Abstract
Medication has a material history that begins long before pharmacies. This article defines a medicine as matter recursively recruited to alter an embodied process that is felt, discerned or symbolically classified as requiring remediation, and reconstructs the history of such recruitment as an accretive stratigraphy of multimaterial relations. Animal self-medication shows the relation at its most basic, with matter recruited in response to felt disturbance and no concept of medicine. From there the stratigraphy runs through found, collected, transformed, compounded, purified and projectively fabricated medicinal matter, each layer added to the earlier ones without replacing them. Modern pharmaceutical research recapitulates these layers within single programmes, as the history of artemisinin shows, so that a contemporary drug carries several of them at once and the opposition between natural remedy and manufactured pharmaceutical dissolves. The distinctive problem of medication is mobility. A medicine must stay the same in the respects on which remediation depends while it crosses climates, supply chains and bodies, a condition called portable sameness and analysed as selective invariance, stabilised by manufacturing standards, packaging, labelling, regulation and law. Because mobility and reproducibility push ownership toward symbolic specification, disputes over patents, generics and traditional knowledge appear as contests between symbolic stabilisations of a relation that is already multimaterial and often very old, in which the burden of reopening a closure falls on the disadvantaged party. Throughout, medication remains multimediated and answerable downward: no dossier or property right can make a molecule work in a body, a climate or a community that does not fit it, and remediation for one body can alter the conditions of remediation for others.
I. Introduction: What Is a Medication?
When does something become a medicine? The question sounds elementary, and it becomes less so the longer one considers it. A chimpanzee in the Mahale Mountains, listless and heavily parasitised, strips the bark and leaves from a shoot of Vernonia amygdalina and chews the bitter pith of a plant her group ordinarily avoids. An orangutan in Sumatra chews the leaves of a liana and presses the paste onto a wound on his face. A household in West Bengal applies turmeric to a cut. Someone drinks a decoction for a fever. A pharmacist dispenses a standardised tablet whose active ingredient was synthesised in one country, formulated in another and packed in a third. A medicinal chemist designs, on a screen, a molecule that has never existed. All of these are medicinal events, and their material histories could hardly differ more.
This article proposes a definition broad enough to hold them together: a medicine is matter recursively recruited to alter an embodied process that is felt, discerned or symbolically classified as requiring remediation. Remediation here means the repair or restoration of a disturbed coordination. The definition deliberately makes no reference to professional authorisation, chemical purity, synthetic manufacture or even explicit intention. Each of these marks a later stabilisation of medicinality, and none can serve as its criterion without cutting off most of its history.
The three verbs of the definition name three ways in which a bodily process can become a candidate for remediation, and they correspond to different levels of recursivity. A disturbance can be felt by the one whose body it is, in the change of texture that makes something salient before it can be named, which is L2. It can be discerned by another being attuned to the sufferer, as when a mother registers that her child is unwell before the child can say so. And it can be symbolically classified, named as a complaint, a symptom or a diagnosis, which happens at L3 and is stabilised into portable categories of indication and treatment at L4. A medicine can be recruited at any of these points. The vocabulary of consciousness cannot draw the distinctions among them. Used broadly, the word covers all of them at once, and used narrowly it draws a single line through their middle. Neither use can say what the history of medication needs to say, which is how far above felt disturbance a given recruitment has climbed. Remediation also has a direction. Its aim is that the disturbed process settles again into coordination that needs no attention, and a successful remediation therefore tends to disappear from view.
The first threshold the definition implies is condition-specific remediation. Eating becomes medicinal when a substance is selectively recruited in relation to some perceived alteration of the body, whatever pharmacological effects the substance may have in other contexts. The distinction between diet and medicine is accordingly relational before it is categorical. Research on animal self-medication supplies the deepest evolutionary entry point, because it has had to specify this relation empirically: a medicinal use must be associated with a particular condition, differ from ordinary diet, be followed by improvement and involve material with relevant biological activity (Huffman 1997). The word recruited carries the criterion. A plant's antiparasitic activity is a fact about the plant, and it becomes medicinal only when a living being takes it up into its further coordination in relation to a bodily condition.
