ethical responsibility, in this context, cannot be reduced to individual intention or moral resolve. It requires attention to how social kinds are structured, what they are rationalized to reproduce, and where their operative goals diverge from the values they publicly profess.
This is also where my model intersects, though not identically, with what Martin Bohle and Eduardo Marone have called geoethics. In their 2021 piece, they describe geoethics as an “epistemic-moral hybrid,” a stance that combines knowledge about Earth systems with moral reflection on human entanglement within them. They suggest “the central tenet of Geoethics is the virtuous and responsible individual (human agent) who pursues a practice that is geosciences-knowledge-based, just, equitable, inclusive, participatory, and ecologically oriented. (Bohle and Marone 2021: 5). This strikes me as broadly valuable. My own framework, however, reframes the issue less in terms of individual moral stance and more in terms of the distribution and coordination of agency. The social kinds theory I’ve articulated does not ask only what individuals should do, but what kinds of semiotic and material infrastructures make certain actions possible in the first place. Rather than rooting ethics in an epistemic posture toward “the Earth system,” it asks how responsibility and transformation are routed through social kinds—some explicitly moral, some infrastructural, some sensory, some nonhuman. If geoethics is a call for situated care, social kinds theory offers an analytic for how situatedness is structured. Both are needed.
Conclusion
In summary, hylosemiotics offers a way to analyze more-than-human culture without collapsing meaning into matter, agency into causation, or responsibility into a flattened ontology. It sharpens the analytic lens through a set of operational distinctions: between causation and interpretation, sender meaning and receiver meaning, voluntary and involuntary signs, material trace and inferential uptake. These are not abstractions. They allow scholars to track how coordination happens. Which signs matter, to whom, under what constraints, with what consequences. A honeybee’s alignment dance. An elephant matriarch’s infrasonic call. A factory shift bell. A climate station’s sensor grid. These are not background conditions or metaphors. They are semiotic events that organize perception, anchor expectation, modulate power across species lines.
My account social kinds also reframes the question of agency. It shifts focus from the actor to the architecture of action, from not just “who did it” to how doing became possible, how intent got routed through material and institutional systems, how forces amplify or dissipate through social kinds. Those kinds braid together biology, infrastructure, historical process. In this sense, I’ve aimed to provide a method for studying coordination, accountability, transformation in entangled multispecies ecologies on a planet undergoing systemic change.
References
Bohle, M., and Marone, E. 2021. Geoethics, a Branding for Sustainable Practices. Sustainability, 13(2), 895, 1-12. https://doi.org/10.3390/su13020895
Godfrey-Smith, P. 2016. Other minds: The octopus, the sea, and the deep origins of consciousness. Farrar, Straus and Giroux.
Herrmann-Pillath, C. 2022. Roundtable: The future of culture in more-than-human worlds of being. Cultural Science (14), 1-2.
Holzhaider, J.C., Hunt, G.R., and Gray, R.D. 2010. Social learning in new Caledonian crows. Learning & Behavior 38, 206–219.
Hunt, G.R. and Gray, R.D. 2003. Diversification and cumulative evolution in New Caledonian crow tool manufacture. Proceedings of the Royal Science Biological Sciences 270(1517), 867-74.
Nencini, A. M. 2025.Metamodernism: A Multispecies Approach to Hermeneutics. Method & Theory in the Study of Religion. 37, 47-56.
Pearson, S. J., and Weismantel, M. 2010. Does “The Animal” Exist?: Toward a Theory of Social Life with Animals. In D. BRANTZ (Ed.), Beastly Natures: Animals, Humans, and the Study of History (pp. 17–37). University of Virginia Press.
Stiegler, B. 2009. Pour une nouvelle critique de l’économie politique. Galilée.
Storm, J.Ā.J. 2021. Metamodernism: The Future of Theory. University of Chicago Press.
Storm, J.Ā.J. 2025. Breaking the Postmodern Deadlock: Metamodernism’s Methodological Revolution. Method & Theory in the Study of Religion. 37, 57-71.
[1] More precisely, I would argue that “interpretation” is a particular type of manifestation of causal powers, one characteristic of sentient or suitably information-sensitive systems. But space prohibits a fuller elaboration of my account of causal powers.
[2] For further discussion of my use of voluntary in the multispecies hermeneutic, see Nencini 2025, Storm 2025.
[3] In capitalist societies, these rationalization processes are often further captured by the logics of capital itself. In this respect, the operative trajectory of a social kind—whatever its initial orientation—tends to be determined by its capacity for commodification. The result is often a balance between functional improvement and what Cory Doctorow calls “enshittification.”
