Split Worlds with Incommensurably Paired Epistemologies: Defining SWIPEs (Series: SWIPEs 1/5)

Binaries are ancient, SWIPEs are recent. This article defines a SWIPE, a Split World with Incommensurably Paired Epistemologies, and distinguishes it from the ordinary oppositions that human classification generates everywhere. Four characteristics are jointly necessary: an analytic difference is reified into independently constituted domains; different legitimate ways of knowing become paired with those domains; transfer of method across the division becomes restricted or unidirectional; and the partition becomes prior to inquiry rather than a finding inquiry might revise. Candidate cases include nature/culture, mind/body, subject/object, fact/value, individual/society, human/nonhuman, matter/meaning, disease/illness, race/culture, and sex/gender, none of which is assumed here to pass the test. The historical argument is that SWIPE formation depends on something more specific than dualistic philosophy. It requires specialized empirical research regimes capable of claiming epistemic jurisdiction over particular parts of reality, a condition that consolidates only in the nineteenth century with laboratories, archives, seminars, professions, journals, and the research university. The positionality argument is that a SWIPE need not be consciously endorsed by anyone assigned to its poles; it can reside in institutional arrangements, disciplinary jurisdictions, pedagogical simplifications, and the descriptions of those who observe disciplines from outside. The dynamic argument is that SWIPEs have very different recursive careers. Some are weakly defended conventions. Others become protective settlements whose reopening is associated with feared political or moral consequences, a condition here called protective sedimentation. SWIPE theory is therefore not a taxonomy of binaries but an account of how distinctions become worlds, how worlds acquire epistemologies, and how the resulting arrangements become difficult to think outside.

Introduction: Who Is Doing the Swiping?

A medical anthropologist distinguishes disease from illness. A philosopher of mind asks how mental states relate to neural ones. A sociologist asks how individual agency relates to social structure. A feminist theorist distinguishes sex from gender. An anthropologist insists that human variation is cultural rather than racial. A historian of science describes modern knowledge as divided between nature and culture.

These look like six different problems, belonging to six different literatures, with six different bibliographies. Each has its own canonical texts, its own controversies, its own sense of what would count as progress. The proposal advanced here is that they may share a single classificatory operation, and that identifying the operation precisely matters more than adding another entry to the long list of dualisms modern thought is accused of harbouring. The list is already long, and lengthening it has not proved illuminating.

The first thing to establish is that "binary" is far too broad a category to do any work. Human cognition, language, and social classification generate oppositions relentlessly. Hot and cold. Alive and dead. Inside and outside. Near and far. Before and after. Raw and cooked. None of these is an epistemological catastrophe, and a theory that treated every polarity as a pathology would have identified nothing more interesting than the fact that people make distinctions.

Nor is an ontological distinction sufficient, however grand. Descartes can separate thinking substance from extended substance, one of the most consequential ontological cuts in the history of European philosophy, while proposing a single method of knowledge for the whole of the resulting territory. A companion genealogy of the nature/culture case argues this point at length: a binary may precede its epistemological alignment by centuries, and an epistemological differentiation may in turn precede its institutionalization as a settled partition (Ecks 2026d). Dualism is not enough. Something further has to happen.

So the foundational question is this. What has to happen to a distinction before it becomes a SWIPE? And immediately behind it stands a historical question that most critiques of dualism never ask: what kind of world has to exist before swiping is even possible?

The answer offered here is not that Western philosophy has always been dualistic, which is both too vague and too flattering to philosophy. Nor is it that Descartes did it, an attribution that survives mainly because it is easy to teach. The answer is that SWIPEs become possible when distinctions can be attached to sufficiently differentiated regimes of knowledge production that those regimes acquire jurisdiction over different parts of reality. Jurisdiction is the operative concept, borrowed from the sociology of professions, where it names a profession's claim to authority over a domain of problems and the competitive processes by which such claims are won, defended, and lost (Abbott 1988). A SWIPE is what results when jurisdictional claims stop being claims about who is competent and start functioning as claims about what there is. This is why the nineteenth century is decisive, and why no amount of pre-nineteenth-century metaphysics quite delivers the phenomenon.

Two clarifications forestall predictable objections. The claim is not that nothing resembling a partition existed before the nineteenth century, only that the specific structure defined here did not, and that earlier candidates fail identifiable conditions rather than merely being weaker versions of the same thing. Nor is the claim that specialization is regrettable. Specialized empirical research regimes are among the most productive arrangements ever devised, and the argument here concerns a particular thing that can happen to them, not the fact of their existence.

The second surprise concerns who holds a SWIPE, and it may be the more consequential of the two. The people supposedly inhabiting a partition are frequently not the people who articulate it. A molecular biologist may pass an entire career without invoking nature and culture. An anthropologist describing molecular biology may invoke them constantly. This asymmetry is not incidental. It suggests that the analysis of a SWIPE must distinguish carefully among holding one, carrying one, reproducing one, suffering one, diagnosing one, and defending one, because these are different relations to the same structure and they are held by different people, often with no contact between them. It also introduces a methodological danger that no critique of dualism can afford to ignore: a sufficiently enthusiastic diagnostician can produce the partition in the act of describing it.

I. What Exactly Is a SWIPE?

A SWIPE is a Split World with Incommensurably Paired Epistemologies. Four conditions are jointly necessary, and a distinction lacking any one of them is not a SWIPE, whatever else it may be.

Ontological reification. An analytic difference is transformed into two independently constituted domains. The problem is not differentiation itself. The problem is the move from A differs from B to A and B are separate kinds of reality, each imaginable, in principle, without the other. Distinguishing the chemistry of a drug from the meaning of taking it is useful. Concluding that pharmacology and meaning belong to different orders of being is the reifying step.

Epistemological pairing. Different legitimate ways of knowing become attached to the two sides. Nature draws measurement, causal analysis, experiment, and the search for general law; culture draws interpretation, contextual reconstruction, and the search for meaning. A SWIPE cuts epistemology at exactly the place it cuts ontology, and the alignment of the two cuts is what gives the structure its stability.

Restricted transferability. Methods appropriate on one side become inappropriate, deficient, or unintelligible on the other. This is where the I in the acronym earns its place, and it needs a qualification. The epistemologies need not be incommensurable in Kuhn's sense, which concerns rival paradigms competing for the same domain, cross-classifying the same entities into mutually exclusive kinds, with no neutral observation language or shared algorithm for theory choice (Kuhn 1962). Kuhn's incommensurability is diachronic and intra-disciplinary: it holds between successors and rivals. What a SWIPE installs is synchronic and inter-disciplinary, and it is better described as jurisdictional incommensurability. The two epistemologies coexist peacefully. They simply do not travel. A naturalist's explanation of a poem and a historian's explanation of a chemical reaction are both, in the relevant institutional settings, category mistakes rather than merely bad arguments.