The account of made realities on which the article draws treats multimateriality as one of five mediations through which a living being remains coordinated with a world, alongside multisensorial embodiment, being-with, multiversal dwelling and multisymbolisation. Embodiment, being-with and dwelling are the zoetic mediations, available to living beings that make nothing. Multimateriality and multisymbolisation are the poietic mediations, in which recursive coordination is deposited into relatively durable material and symbolic form. Multimateriality has a layered evolutionary history. It rests on two preconditions, material dependence on conditions of dwelling and metabolic externalisation, and unfolds through six layers of relation to matter: recruited self-produced matter, recruited found matter, selected and displaced matter, transformed matter, combined matter and projectively fabricated matter. The criterion throughout is recursive uptake, since a material difference becomes multimaterial only when it enters a living being's further coordination. The layers accumulate rather than succeed one another, and found and made name relations to aspects of a material configuration, not classes of object.
The article develops four arguments. The first is that medication can be reconstructed through this stratigraphy, in which later forms of medicinal matter are added to earlier ones without replacing them. The second is that modern pharmaceutical production does not supersede earlier medicinal materialities. Found plants, dried herbs, extracts, compound preparations, purified alkaloids and synthetic molecules remain layered together in contemporary therapeutics, and a single drug frequently carries several of these relations inside it. The third is that medicines acquire an especially intense form of poietic stabilisation because they are mobile remediations. They must preserve enough material and symbolic identity to travel between places, institutions and bodies and still produce the anticipated effect. The fourth is that medication remains fully multimediated and answerable downward, since the most elaborately specified remedy still has to fit a body, a climate and a community.
The question organising the article follows from these arguments. How did living beings move from opportunistically recruiting substances in response to bodily trouble to manufacturing globally mobile substances whose composition, indication, dose, ownership and legitimate circulation can all be specified before the medicine reaches a patient?
II. Before Medicine: From Diet to Condition-Specific Remediation
The history of medication begins below any formal category of medicine. Every heterotrophic organism must recruit external matter into its metabolism, and much of what later becomes materia medica begins within this vastly older nutritional relationship. The evolutionary problem is therefore differential recruitment rather than first ingestion: how certain materials come to be taken up at particular times and for particular conditions, in ways that differ from their uptake as food.
A continuum can be drawn across this problem. At one end lies ordinary consumption; then unusual consumption; then consumption under bodily disturbance; then repeated recruitment associated with that disturbance; and finally socially or symbolically stabilised medicinal use. Animal self-medication shows how far along this continuum nonhuman life has travelled. Huffman and Seifu (1989) described a visibly ill chimpanzee at Mahale chewing the pith of Vernonia amygdalina, a plant chimpanzees ordinarily avoid and whose bitterness signals compounds with antiparasitic activity, and resuming normal activity by the following day. Chimpanzees and bonobos in several populations also swallow rough, hairy leaves whole, without chewing, in the rainy season when nodular worm infections peak, and the leaves pass through the gut carrying worms with them (Huffman 2001). Work at Budongo has since combined behavioural observation with pharmacological testing of plants eaten outside ordinary diet, finding that most of the extracts tested inhibited bacterial growth and about a third had anti-inflammatory activity (Freymann et al. 2024). The orangutan case adds external application. Rakus, a male at Suaq Balimbing, chewed the leaves of Fibraurea tinctoria, a liana orangutans rarely eat, applied the juice repeatedly to a fresh facial wound and then covered it with the chewed leaves; the wound healed without visible infection (Laumer et al. 2024).
Much of the matter recruited in these cases was produced by other organisms for purposes of their own. Many of the bitter and bioactive compounds of plants are secondary metabolites that plants make in relation to herbivores, pathogens and competitors, and many antibiotics are secretions of soil microbes that evolved in competition with other microbes (Davies and Davies 2010). What is a secretion for the organism that makes it becomes found medicinal matter for the ape or the human who recruits it. The same molecule can therefore stand in different multimaterial relations to different beings: self-produced for its maker, found for the chimpanzee that chews it and, later, selected, transformed and designed for the chemists who work on it.