Juha Hiedanpäa, LUKE, Finland
you have written a wonderful piece of short writing. The concept of hylosemiotics and the way you frame and articulate it will be of great interest to the journal’s readers. I don’t have any critical comments, but I would like to show you a couple of places where you could sharpen your focus and help readers to understand what you are up to and why.
Your starting point on the inherent asymmetry of sign production and sign consumption is simple but surprisingly rewarding, and how you move on to the continuum of voluntary and involuntary cues or signs. Already at this point, it made me wonder; indeed, there is a strong applicability beyond theoretical discussions. Immediately, for example, I started thinking about how to apply this to failing biodiversity policies. (But not yet here, I could not grasp the full potential.)
Indeed, interpretation does not require intentionality on the part of the sign producer, e.g., a sound of a broken twig. One thing, however, made me wonder. You write “All that is to say, hylosemiotics distinguishes receiver meaning (what the interpreter infers) from sender meaning (what the emitter hopes will be inferred)“. It is an interesting choice of words to say that there is hopefulness on the side of a voluntary (and involuntary) sign. What do you think about nonhuman species’ hopefulness, e.g., the wolf marking a territory or a plant transmitting certain chemicals? Some tiny clarification, please.
It’s also a very good point that all the signs are materialised, one way or the other. And in a very important way, you emphasise the difference between hylosemiotics and the New Materialism or multispecies studies by claiming that matter itself is not meaningful, nor that any material interaction automatically counts as semiosis: material does not constitute agency or meaning, but, according to hylosemiotics, it requires an interpreter. But not just any interpreter, but a self – something capable of acting on its own behalf. (I will come back to this in my summary below and ask a tiny clarification again.)
And now, there is an obvious tension between individualism and more emergent or holistic social kinds in your articulation of hylosemiotics. Your piece would very much benefit if you could clarify this tension between the individual, as an interpreting self, and the social kind. As I see it, and as a compassionate reader, these two are brought together by the concept of semiotic coordination or materialized signs. Semiotic architecture is the context- and situation-specific coordination context for individual interpretations and actions, emergent functions and social kinds, and the consequences of what happened. Jason, this is how I understand your thoughts. As you see, there might be a place for clarification. Anyway, interestingly, this is at the heart not only of classical pragmatism but also of the old institutional school of economics. Institutions scaffold individual action opportunities and potential, not directly, but mediated by practices and social kinds.
To summarise, first, your paper would benefit if you could clarify your thinking with the Peircean understanding of interpretant, i.e., not only as a self in sign process but also, or should I say, especially, the interpretant as an effect of sign process. Second, please clarify the relations between the individual, social kind, and semiotic coordination. This would help a reader to follow the thinking that you articulate, for example, in this wonderful quote: “The social kinds theory I’ve articulated does not ask only what individuals should do, but what kinds of semiotic and material infrastructures make certain actions possible in the first place. Rather than rooting ethics in an epistemic posture toward “the Earth system,” it asks how responsibility and transformation are routed through social kinds—some explicitly moral, some infrastructural, some sensory, some nonhuman.”
Third, this brings me back to the hardship of enabling biodiversity recovery on private lands in Finland, and beyond. Your text made me think about the kinds of semiotic and material infrastructures needed to coordinate individual forest owners and their lands into social kinds that would let biodiversity recover. Indeed, how to design such (policy) architecture, and what would, from policymakers’ perspectives, be the concrete, voluntary signs or cues to prompt voluntary or involuntary actions for biodiversity? Until now, due to a lack of workable policies, policymakers have mostly sent involuntary signs not to act for biodiversity. The challenge is to understand, you made me think, how certain social kinds are produced within complex assemblages of land, individuals, policies, and consequent semiotic coordination. (I would rather not use systems science language here.) As you can see, your piece opens multiple lines of thought. Please, give a moment to these three aspects and help people, like me, who try to apply this in theory and practice.
Thank you, Jason, for this wonderful piece of work.
Robert Yelle, LMU Munich
I read Jason Josephson-Storm‘s short essay with real interest and sympathy. (I also reread his long chapter on “Hylosemiotics” from his Metamodernism book, which is the basis for much of what he says in this short essay.) Let me interrogate the theory by probing some of its boundaries, where it might be improved. As I understand it, this is the purpose of this sort of exchange in any case.
“Hylosemiotics” seems an unhappy coinage to me. Josephson-Storm argues that all signs are material; and who would disagree? However, as he is aware and acknowledges, signs only exist for (Peircean) interpretants, which are sentient minds. When Robinson Crusoe finds a footprint on his island, he infers that another human being is there. An animal might do the same. There must be some visible mark, a criterion as it were, testifying to this fact. But without someone to make the inference, there is no sign. On this we agree. Such a semiotic transcends the material world while remaining embedded in it.