Priority to inquiry. This is the strongest and most diagnostic criterion. The distinction is no longer something research might discover or overturn; it structures research before investigation begins. Researchers allocate a phenomenon to one side and then ask what follows, rather than asking whether the phenomenon warrants the allocation at all. A partition still experienced as a philosophical problem has not yet reached this condition. Kant's chasm between nature and freedom is enormous, explicit, and load-bearing, and Kant spends an entire Critique trying to mediate it; that effort is evidence that the partition had not yet become prior to inquiry in the required sense (Ecks 2026d).

It follows that a number of things resembling SWIPEs are not SWIPEs. A polarity is not a SWIPE. A gradient is certainly not. A difference in methods is not, since methods differ for all sorts of practical reasons that carry no ontological freight. Disciplinary specialization is not, and neither is ontological or methodological pluralism, both of which may be entirely reasonable responses to a heterogeneous world. The distinction that matters can be put in a sentence. Specialization says that different questions may require different techniques. A SWIPE says that different techniques belong to different worlds.

A worked negative case clarifies the test. Consider the difference between studying an earthquake seismologically and studying it as a social catastrophe. Two literatures exist; two sets of methods are used; two professional communities are involved; publications rarely overlap. This satisfies none of the four conditions on inspection. Nobody holds that the seismic and social aspects are separate kinds of reality, since everyone agrees there is one event. Nobody holds that seismological method is inappropriate to the social question in principle, merely that it does not answer it. And nobody begins by allocating the earthquake to one domain and then asking what follows; the allocation is made per question, revisably, and disaster research routinely combines the two without anyone reporting a category violation. Division of labour, however entrenched, is not partition. If the framework cannot distinguish this case from disease/illness, it is not measuring anything.

In the vocabulary of Living Value Theory, a SWIPE is an L4 structure: a stabilized classificatory arrangement capable of organizing institutions, curricula, evidentiary standards, and habitual reasoning (Ecks 2026a). This is what allows a SWIPE to persist without anyone continually asserting it, and it is why the analysis cannot remain at the level of propositions held by individuals. An L4 arrangement is carried by departments, examinations, funding categories, and journal scopes at least as much as by anybody's beliefs.

The candidate family can be set out provisionally.

These are candidates, not ten established results. Each requires its own genealogy, and the four-part test must be capable of returning a negative verdict. A framework that discovers SWIPEs everywhere it looks has become exactly the kind of overburdened L4 classification it was designed to diagnose.

II. Before SWIPE: Why Ancient Binaries Were Not Enough

Binary thinking is old, and the point of the following is not to minimize its force.

Fifth-century Greek thought argues fiercely over physis and nomos, what exists by nature against what exists by custom, convention, or law. Sacred and profane organize ritual life across an enormous range of societies. Distinctions between soul and body, or between the perishable and the imperishable, structure ancient philosophy and theology alike. These are not weak distinctions. Some of them were worth dying for.

What none of them delivers is paired epistemic jurisdiction. Nothing in Greek thought installs a doctrine that physis must be studied through mathematical-causal explanation while nomos requires a hermeneutic method unavailable to students of nature; the very idea of two methodologies answerable to two epistemic constituencies is absent, and ancient knowledge practices moved across the supposed boundary constantly, with Hippocratic medicine reasoning about winds, waters, places, custom, and constitution in a single register (Ecks 2026d). A very strong binary, in other words, is fully compatible with a completely unified epistemology.

The sacred/profane distinction is worth a separate word, because it is the case most likely to be offered as a counterexample. It is ancient, it is ontologically strong, and Durkheim's formulation is about as absolute as such formulations get: in all the history of human thought, he writes, there exists no other example of two categories of things so profoundly differentiated or so radically opposed to one another, conceived as two worlds between which there is nothing in common (Durkheim 1995, Bk. I, ch.1). Yet even here the second and third conditions fail. No ancient or medieval tradition holds that the sacred requires an epistemology structurally unavailable for the profane; the same ritual expertise, the same textual authority, and the same reasoning apply to both, and the specialists who handle the sacred are usually the same people who adjudicate the profane. Absolute ontological heterogeneity, it turns out, is compatible with complete epistemological unity. That combination is precisely what modern partitions do not permit, and noticing it is the quickest way to see that the four conditions are not redundant.

Medieval universities divided knowledge extensively, and the divisions were institutionally serious: theology, law, and medicine as higher faculties, with the arts faculty covering grammar, rhetoric, logic, arithmetic, geometry, music, and astronomy. Career paths ran through them; jurisdictional disputes occurred. But a hierarchy of disciplines inside a shared epistemic order is not a partition of reality into mutually inaccessible domains. Natural philosophy sat within that order rather than against it, and the ordering principles were theological and pedagogical rather than anything resembling a scientific/humanistic split. Crucially, the medieval hierarchy is vertical, arranging disciplines by dignity and by the sublimity of their objects under a single conception of knowledge, whereas a SWIPE is horizontal, placing domains beside one another with no common ground at all. A ladder and a wall are different structures, and only the second is at issue here.

Early modern dualism supplies materials rather than the finished structure. Descartes writes across optics, meteorology, geometry, physiology, and the passions of the soul without ever suggesting that these require different epistemic procedures; the method of clear and distinct ideas is offered for the whole tree of philosophy. Cartesian dualism is therefore an ontological differentiation joined to an emphatically monist epistemology, which is nearly the opposite of a SWIPE. This is a substantial correction to the conventional intellectual history in which Descartes appears as the origin of every subsequent modern partition.

The eighteenth century assembles further components without assembling them into a technology. Vico supplies a differential epistemic accessibility, arguing that the world of nations can be known with a certainty unavailable for the world God made. Rousseau intensifies nature against civilization into a live normative axis. Kant produces the nature/freedom gulf and immediately sets about bridging it. And, on a separate track, the word "culture" completes a grammatical revolution, ceasing to require an object, becoming a free-standing and then pluralizable noun (Williams 1983; Ecks 2026b). Each of these matters. None of them, individually or together, produces two domains with two nontransferable epistemologies. Vico's asymmetry is a metaphysical claim about certainty, not a rule about who may study what. Rousseau reasons about human nature and social institutions with the same blend of historical speculation and moral argument throughout, shifting register nowhere. And a freed noun, however consequential, is a lexical fact rather than an epistemological one.