These cases call for a distinction between remediation and medicine. Remediation is the more basic relation, and an animal can recruit matter remedially without possessing any symbolic category corresponding to medicine. The remediative relation can therefore precede the explicit category of medicine by millions of years. In Living Value Theory's terms, the relation begins at L2. A bodily discrepancy becomes salient without diagnosis: something feels, moves or functions differently, and a material engagement is recruited in response. The claim requires no conceptual self-diagnosis on the animal's part, only recursive susceptibility, the fact that prior bodily consequences alter subsequent material engagement, so that a felt misalignment and the matter that eases it become coupled in the organism's further coordination. Such recruitment can be sustained by what Godfrey-Smith (2024), following Wright (1976), describes as feedback from past cases: what has eased a trouble before is sought again, through individual trial and error or, in some populations, through social learning (Huffman 1997). Recruitment guided by a represented outcome that nothing has yet achieved belongs to a much later layer of the stratigraphy.
Human medicine did not begin by drawing a clean boundary between food and medicine either. De Vos's (2010) study of twelve Mediterranean and European pharmacy texts from the fifth century BCE to the nineteenth century identified 439 commonly used simples, 78 per cent of them from plants and the remainder from animals and minerals, and many of them served purposes beyond therapeutics. Turmeric in South Asia is at once spice, colourant, cosmetic and medicine, and the same root may pass through all four uses in a single household on a single day. Where pharmaceuticals enter everyday life, the boundary remains porous in the other direction as well. In India, psychopharmaceuticals are widely understood through an idiom of digestion, so that a pill is assessed as something eaten and worked upon by the belly (Ecks 2013).
The first proposition of the article can now be stated in its simplest form. Medicine begins when ordinary matter acquires a condition-specific remediative relation to embodiment, long before any special class of medicinal substances appears.
III. A Multimaterial Stratigraphy of Medication
The relations through which matter becomes medicinal can be ordered as a stratigraphy that specialises the general layers of multimateriality to the case of remediation. The ordering is partial rather than linear. Some relations build on others, and later relations incorporate earlier ones without superseding them, but the stratigraphy makes no claim that every medical tradition passed through them in the same sequence. What it describes is the expanding range of relations through which living beings have recruited matter to remediate embodiment, and the way these relations stack within a single therapeutic practice or a single drug.
Beneath the six layers lie the two preconditions of all multimateriality. Every remediation presupposes a body that depends on material conditions it did not make and whose metabolism takes matter in, transforms it and releases it again. A medicine enters the body through this metabolism and leaves through it, and what leaves returns to the conditions of dwelling, a point on which Section VI insists.
The first layer is recruited self-produced remediation. Many mammals lick their wounds, and human saliva contains histatins, peptides that promote the closure of epithelial wounds in cell culture (Oudhoff et al. 2008). Recruited self-produced remediation is real but narrow, and medication as a history of made remediation begins largely with matter from elsewhere.
The second layer is recruited found medicinal matter. Plants, bark, leaves, clay, minerals and resins are encountered and recruited with minimal transformation. Most animal self-medication belongs here: the chimpanzee chews the pith where the plant grows. Geophagy, the eating of clays and earths, which occurs across many primates and many human populations and may bind toxins or protect the gut, belongs here too. The criterion concerns the material form rather than its naturalness: the relevant form substantially precedes its medicinal recruitment.
The third layer is selected and displaced medicinal matter. Medicinal matter is sought, collected, stored or carried for later remediation. The change is evolutionarily enormous, because prospective medicinal relevance can now survive separation from the immediate episode of illness. The gathered herb drying in a shelter differs from the plant chewed where it grows: it has become a portable remedy, held in readiness for a trouble that has not yet arrived. Displacement is also the first form of medicinal mobility, and the problem this article pursues begins here. A dried leaf carried from a hillside to a village must still be the same leaf when someone needs it, and storage already raises the questions of decay, contamination and substitution that pharmaceutical regulation will later formalise. Long-distance trade magnified the problem. When cinchona bark reached Europe from the Andes in the seventeenth century, its value against intermittent fevers made it worth adulterating, and apothecaries and physicians had to learn to tell true bark from inferior or substituted barks by colour, taste and texture, long before its active principle could be named. Displaced medicinal matter needed marks of identity that could travel with it.