The impulse of the piece is avowedly to take seriously the fact that we inhabit a world of signs, and to extend signification throughout the animal (and even vegetal) kingdoms, as a mark of our kinship with all life, and an effort to transcend a narrow anthropomorphism and privileging of linguistic signs. For many if not most purposes, this feels right. Animals use tools, interpret signs and change their behavior accordingly, even lie (recall Umberto Eco’s definition of semiotics as a “theory of the lie”). Animals are sentient, and seem to feel pain. Of course, we exist on a continuum with other animals, being animals ourselves.
But should we call the signs and behaviors that animals produce “culture”? And should we equate such practices with the production of meaning? Those are more difficult questions to answer. The example of the corvid sculpting a leaf for the purpose of removing tasty insects from wood, with which Josephson-Storm begins his piece, is an example of pragmatic behavior. He notes that such behaviors are observed, imitated, and passed down within corvid groups. It seems significant that the transmission of this tool-making technique is limited to those in the immediate vicinity. There is no recipe book, no manual of instruction, that could be conveyed to other corvids and deciphered by them. That is because corvids possess neither language nor a system of writing.
E. B. Tylor defined culture famously as “that complex whole which includes knowledge, belief, art, law, morals, custom, and any other capabilities and habits acquired by man as a member of society.” How many of these components of Tylor’s capacious definition are found widely distributed in the animal kingdom? Several of them appear to refer to religion. There are cases of animal behaviors that are learned and that appear to have no practical function; such cases might be interpreted as rituals. There is still nothing like Neanderthal burials, which involved arrangement of the bones and added objects. H. sapiens burials were more complex yet, involving arrangement of bones in the fetal position and sometimes ochre. A symbol of rebirth? Perhaps. We can’t be sure. Conversely, the idea of an elephants’ graveyard appears to be a legend.
What about laws? Social behaviors are learned and presume the recognition of norms or customs. Does this yet constitute what we mean by a law? I am not sure. In human societies, laws are increasingly codified and language-bound. Moreover, these aim to restrain instinct. Famously, every human society appears to have some version of a prohibition against incest; even though this may be violated. Claude Lévi-Strauss devoted his dissertation to explaining this fact in structural and sociological terms. Animals do not observe any incest taboo, so far as I am aware.
I think it may be even more problematic to attribute “meaning” to the signs that animals produce (and consume) (see Metamodernism, 170-71, relying on von Uexküll). Animals adjust their behavior in accord with their surroundings or Umwelt, to be sure. But what we call meaning– the semantic as distinguished from the pragmatic function of language– seems to be specific to language, unless we adopt some overly expansive definition of meaning. Peacocks, like some other animals, have elaborate displays as part of mating rituals. But they do not have a system of fashion or haute couture, as humans do. Roland Barthes famously argued that this requires speech to be meaningful: “If we go beyond a few rudimentary signs…, can clothing signify without recourse to the speech that describes it, comments upon it, and provides it with signifiers and signifieds abundant enough to constitute a system of meaning? Man is doomed to articulated language, and no semiological undertaking can ignore this fact. … Thus, as soon as we observe Fashion, we discover that writing appears constitutive…” (Barthes, The Fashion System (Berkeley: University of California Press, 1990), xi).
Josephson-Storm is well aware of such claims for the uniqueness of language, and thus of meaning and culture. Yet he seeks to go beyond them, in the interest of outlining a more expansive semiotics that would extend the empire of signs throughout the animal kingdom (and beyond). I am sympathetic, and agree with his rejection of rigid dichotomies. However, it is easer to argue that certain behaviors inhabit the same cline or continuum, than it is to define the distinction between linguistic and non-linguistic modes of semiosis. The ritual verbal formulas called mantras in Asian cultures, as Frits Staal earlier pointed out, exhibit elaborate syntactic patterns. In this regard, they resembled bird songs, and like the latter were devoid of meaning, according to Staal. I have argued against some of Staal’s conclusions myself, by demonstrating that certain recursive and chiastic patterns in otherwise meaningless bija-mantras represented elaborate forms of symbolism: by rearranging the sequence of phonemes, they diagram various culturally conceived processes of creation. Mantras are an example of what the late Michael Silverstein called “indexical icons.” We can scarcely imagine anything of this complexity among animals that lack language. Such devices are key instantiations of what Silverstein’s teacher Roman Jakobson called the “poetic function,” and they transcend the divide between the phonemic, syntactic, semantic, and pragmatic dimensions of language. Can bird songs do the same?