There is also comparative evidence that the pair is not a universal feature of human classification. Ethnographic work has shown that the nature/culture opposition, and the gendered symbolism frequently attached to it, cannot be located in some societies at all, which is a serious difficulty for any account treating the partition as a cognitive inevitability rather than a historical product (Strathern 1980). Broader comparative ontology reaches a compatible conclusion, treating the modern Western configuration, naturalism, in which physicalities are assumed continuous across beings while interiorities are treated as discontinuous, singling humans out, as one arrangement among four rather than the default from which others deviate (Descola 2013). If the partition were a deep feature of human cognition, it should be far more widely distributed than it is.

The lesson generalizes beyond the nature/culture case, and it is worth stating as a rule for the whole series. A binary may precede epistemological differentiation. Epistemological differentiation may precede institutional partition. The three achievements are separable, they occur on different timescales, and reading the third back into the first is the characteristic error of genealogies written from inside a settled classification.

So what was still missing around 1800? Not empirical evidence: observation, measurement, dissection, collection, and experiment all long predate the nineteenth century, and it would be absurd to suggest that earlier natural philosophers had no evidence. What was missing was the infrastructural organization of specialized empirical knowledge production, and with it the possibility that different communities could hold recognized, defensible, and mutually exclusive authority over different classes of object. That condition is the subject of the next section.

III. When the World Became Swipeable: The Nineteenth-Century Conditions

The decisive transformation is the rise of research as an organized profession, and the point is not that earlier scholars lacked evidence but that evidence production itself becomes infrastructurally organized.

Increasingly, across the nineteenth century, legitimate scholarship requires specialized training, standardized methods, instruments or documentary materials, systematic record-keeping, publication in specialized venues, evaluation by peer communities, and an institutional location. In the Prussian professoriate, research productivity came over the first half of the century to function as an increasingly decisive criterion for appointment and advancement, operating alongside patronage, faculty recommendation, and teaching reputation rather than simply replacing them, and driven as much by career structure and competition among states for professors as by neo-humanist ideology (Turner 1971). Scholarship becomes a career with measurable outputs. Weber, lecturing in Munich in 1917, could already describe the consequence as settled: "A really definitive and good accomplishment is today always a specialized accomplishment" (Weber 1946, p.135).

Laboratory epistemology develops in exactly this period. Liebig arrived at Giessen in 1824 and converted a disused artillery guardhouse into a chemical laboratory, running instruction as a private, fee-paying institute for pharmacy outside the university budget until the state took it over as a university institute in 1833; the building acquired its lasting form in 1839 with the addition of an analytical laboratory, a pharmaceutical laboratory, a library, and an auditorium. Giessen became the exemplary case of collective, instruction-based chemical research training, though recent scholarship has substantially complicated the story in which one man's genius invents the research laboratory, showing instead a fusion of teaching and research that consolidated gradually and depended on specific analytic techniques and apparatus (Morrell 1972; Holmes 1989; Rocke 2023). Physiology, experimental medicine, and physics follow comparable trajectories, with the Königsberg physics seminar under Franz Neumann producing not merely results but a transmissible practice of precision measurement, error analysis, and least-squares reduction, taught as a discipline (Olesko 1991). Knowledge ceases to be principally something a learned person argues for and becomes something a specialized apparatus, staffed by trained personnel, produces and warrants.

Here the analysis must resist a temptation that has disfigured most accounts of the two cultures, namely the equation of science with data and the humanities with speculation. Nineteenth-century history and philology became fiercely, competitively empirical. The philological seminar developed transmissible techniques of source criticism, textual comparison, manuscript collation, chronology, and documentary authentication, and it did so well before Giessen: Wolf's Seminarium Philologicum at Halle from 1787 is the best-known instance but not the first, since Göttingen had a philological seminar from 1737 under Gesner, directed by Heyne from 1763 (Clark 2006). Archives were organized, catalogued, and opened. Editions were established on explicit principles. Professional infrastructure followed: the Historische Zeitschrift in 1859, the American Historical Association in 1884, the English Historical Review in 1886.

Two empirical regimes therefore develop simultaneously, and this simultaneity is the crux. The nineteenth-century problem is not evidence against no evidence. Everyone increasingly agrees that claims require evidence. Disagreement grows instead over what counts as evidence, and the candidates multiply: a tissue section, a calibrated measurement, a textual variant, an archival document, a reported intention, a statistical regularity, an observed behaviour, a narrated experience. Plural empirical regimes, each internally rigorous, each with its own criteria of adequacy, are the actual condition of possibility for what follows.

What such regimes install is not merely a preference but a trained perception. Fleck's terms are exact and worth using: a Denkkollektiv, a community maintaining intellectual interaction, shares a Denkstil, which he defines as a readiness for directed perception together with the corresponding assimilation of what is perceived, characterized by the problems it finds interesting, the judgments it treats as evident, and the methods it applies as means of cognition. The consequence he draws is precisely the one this essay needs: what is a fact to the members of one thought collective sometimes does not exist for the members of another (Fleck 1979). Training produces competence and its limits together. This explains how a partition can operate without being believed. Nobody need assert that the archive and the laboratory address different realities; it is enough that practitioners trained in one cannot readily recognize the other's output as evidence bearing on their own question. Nor is such differentiation a neutral sorting. The scientific field, on Bourdieu's account, is the locus of a competitive struggle in which the specific stake is the monopoly of scientific authority, defined inseparably as technical capacity and social power, or equivalently the socially recognized capacity to speak and act legitimately in scientific matters (Bourdieu 1975, p.19).

From plural regimes comes the transition that matters: method becomes competence, and competence becomes jurisdiction. A laboratory scientist does not merely know something; she holds a recognized authority grounded in techniques unavailable to nonspecialists, defended against neighbouring claimants, and institutionally certified. The historian acquires a different jurisdiction, the philologist another, the clinician another. Professions compete for these territories, and on Abbott's account jurisdictional claims are made and contested in three arenas of differing formality: the workplace, where settlements are often informal and blurred; public opinion; and the legal system, the most rigid of the three (Abbott 1988). Practitioners also participate actively in drawing the boundaries, attributing selected characteristics to the institution of science itself, to its practitioners, methods, stock of knowledge, values, and work organization, in order to construct a social boundary marking off some activities as non-science, a rhetorical and institutional labour Gieryn calls boundary-work (Gieryn 1983, p.782).