The fourth layer is transformative medication. Grinding, pounding, drying, soaking, boiling, smoking, fermenting, roasting, extracting and dissolving alter the substance before use. The orangutan's chewing is a minimal transformation, turning a leaf into juice and paste. Human traditions elaborated transformation into a large part of what healing knowledge consisted of. Dioscorides' De materia medica, compiled in the first century CE and consulted for more than sixteen centuries, recorded for each substance where to gather it, how to prepare and preserve it, what it treated and in what quantity (Riddle 1985; De Vos 2010). Processing in such traditions increasingly becomes part of what the medicine is, rather than a preparation applied to a medicine that already exists. A decoction and a tincture of the same root are different medicines, and the difference lies wholly in the transformation.
The fifth layer is combinatorial medication. Several independently recruited materials are deliberately assembled into decoctions, plasters, ointments, pills and compound preparations. The important transition runs from the claim that this substance helps this trouble to a stable formulation whose effect is attributed to a relation among ingredients. Historical materia medica made the distinction symbolically explicit by separating simples from compounds, and some compounds became institutions in their own right. Theriac, the celebrated antidote of Greek, Roman, Islamic and European pharmacy, combined dozens of ingredients, among them opium and viper flesh, and its public preparation in Venice was supervised by officials into the early modern period (Griffin 2004). Ayurvedic, Unani and Chinese pharmacologies developed equally elaborate doctrines of formulation, in which the relations among ingredients carried as much therapeutic weight as the ingredients themselves.
Purification and standardisation form a late intensification of the fourth layer. They appear only where extraction and analysis reach a certain degree of control, and historically they follow the rise of compounds. When Friedrich Sertürner isolated morphine from opium in the early nineteenth century, and Pelletier and Caventou isolated quinine from cinchona bark in 1820, the pharmacologically relevant relation was separated from most of the source material. The willow, the bark and the poppy need no longer accompany the compound. Concentration can be controlled and dose specified by weight, and the identity of the medicine becomes steadily less dependent on the visible organism it came from. Purification is a form of material abstraction: it strips away the matter that does not answer to the remediative relation, and it makes what remains far more portable, storable and measurable than the plant.
The sixth layer is projectively fabricated medication. Here the recursive coupling of multimateriality and multisymbolisation becomes decisive. Chemical structure, dose, release profile, formulation, manufacturing method and intended indication can be specified before a batch exists, and medicinal molecules can themselves be designed, modified and screened symbolically before they are synthesised. A medicine can exist first as a structure, a formula, a target profile, a synthetic route, a trial protocol and a regulatory dossier, and matter must then be made to answer to these specifications. Aspirin marks an early threshold. Salicin had been extracted from willow bark in 1828; chemists at Bayer acetylated salicylic acid in 1897 to obtain a derivative less irritating to the stomach than the salicylates then in use, and the compound was marketed in 1899 under a trade name (Sneader 2000). The molecule was designed against a specified bodily consequence before it was made.
Projection adds a second route by which something can come about because of its effects. Godfrey-Smith (2024), drawing on Wright (1976), distinguishes feedback from past cases, in which earlier successes explain why a form exists, from the representation and pursuit of a future outcome that nothing has yet produced. The first five layers of medicinal recruitment can be sustained by the first route alone, through trial and error, social learning and the accumulated practice of healing traditions. The projective layer adds the second. A molecule can be designed for a target that no earlier remedy has reached. In practice the two routes are interleaved, since designed and screened candidates must be tested in living systems, and most of them fail before they reach patients (Sun et al. 2022).
Non-supersession runs through the entire stratigraphy. Contemporary therapeutics still contains plants used with little modification, household herbal preparations, standardised extracts, purified natural products, semisynthetic and fully synthetic compounds, vaccines, biologics and highly engineered delivery systems. A single pharmacy shelf may hold several layers at once, and a single patient may use all of them in a week. A single tablet is found in some respects and made in others: the lineage of its active principle may run back to a willow or a poppy, its crystalline form may be engineered, and its excipients and packaging belong to the fifth and sixth layers. The pharmaceutical world is a stack of relations to medicinal matter, in which the newest have been added to the oldest.