I agree with Josephson-Storm’s contention that “Signification requires two things simultaneously: a material form in the world and a situated inferential act by a sentient being.” However, I do not concur that this is simply a restatement of the sentence immediately preceding, which contends that “Meaning arises when an interpreter takes [a material] trace as carrying meaning for them.” Meaning requires something more than signification, and that something more is bound up with language. Although signification extends throughout the animal kingdom, only humans appear to possess language. The fact that behaviors inhabit the same continuum is not enough to establish their identity. Even when the difference is scalar rather than binary–analog rather than digital–a difference in degree may amount to a difference in kind. Presumably, this is analogous to the threshold at which a complex system attains consciousness. I note that we are still unable to define this threshold.
Animals can coordinate their behaviors as social beings. The dances of bees may diagram and map the route to a food source. Bees have been regarded as social animals par excellence, from classical antiquity to Hobbes and Mandeville. But actions learned and reinforced by pragmatic results do not seem to rise beyond the level of what can be accounted for in behavioristic terms (although Josephson-Storm disavows B. F. Skinner’s model, in Metamodernism, 171).
Josephson-Storm appropriately seeks to steer the narrow path between “the Scylla of anthropomorphism and the Charybdis of anthropocentrism” (quoting Pearson and Weismantel). However, I wonder if his theoretical model has managed to steer clear of language, the stone wall that divides human from animals?
Author response to reviews
Thank you for the thoughtful, generous, and intellectually engaged responses to the article. I very much appreciate the care with which both reviewers read the piece.
I am grateful to Juha Hidenpää for his generous and constructive feedback. In response, I have made several targeted clarifications intended to sharpen rather than expand the argument. First, I revised the formulation of sender meaning to avoid any implication of subjective “hopefulness,” specifying instead that signs may be oriented toward eliciting interpretation without presuming reflective intention, particularly in nonhuman cases. Second, I added a brief clarification of the Peircean interpretant, emphasizing it as the inferential effect of a materialized sign within a situated semiotic ecology, along with a short plain-language gloss to improve accessibility. Third, I inserted a bridging sentence clarifying how individual interpretation relates to social kinds through facilitating the coordination of materialized signs. Finally, I included a short illustrative sentence in the conclusion connecting the framework to biodiversity policy in order to make the potential applied implications more explicit while keeping the article primarily theoretical.
I would also like to thank Robert Yelle for a careful and substantively engaged critique. Given his own important contributions to the study of semiotics and religion, I especially appreciate the seriousness with which he has taken the argument and his willingness to revisit the Metamodernism chapter in order to deepen that engagement. This is very much appreciated.
His comments really helped me identify places where implicit definitional assumptions might create unnecessary friction for readers. I have made several modest clarifications accordingly without altering the paper’s central claims. In the introduction, I now provide a working definition of culture as socially transmitted patterns of coordination mediated by materialized signs across generations. This makes explicit the sense of “culture” likely to be most productive for readers of Cultural Science and clarifies why intergenerationally stabilized nonhuman practices, such as corvid tool traditions, fall within the scope of the argument. I have also added a few additional ethological references on socially transmitted behavioral traditions across species to strengthen the empirical grounding of the discussion and to show that the corvid example is part of a wider evidentiary pattern. In the hylosemiotics section, I restate more explicitly that the account treats meaning as situated inference rather than the decoding of self-contained semantic contents, thereby clarifying that the framework is not limited to language-bound semantics. I also added a brief sentence acknowledging distinctive human linguistic affordances—recursion, symbolic abstraction, and cumulative scale—while emphasizing that these are not treated here as prerequisites for semiosis or culturally mediated coordination. Finally, I include a short clarifying line distinguishing inferential semiosis from simple stimulus-response conditioning (while acknowledging the empirical difficulty of drawing that boundary from the outside), since this distinction is one of the pressure points his review helpfully highlights.
On the remaining point of disagreement: Professor Yelle’s concerns largely turn on whether “meaning” and “culture” should be reserved for language-centric phenomena. The article deliberately questions that restriction. I fully agree that human linguistic capacities enable forms of cumulative symbolic complexity unmatched elsewhere, and nothing in the essay is intended to minimize those differences. The aim, rather, is to reconsider whether language must function as the gatekeeping criterion for semiosis or culture as such. My proposal is that semiosis and culturally stabilized coordination can be analyzed across species without collapsing differences or lapsing into anthropomorphism. I hope the added clarifications make that boundary-setting clearer, even where our conceptual preferences ultimately diverge.
I am grateful to both reviewers for the care and generosity with which they engaged the piece and for the opportunity to refine the argument in response.