The dangerous step is now available. Instead of saying we happen to use different methods, one can say we use different methods because our objects are different kinds of reality. Jurisdiction becomes ontology. That conversion is the birth condition of SWIPE, and it can be set out as a sequence:

empirical intensification → methodological specialization → professional jurisdiction → epistemological differentiation → allocation of domains → ontological reification → restricted transferability → SWIPE.

Nothing in this sequence is automatic, and each arrow is a historical accomplishment rather than a logical entailment. But the sequence explains why the phenomenon is late. It requires laboratories, hospitals, archives, museums, universities, departments, libraries, journals, professional associations, examinations, funding structures, and training regimes. A SWIPE therefore has a multimaterial history and a dwelling history as much as a symbolic one, which is a considerably stronger explanation than the observation that some philosophers became dualists.

It also explains something the philosophical account cannot: why these partitions are so resistant to argument. A doctrine can be refuted. An arrangement of buildings, appointments, degree requirements, and evidential conventions cannot, because nothing in it is asserted and therefore nothing in it can be contradicted. The refutation lands in a journal, and the journal is itself one of the instruments by which the division is maintained.

IV. From One SWIPE to a Family: A Comparative Morphology

Once SWIPE exists as a classificatory technology, it becomes available for repeated use, and what follows compares forms of derivation rather than offering ten miniature histories.

Foundational cases. Nature/culture is the master case, and its genealogy yields a result that constrains everything else: no single thinker authored the settlement, which emerged through distributed institutional differentiation, translation accidents, pedagogical simplification, and retrospective alignment (Ecks 2026d). Mind/body is the sharpest test of the nineteenth-century thesis, because the ontological dualism is very old while the SWIPE, if there is one, should date only from the point at which the body acquires experimental physiological and clinical jurisdictions and mind acquires other modes of access. The prediction is precise and falsifiable: if the mind/body SWIPE is Cartesian, it should be visible in seventeenth-century practice; if it is jurisdictional, it should appear only once physiology, neurology, and psychiatry become distinct professions with distinct training, laboratories, and licensing, which is to say in the second half of the nineteenth century. The subsequent history of psychosomatic medicine, a field that exists precisely because the allocation keeps failing, is the natural place to look for confirmation.

Subject/object may be more foundational still, but the same discipline applies: the partition should be dated not to the first philosophical distinction between knower and known but to the moment at which "objective" and "subjective" become names for epistemic jurisdictions with institutional homes. That transformation is now well documented, and its timing is instructive, since the ideal of a self-effacing observer whose personal judgment is a source of error rather than authority is itself a mid-nineteenth-century development with a traceable history in scientific atlases and image-making practices (Daston and Galison 2007). Objectivity, in other words, is not the timeless opposite of subjectivity but a historically specific epistemic virtue that had to be invented, trained, and enforced. One qualification their account insists on matters for this framework: mechanical objectivity did not replace the earlier virtue of truth-to-nature, nor was it replaced in turn by trained judgment. The virtues accumulate and coexist. That is a useful caution against assuming that a new epistemic regime displaces its predecessor rather than being added alongside it, and Section VII returns to the point in another form.

Fact/value is a particularly instructive case, because an epistemic difference is reified into a supposed feature of reality itself: facts are found, values are added. That the dichotomy collapses under examination, not least because thick concepts such as cruel or brave are simultaneously descriptive and evaluative, has been argued at length without much disturbing its institutional life (Putnam 2002). The gap between the strength of the refutation and the stability of the arrangement is itself a datum, and it anticipates the argument of Section VII: philosophical demolition is remarkably ineffective against a partition that is institutionally load-bearing.

Disciplinary cases. These take a general architecture and install it inside a particular field. Disease/illness is the cleanest example: disease receives biomedical jurisdiction, illness receives interpretive jurisdiction, and the two are assigned to different professions with different training (Kleinman 1980). The critical point is that this distinction did not invent the split. It imported an architecture that had already spent a century sedimenting in the modern university and applied it to the body of the sick person (Ecks 2026c). Individual/society divides psychological from sociological jurisdiction, and the extraordinary volume of structure-and-agency theorizing may itself be evidence of repeated repair work around a cut that keeps failing to hold.

Boundary-of-the-human cases. Human/nonhuman allocates meaning, intentionality, and history to one side and behaviour, mechanism, adaptation, and causation to the other, an arrangement made increasingly unstable by ethology, animal cognition research, and multispecies ethnography. The instability is diagnostic: when a chimpanzee's behaviour has to be described either as culture or as adaptation, and the choice determines which journal will consider it, the allocation is doing ontological work under the guise of a disciplinary convention. Matter/meaning places causal materiality on one side and signification on the other, and here the useful observation is that the subsequent material turn in the human sciences is interpretable as recursive reopening, though whether it reopened the partition or merely renegotiated its terms is exactly the sort of question the framework should be able to ask. A turn that consists in insisting that matter, too, has agency may be reasserting the very inventory it means to overturn, since it accepts that the world comes sorted into matter and meaning and then redistributes properties between the two piles.

Politically protective cases. Race/culture and sex/gender belong in a class of their own, and the reason is not that they are stronger SWIPEs but that their stability has an additional source. The Boasian separation of race from culture did indispensable anti-determinist work, establishing culture as a phenomenon dependent on historical causes regardless of race (Boas 1911; Stocking 1968). That achievement can itself sediment the distinction, and the sedimentation has costs that later anthropology has had to confront. Visweswaran's argument is sharper than a claim about neglect: the attempt to expunge race from social science by assigning it to biology, as Boas and his students did, helped legitimate the scientific study of race, an active disciplinary division of labour rather than an oversight (Visweswaran 1998, p.70). Meanwhile "culture" proved perfectly capable of doing exclusionary work in its own right, functioning, in Balibar's formulation, like a nature, and in particular as a way of locking individuals and groups a priori into a genealogy, into a determination that is immutable and intangible in origin (Balibar 1991, p.22; Stolcke 1995).

Sex/gender has a comparable structure. The distinction between bodily classification and socially acquired role was elaborated in mid-century clinical psychology, where "gender role" was defined and operationalized, and then imported into feminist theory, opening the space in which biological determinism could be challenged (Money, Hampson and Hampson 1955; Stoller 1968; Oakley 1972; Rubin 1975). Subsequent work destabilized the clean allocation from within. Butler's formulation is the sharpest: if the immutable character of sex is contested, perhaps this construct called sex is as culturally constructed as gender, "indeed, perhaps it was always already gender, with the consequence that the distinction between sex and gender turns out to be no distinction at all" (Butler 1990, p.7). Others documented how bodily sex is itself produced through classificatory and clinical practice (Fausto-Sterling 2000; Germon 2009).