IV. The Pharmaceutical Laboratory as Stratigraphic Recapitulation
Modern pharmaceutical research often passes, within a few years, through the whole sequence of relations that the stratigraphy distinguishes. A typical path from source to product runs from finding an organism or a therapeutic practice, through selecting a candidate, collecting material, extracting, fractionating and identifying an active constituent, to modifying the molecule, synthesising analogues, combining the chosen compound with excipients, formulating, stabilising and manufacturing. Natural products research moves routinely among these steps. Newman and Cragg (2020) found that natural products, their derivatives and synthetic compounds modelled on them accounted for a large share of the small-molecule drugs approved worldwide between 1981 and 2019.
The pattern, which implies no grand claim that pharmacology replays evolution, is better called stratigraphic recapitulation: industrial research repeatedly passes through the same kinds of material relation that the general stratigraphy distinguishes, because each kind solves a problem that the next presupposes.
Artemisinin shows the recapitulation with unusual completeness. In 1969 Tu Youyou joined a Chinese research programme seeking new antimalarials, and her team screened traditional remedies, among them qinghao, sweet wormwood (Artemisia annua). Early extracts gave inconsistent results. Tu returned to a fourth-century text, Ge Hong's Handbook of Prescriptions for Emergencies, which instructed that a handful of qinghao be soaked in water and its juice wrung out and drunk, and inferred from the absence of heating that high temperatures might destroy the active principle. Her team switched to low-temperature ether extraction, obtained a fully effective extract in 1971 and isolated the active compound, artemisinin, in 1972 (Tu 2011). Each subsequent step added a further layer. Chemists modified the molecule into more soluble and potent semisynthetic derivatives such as artesunate and artemether. Clinicians combined these with partner drugs into artemisinin-based combination therapies, now the standard first-line treatment for uncomplicated falciparum malaria, partly to slow the evolution of resistance. Synthetic biologists engineered yeast to produce artemisinic acid, a precursor that could then be converted chemically into artemisinin, in an attempt to free supply from the fluctuations of agricultural harvest (Paddon et al. 2013). The attempt met market resistance. Sanofi began industrial production in 2013, made no semisynthetic artemisinin in 2015 and put its plant up for sale, in a market whose economics proved hard to change (Peplow 2016). The cultivated plant, its ancient preparation, the purified compound, the semisynthetic derivatives and the fixed combination all remain in use, and the fate of the engineered yeast shows that the newest layer does not automatically displace the oldest.
A contemporary medicine therefore frequently carries residues of every layer at once. Its compound may descend from a found plant and a practice of use. Its active principle may have been isolated. Its structure may have been transformed. Its dosage form is combinatorial, since tablets consist largely of excipients that bind, disintegrate, coat, preserve and colour. Its production is projectively specified down to the particle size of its powders. This dissolves the familiar binary between natural medicine and manufactured pharmaceutical. The more productive questions concern how many layers of transformation, stabilisation and symbolic projection intervene between source matter and medicinal use, and which of them a given actor can see.
Artemisinin also shows how the recapitulation redistributes knowledge across the layers. Tu's decisive step depended on reading a symbolically stabilised trace of an older transformative practice, a recipe preserved for some sixteen centuries, as information about the material conditions under which a compound survives. The ancient text became relevant to the laboratory as a record of what the plant's chemistry tolerated, something its authors could not have described in those terms. Godfrey-Smith (2024), following Henrich (2016), stresses that traditions often carry effective techniques whose rationale their practitioners cannot state, and that observers cannot reliably sort the useful elements from the arbitrary ones by inspection. The qinghao recipe shows both edges of this. Its instruction to soak and wring rather than boil looks like an incidental detail until an extraction experiment shows what it protects. Older layers of the stratigraphy thus remain available to newer ones both as material relations and as the symbolic deposits that recorded them.