These two cases also show why the framework must resist a certain kind of triumphalism. Both partitions were built by people trying to prevent a specific harm, and both largely succeeded. To observe that a distinction has hardened into a jurisdictional arrangement is not to recommend its removal, and the analysis of a protective partition is incomplete until it has stated clearly what the partition is protecting against and whether anything else would do the job.

Family resemblance without reduction. It is tempting to posit a master SWIPE from which the others descend, and the temptation should be refused. Nature/culture plausibly influences disease/illness. Mind/body may reinforce subject/object. Race/culture and sex/gender may recruit nature/culture for authority. But these are historical relationships to be demonstrated, case by case, not algebraic derivations to be assumed. SWIPE theory is genealogical rather than deductive, and a genealogy that already knows its conclusion has stopped being one.

V. Who Is Doing the Swiping?

The ordinary assumption is that if a discipline operates within a distinction, its practitioners believe in it. This is far too mentalistic, and correcting it opens the most productive line of inquiry the framework offers.

A molecular biologist need hold no explicit theory of nature and culture. A surgeon need not endorse Cartesian dualism. A historian need not consciously affirm understanding against explanation. Yet the institutions in which all three work may reproduce jurisdictional boundaries with great precision, through curricula, examinations, hiring lines, journal scopes, and funding categories. The SWIPE can be fully operative while nobody in the vicinity holds it as a proposition.

Six positions can be distinguished, and they should not be collapsed.

The doctrinal holder explicitly argues for the partition. This is the easiest case to document and perhaps the least historically important, since doctrinal statements are comparatively rare and frequently postdate the arrangements they justify.

The institutional carrier is not a person at all. Departments, degree structures, professional bodies, journals, funding schemes, ethics review categories, and legal classifications reproduce the partition without requiring propositional belief from anyone.

The practical reproducer works inside the arrangement without reflecting on it. They do not believe the SWIPE; they encounter a settled division of labour and get on with the work. This is the largest group by far, and treating them as adherents is a straightforward error of attribution.

The boundary sufferer studies objects that repeatedly cross jurisdictions. Medical anthropology is paradigmatic, but so are psychosomatic medicine, psychiatry, environmental humanities, human-animal studies, and much of science and technology studies. For these researchers the partition is a daily practical obstacle rather than a theoretical position.

The external diagnostician names the SWIPE. This is usually an anthropologist, historian, philosopher, sociologist, or STS researcher, and the diagnosis is typically made from outside the disciplines being described.

The political custodian actively defends the partition because of what they believe would follow from its collapse. Race/culture and sex/gender make this position unusually visible, but it exists wherever a distinction is doing protective work.

These positions are not ranked, and a single person can occupy several at once, or move between them over a career. A clinician may be a practical reproducer at work and an external diagnostician in print. More importantly, the positions are frequently not in communication. The diagnostician typically addresses other diagnosticians; the institutional carrier has no mechanism for receiving an argument; and the practical reproducer is rarely aware of being described at all. This is one reason the critical literature on dualism has accumulated for a century with so little institutional consequence. It has been addressed almost entirely to the one position that was never holding the partition in place.

From this typology follows a hypothesis worth testing across cases: a SWIPE may be least recursively salient at its centres and most salient at its boundaries. A physicist can work productively for decades without the nature/culture partition ever becoming an issue, because nothing in the daily practice of the discipline brings the allocation into question. A medical anthropologist cannot avoid it, because the ethnographic object violates the allocation continuously and in ways that cannot be deferred.

This raises an uncomfortable possibility about the nature/culture case specifically. The partition is disproportionately articulated from the culture side. Dilthey asked how the human sciences should relate to natural science; anthropology and STS have asked versions of the same question ever since. Natural science, for the most part, has simply proceeded. It is therefore worth entertaining the hypothesis that the nature/culture divide is partly a self-description of the human sciences produced under conditions of comparison, and that the description has been mistaken for a shared settlement to which both sides subscribe.

Which leads to the framework's central methodological danger. A critic can manufacture a SWIPE in the act of diagnosing it. If an analyst declares all laboratory work to be Nature and all interpretive work to be Culture, and then criticizes scientists for separating nature from culture, the analyst has performed the partition and attributed it to others. Latour's account of modernity as the simultaneous work of purification, which creates two entirely distinct ontological zones, that of human beings and that of nonhumans, and translation, which creates mixtures between wholly new types of beings, hybrids of nature and culture, is illuminating precisely here, and also exposed to the same risk, since the purified poles have to be specified by the analyst before the hybrids can be counted (Latour 1993). A hybrid is only hybrid relative to a prior purity, and if the purity is supplied by the description rather than found in the practice, the analysis has generated its own object.

Evidence that a SWIPE is operating cannot therefore consist of critical literature asserting that one exists. Nor can it consist of the observation that two disciplines study different things using different methods, which is true of nearly every pair of disciplines and establishes nothing.

What would count as evidence? A workable protocol includes: explicit methodological prohibition; formal institutional jurisdiction; separated training pathways; incompatible evidential standards; divided professional responsibility for the same phenomenon; systematic non-formulation of certain questions; classification embedded in funding categories or publication scopes; and the recurrent treatment of cross-boundary phenomena as anomalies requiring special handling. These are observable. They can be found or not found. Every subsequent article in this series should be answerable to them.

VI. How SWIPEs Are Recursively Opened

A sedimented partition can become recursively available again when something disrupts its fit, and the mechanisms are various. Opening is not intrinsically emancipatory, and nothing in what follows implies that a reopened SWIPE is thereby improved.

Empirical opening occurs when the phenomenon refuses its allocation. Disease turns out not to be separable from experience. Bodies turn out to vary with social conditions in ways that biology and history cannot address separately. Lock's comparison of Japanese kōnenki with North American menopause found hot flashes reported by around a eighth of Japanese women aged forty-five to fifty-five against roughly a third in Manitoba and Massachusetts, with stiff shoulders rather than hot flashes the salient complaint, a difference she attributes to diet, endocrine baseline, and cultural framing acting together rather than to biology alone or to reporting convention alone (Lock 1993). Animals display capacities reserved for the human side. Scientific facts turn out to depend on material and social arrangements, which was the burden of the experimentalist historiography showing that matters of fact are produced through material, literary, and social technologies that each embed the others (Shapin and Schaffer 1985), and of Rheinberger's account of experimental systems as systems of manipulation designed to give unknown answers to questions that the experimenters themselves are not yet able clearly to ask (Rheinberger 1997, p.28). Empirical recalcitrance is probably the most powerful opening mechanism available, because it does not depend on anyone finding the partition philosophically objectionable. The phenomenon simply keeps producing results that neither jurisdiction can absorb, and the pressure accumulates in the practical business of publishing, funding, and explaining anomalies to colleagues.

Boundary opening occurs where researchers positioned between disciplines encounter incompatible jurisdictional requirements as a practical condition of work. A researcher must satisfy two sets of reviewers with two conceptions of adequate evidence, or choose one and accept marginality in the other. Boundary fields therefore function as engines of recursive opening, and their theoretical output is often better read as a record of institutional friction than as free-standing speculation. It also follows that boundary work has characteristic pathologies, since a field that lives on a fault line has strong incentives to keep the fault line interesting, and may become invested in the partition it complains about.

Conceptual opening is the classic route, in which a theorist asks whether the division is coherent. Whitehead's protest is exemplary: he objects to bifurcating nature into, on one side, the nature apprehended in awareness, holding within it the greenness of the trees and the warmth of the sun, and on the other the conjectured system of molecules and electrons that causes the awareness (Whitehead 1920). The objection is not that the two descriptions differ but that treating them as two natures makes the relation between them insoluble by construction. So too Putnam on fact and value, and Strathern's demonstration that the pair could not be located in Hagen at all (Strathern 1980). Conceptual critique matters, but it is only one route and probably not the most consequential, since arguments can be answered, absorbed, or simply ignored in ways that recalcitrant phenomena cannot. Whitehead's objection is more than a century old and has changed no curriculum.

Technological opening occurs when new instruments cross previously separate jurisdictions. Neuroimaging unsettles simple interior/exterior distinctions by producing images of something that had been defined as unobservable from outside. Genetic technologies can destabilize race/culture allocations or, just as easily, re-entrench them under new descriptions, since population-genetic categories are readily misread as the biological races the Boasian settlement had removed from the table. Digital traces blur the line between subjective report and objective record, since a behavioural log is simultaneously a self-description and an external measurement. Technological novelty is therefore not reliably on the side of opening; it reopens and reinforces in roughly equal measure, and which it does in any given case is an empirical question rather than something that can be read off the technology.

Political opening occurs when an allocation is challenged because of the inequality or exclusion it produces. Sex/gender and race/culture are the paradigm cases, and both demonstrate that political reopening can travel in several directions at once. The same argument, that a partition assigns some people to nature and others to culture, can be made in the service of dismantling the assignment or of relocating particular groups within it.

The distinction between internal and external opening deserves particular emphasis. Internal opening occurs when practitioners themselves encounter the inadequacy in the course of their own work. External opening occurs when another field diagnoses the problem from outside. These are not equivalent, and conflating them produces predictable confusion. An externally diagnosed SWIPE may not be recognized by its alleged holders at all, which generates a dispute not merely about the solution but about whether there is a problem, and the resulting stalemate is frequently mistaken for resistance when it may simply be non-recognition.

Finally, opening admits of degrees, and the degrees should be kept distinct. A partition can be locally questioned in particular cases, while remaining intact everywhere else. It can be methodologically softened, so that practitioners are encouraged to borrow across the line without the line moving. It can be ontologically questioned, so that the reality of the two domains is explicitly disputed. It can be institutionally destabilized, so that departments merge, training changes, or funding categories are rewritten. Or it can be abandoned. Most theoretical turns reach the second or third stage. Very few reach the fourth, and the fourth is the only one that survives a change of intellectual fashion, because it alters what the next generation is trained to find obvious. Mistaking a methodological softening for an abandonment is among the more common forms of premature celebration in the human sciences, and it is easy to detect: if the vocabulary has changed while the departments, degrees, journals, and evidential standards have not, nothing has been abandoned.

VII. How SWIPEs Are Recursively Reclosed

Empirical failure frequently produces repair rather than abandonment, and the forms of repair are worth cataloguing, because each of them can look like progress.

Hard closure simply reasserts the distinction. These are fundamentally different things, and the anomaly is a curiosity.

Jurisdictional closure requires no ontological argument at all. The institution polices the boundary and hands the problem back to the proper discipline. The phenomenon is not denied; it is referred elsewhere. This is the commonest form by a wide margin, and also the least visible, because a referral does not look like a theoretical position. A grant panel declining a proposal as insufficiently within scope, a journal returning a manuscript as better suited to another venue, a clinic transferring a patient whose complaint has resisted investigation: none of these involves an assertion about the structure of reality, and all of them reproduce it.

Bridge closure is the most important and the most easily mistaken for a solution. A concept is invented to reconnect the poles: embodiment, biosociality, the psychosomatic, structure-agency mediation, natureculture, the material-semiotic, the biocultural, the biopsychosocial (Engel 1977; Rabinow 1992; Haraway 1988, 2003). These concepts frequently do excellent empirical work, and none of them should be dismissed; the biopsychosocial model reorganized medical education, and biosociality named something genuinely new about how groups form around shared biological classifications. But a bridge presupposes two shores. The pattern is: SWIPE, then felt misalignment, then a repair concept, then apparent advance, then persistence of the SWIPE beneath the repair. Every hyphen, as the companion analysis of medical anthropology puts it, commemorates the wound it was meant to heal (Ecks 2026c). The diagnostic question to ask of any bridge concept is whether it could be stated at all if the partition were abandoned. If the concept becomes unnecessary rather than merely differently expressed, it was a bridge, and its success is evidence that the shores remain.

Porosity closure changes the vocabulary rather than the structure. The boundary is described as blurred, the domains as interacting, the distinction as porous, the poles as entangled or co-produced. If the analysis still begins with two entities whose entanglement then has to be explained, the partition has survived the softening intact.

Boundary redrawing saves the partition by moving difficult phenomena across it. A trait formerly assigned to biology is reassigned to the social, or the reverse. The line moves; the existence of a line is never in question. Disputed biological and social questions are especially prone to this manoeuvre, which can be repeated indefinitely.

Protective closure is the form that most deserves a name of its own. A partition may persist not because its ontology remains convincing but because reopening it is associated with a feared regression. Race/culture is protected by the fear of biological racial determinism, a fear with the most solid historical warrant imaginable. Sex/gender is protected, depending on the position occupied, by the fear either of biological determinism or of erasing material sexual difference, which is why the same distinction is defended from opposed directions by people who agree on almost nothing else. Fact/value is protected by the fear of politicized science. Disease/illness is protected, again depending on position, by the fear of biomedical reductionism or by the fear of losing diagnostic realism.

In each case the protective work is real, historically grounded, and frequently justified. Protective sedimentation is therefore not an accusation, and it is essential that it not be read as one. It is a description of a stabilizing mechanism that operates independently of whether the partition is accurate. A distinction can be philosophically indefensible and politically indispensable at the same time, and someone who defends it under those conditions is not confused. What the framework adds is the observation that a partition sustained this way has effectively been removed from empirical adjudication, since the argument for keeping it no longer refers to the phenomena it organizes. That is worth knowing even, and perhaps especially, when the protection is warranted.

The converse danger is weaponized reopening. Someone can reopen a partition for reasons entirely unlike those of another critic. Challenging the race/culture separation can serve sophisticated biosocial research or it can serve racial essentialism, and the two are not distinguishable by the mere fact of reopening. Challenging sex/gender can serve antagonistic projects simultaneously. Recursive opening therefore has no intrinsic political virtue, and any framework that treats the dissolution of binaries as automatically progressive has failed to notice who else is dissolving them.

Which yields the comparative proposition this section is built toward. The stability of a SWIPE may bear surprisingly little relation to its philosophical coherence. A partition may instead be stabilized in proportion to the magnitude of the catastrophe that people imagine would follow from abandoning it. Fact/value has been refuted repeatedly and remains in excellent institutional health. Nature/culture has been declared obsolete for four decades by fields that continue to be organized by it. If the proposition holds, this is not a puzzle but a prediction, and it should be testable comparatively: the partitions hardest to reopen ought to be those whose collapse is imagined to release the worst consequences, regardless of how well they are defended in argument.

Two implications follow. Philosophical refutation is close to useless as a strategy on its own, since it addresses a warrant nobody is relying on. And the analysis of any protected partition is incomplete without an account of what its abandonment is feared to release, which means that the study of a SWIPE is unavoidably also a study of the fears organized around it.

VIII. Toward a General Theory of SWIPE Careers

No universal life cycle governs all cases, but a general morphology can be stated:

difference → methodological differentiation → epistemic jurisdiction → ontological allocation → institutional sedimentation → local empirical mismatch → recursive opening → conflict over the meaning of the opening → repair, reassertion, redrawing, or abandonment → resedimentation.

This is not linear. An existing L4 closure can itself generate new L2 misalignment as practice drifts from the arrangement that organizes it; those misalignments generate rival L3 articulations; and each new candidate L4 unsettles the others in turn. The nature/culture genealogy demonstrates exactly this recursive rather than stair-step structure, with Windelband's methodological reformulation unsettling Dilthey's object-based one and Weber's fusion unsettling both (Ecks 2026d). Nor is the sequence guaranteed to run to completion. Many differences stop at methodological differentiation and never acquire jurisdiction. Many jurisdictions never reify into ontology. The morphology describes a path that is available, not one that is taken, and identifying where a given case stopped is more informative than confirming that it started.

For each candidate SWIPE, a standard comparative protocol can then be applied.

1. What are the two poles?

2. When did they become independently constituted domains?

3. Which epistemologies became paired with them?

4. When did transfer across the division become problematic?

5. When did the distinction become prior to inquiry?

6. What material and institutional infrastructures carry it?

7. Who explicitly articulates it?

8. Who merely reproduces it?

9. Who experiences misalignment because of it?

10. Who diagnoses it from outside?

11. How is it recursively opened?

12. How is it reclosed?

13. What feared consequence protects it from reopening?

14. Does the candidate actually survive this test?

The last question carries the most weight. A framework that cannot return the verdict there was no SWIPE here is not doing analytical work; it is performing a ritual. Several of the candidates listed in Section I may fail, and the failures will be as informative as the confirmations, because they will show where differentiation stopped short of reification and why. Partial results should also be expected and reported as such: a case may satisfy the first three conditions and not the fourth, which would make it a strong methodological division rather than a SWIPE, and precisely that finding would explain why it has proved comparatively easy to renegotiate.

Questions six through ten are the ones most often skipped, and skipping them is what produces genealogies that read as histories of ideas held by philosophers. A partition carried by examination syllabuses and funding categories will not be found by reading treatises, and the people who could report the misalignment most precisely are rarely the people whose writings are cited when the partition is discussed.

Conclusion: Split Worlds Have to Be Made

People have always differentiated. People have always argued about oppositions. People have always developed specialized expertise, and some have always claimed authority on its basis.

A SWIPE requires considerably more. It requires a world in which knowledge practices have become sufficiently specialized, empirical, institutionalized, and professionally jurisdictional that differences among methods can be mapped onto differences among realities. That is why the nineteenth century matters, and why no single component of the story suffices on its own. The laboratory does not create a SWIPE. The archive does not. Empiricism does not. Professionalization does not. Together, and only together, they make it possible to say: these are different objects, they require different evidence, they require different experts, and the methods of one side cannot settle the questions of the other. At that point a difference has become a world.

The positionality finding then complicates the picture in a way that ought to change how such partitions are studied. A SWIPE may not reside principally in the minds of the people occupying either pole. It can reside in institutional arrangements, disciplinary jurisdictions, pedagogical simplifications, and critical descriptions, and those most troubled by it are frequently boundary workers or outside observers rather than practitioners at either centre. This makes reflexivity indispensable rather than decorative. A framework designed to detect over-reified partitions must not produce them through its own diagnosis, which is why the evidential protocol in Section V is not an afterthought but a condition of the framework's honesty.

The series this article opens will proceed accordingly. The nature/culture genealogy shows how one SWIPE emerged from distributed processes with no author (Ecks 2026d). The disease/illness analysis shows how a mature partition can be imported wholesale into a discipline and installed in the body of the sick person (Ecks 2026c). Mind/body will test whether an ancient dualism becomes a genuine SWIPE only after nineteenth-century empirical specialization. Race/culture and sex/gender will show how partitions become politically protected and therefore unusually difficult to reopen. Others will reveal different careers, and some will not survive the test.

SWIPEs are neither timeless errors of human cognition nor merely philosophical doctrines. They are historical achievements, requiring concepts, practices, infrastructures, institutions, jurisdictions, and people positioned differently in relation to the resulting divisions. They can be carried without being believed, defended without being thought true in every respect, attacked by people who do not inhabit them, and reconstructed by critics who may inadvertently reproduce the very partitions they set out to expose. To study a SWIPE is therefore not simply to ask where a line was drawn. It is to follow the recursive career through which a difference became two worlds, through which those worlds acquired different ways of knowing, and through which the resulting arrangement came to seem difficult, dangerous, or simply unnecessary to reopen.

A world is not split because two things differ. A SWIPE begins when the difference determines who may know what, by which means, and before the inquiry has even begun.

References

Abbott, A., 1988. The System of Professions: An Essay on the Division of Expert Labor. Chicago: University of Chicago Press.

Balibar, E., 1991. Is there a "neo-racism"? In E. Balibar and I. Wallerstein, Race, Nation, Class: Ambiguous Identities, trans. C. Turner. London: Verso, pp.17–28.

Boas, F., 1911. The Mind of Primitive Man. New York: Macmillan. Rev. ed. 1938.

Bourdieu, P., 1975. The specificity of the scientific field and the social conditions of the progress of reason. Social Science Information, 14(6), pp.19–47.

Butler, J., 1990. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge.

Clark, W., 2006. Academic Charisma and the Origins of the Research University. Chicago: University of Chicago Press.

Daston, L. and Galison, P., 2007. Objectivity. New York: Zone Books.

Descola, P., 2013 [2005]. Beyond Nature and Culture. Trans. J. Lloyd. Chicago: University of Chicago Press.

Durkheim, E., 1995 [1912]. The Elementary Forms of Religious Life. Trans. K.E. Fields. New York: Free Press.

Ecks, S., 2026a. The impossible object: beyond social theory. Living Value Theory, livingvaluetheory.org.

Ecks, S., 2026b. Lead the pig-lamb-calf around the field: a deep history of the word "culture." Living Value Theory, livingvaluetheory.org.

Ecks, S., 2026c. Medical anthropology theory: where it all went wrong and how to move on. Living Value Theory, livingvaluetheory.org.

Ecks, S., 2026d. The long incubation of a SWIPE: how nature and culture became incommensurable ways of knowing. Living Value Theory, livingvaluetheory.org.

Engel, G.L., 1977. The need for a new medical model: a challenge for biomedicine. Science, 196(4286), pp.129–136.

Fausto-Sterling, A., 2000. Sexing the Body: Gender Politics and the Construction of Sexuality. New York: Basic Books.

Fleck, L., 1979 [1935]. Genesis and Development of a Scientific Fact. Eds. T.J. Trenn and R.K. Merton, trans. F. Bradley and T.J. Trenn. Chicago: University of Chicago Press.

Germon, J., 2009. Gender: A Genealogy of an Idea. New York: Palgrave Macmillan.

Gieryn, T.F., 1983. Boundary-work and the demarcation of science from non-science: strains and interests in professional ideologies of scientists. American Sociological Review, 48(6), pp.781–795.

Haraway, D., 1988. Situated knowledges: the science question in feminism and the privilege of partial perspective. Feminist Studies, 14(3), pp.575–599.

Haraway, D., 2003. The Companion Species Manifesto: Dogs, People, and Significant Otherness. Chicago: Prickly Paradigm Press.

Holmes, F.L., 1989. The complementarity of teaching and research in Liebig's laboratory. Osiris, 5, pp.121–164.

Kleinman, A., 1980. Patients and Healers in the Context of Culture: An Exploration of the Borderland Between Anthropology, Medicine, and Psychiatry. Berkeley: University of California Press.

Kuhn, T.S., 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

Latour, B., 1993 [1991]. We Have Never Been Modern. Trans. C. Porter. Cambridge, MA: Harvard University Press.

Lock, M., 1993. Encounters with Aging: Mythologies of Menopause in Japan and North America. Berkeley: University of California Press.

Money, J., Hampson, J.G. and Hampson, J.L., 1955. An examination of some basic sexual concepts: the evidence of human hermaphroditism. Bulletin of the Johns Hopkins Hospital, 97(4), pp.301–319.

Morrell, J.B., 1972. The chemist breeders: the research schools of Liebig and Thomas Thomson. Ambix, 19(1), pp.1–46.

Oakley, A., 1972. Sex, Gender and Society. London: Temple Smith.

Olesko, K.M., 1991. Physics as a Calling: Discipline and Practice in the Königsberg Seminar for Physics. Ithaca: Cornell University Press.

Putnam, H., 2002. The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge, MA: Harvard University Press.

Rabinow, P., 1992. Artificiality and enlightenment: from sociobiology to biosociality. In J. Crary and S. Kwinter, eds., Incorporations. New York: Zone Books, pp.234–252.

Rheinberger, H.-J., 1997. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford: Stanford University Press.

Rocke, A.J., 2023. Origins and spread of the "Giessen model" in university science. In K. van Berkel and E. Homburg, eds., The Laboratory Revolution and the Creation of the Modern University, 1830–1940. Amsterdam: Amsterdam University Press, pp.33–64.

Rubin, G., 1975. The traffic in women: notes on the "political economy" of sex. In R.R. Reiter, ed., Toward an Anthropology of Women. New York: Monthly Review Press, pp.157–210.

Shapin, S. and Schaffer, S., 1985. Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life. Princeton: Princeton University Press.

Stocking, G.W., Jr., 1968. Race, Culture, and Evolution: Essays in the History of Anthropology. New York: Free Press.

Stolcke, V., 1995. Talking culture: new boundaries, new rhetorics of exclusion in Europe. Current Anthropology, 36(1), pp.1–24.

Stoller, R.J., 1968. Sex and Gender: On the Development of Masculinity and Femininity. New York: Science House.

Strathern, M., 1980. No nature, no culture: the Hagen case. In C.P. MacCormack and M. Strathern, eds., Nature, Culture and Gender. Cambridge: Cambridge University Press, pp.174–222.

Turner, R.S., 1971. The growth of professorial research in Prussia, 1818 to 1848: causes and context. Historical Studies in the Physical Sciences, 3, pp.137–182.

Visweswaran, K., 1998. Race and the culture of anthropology. American Anthropologist, 100(1), pp.70–83.

Weber, M., 1946 [1917]. Science as a vocation. Lecture delivered at Munich, November 1917; first published 1919. In H.H. Gerth and C.W. Mills, eds. and trans., From Max Weber: Essays in Sociology. New York: Oxford University Press, pp.129–156.

Whitehead, A.N., 1920. The Concept of Nature. Cambridge: Cambridge University Press.

Williams, R., 1983 [1976]. Keywords: A Vocabulary of Culture and Society. Rev. ed. Oxford: Oxford University Press.