The projective layer's characteristic promise is that a medicine can be designed for an outcome before it exists, and the laboratory keeps returning that promise to living bodies for confirmation. Candidate molecules must be tested for toxicity and effect in living systems, and for a century this has meant passing them through the bodies of other animals, rodents above all, and dogs and primates in some regulatory toxicology. Godfrey-Smith (2024) argues that if felt experience is widespread among animals, the whole lives of laboratory animals, and not only the procedures performed on them, belong in the reckoning, and he would end harmful experiments on primates and dogs even at the cost of slowing some medical advances. Whatever view one takes of that argument, it shows how a medicine specified in symbols remains answerable to the embodiment of beings other than the patient, and that its answerability runs through other lineages before it reaches its own. Some regulators have begun to admit nonanimal methods, as United States law has done since 2022 (United States Congress 2022), and computational and cell-based models now stand in for some tests. Such stand-ins are validated by how well they predict outcomes in living bodies, so that the substitution moves the point of answerability without removing it.
V. Mobile Remediation: Why Medicines Need Exceptional Stabilisation
A Roman bridge, the paradigm of projective multimateriality, remains tied to one gorge, one geology and one river for the whole of its existence. Its fit is achieved in place, and the principle that governs it is that later layers remain answerable to the older material and zoetic relations they transform: the bridge reorganises the relation between two banks without suspending the river. Medication poses a different poietic problem. A tablet manufactured in Hyderabad, Basel or New Jersey may be expected to produce sufficiently predictable effects months later in Edinburgh, Accra or São Paulo, in bodies its makers will never meet, after passing through ships, warehouses, trucks, pharmacies and bathroom cabinets. It must become portable without becoming indeterminate. The bridge has to fit where it is; the medicine has to remain capable of fitting after it has travelled.
The problem can be named portable sameness. A pharmaceutical must preserve enough material and symbolic identity for widely separated actors to treat distributed instances as instances of the same medicine. Portable sameness is always selective. No two tablets are identical at the molecular level, and much about them may vary, including their colour, their coating, their country of manufacture and their price. What must remain invariant is a specified set of properties within specified limits: the identity of the active substance, its concentration and purity, the dose per unit, the dosage form, the rate at which the substance dissolves and becomes available to the body, and the absence of harmful contaminants. Portable sameness is therefore selective invariance, the stabilisation of exactly those relations on which remediation depends while everything around them changes.
What must stay invariant includes more than the effect for which a medicine was recruited. Recruitment is selective on an effect, but the matter arrives with all of its properties. In Godfrey-Smith's (2024) terms, an activity can bring about changes that are its point and changes that merely follow from it, and the same holds for medication: the sedation that accompanies an antihistamine, the gastric irritation that accompanies aspirin and the tolerance that follows repeated opioid use are consequences of a recruited remediation that nobody recruited. Portable sameness has to hold these unrecruited effects as steady as the recruited ones. Pharmacovigilance, the monitoring of medicines after they have entered circulation, exists because some of them appear only when a medicine has passed through more bodies than any trial could include.
The band of invariance also differs between kinds of medicine. For drugs with a narrow therapeutic index, such as lithium, warfarin or levothyroxine, small differences in the amount reaching the blood can separate effect from toxicity, and prescribers treat switching between versions with particular caution. Biological medicines, large proteins produced in living cells, cannot be copied molecule for molecule at all, since their structure depends in part on the cell lines and processes that make them. Regulators accordingly approve their copies as biosimilars, shown to have no clinically meaningful differences from the reference product, rather than as identical. The biosimilar is an institutional acknowledgement that portable sameness is a stabilised threshold rather than a natural fact, and that the threshold has to be set according to what living bodies can tell apart.
Achieving it requires several stabilisations at once. Good manufacturing practice seeks to ensure that products are consistently produced and controlled according to quality standards appropriate to their intended use (WHO 2014). Stability testing establishes how long a product keeps its specified properties at given temperatures and humidities, which is why the world is divided into climatic zones for the purpose. Good storage and distribution practices extend quality assurance through the whole chain of transport and storage, because heat, humidity, light and handling can alter the medicinal object after it leaves the factory (WHO 2020). Generic medicines must show bioequivalence, delivering the active substance into the bloodstream at a rate and to an extent within agreed limits of the reference product, which makes the body itself the measure of sameness. Vaccine vial monitors, heat-sensitive labels that darken irreversibly with cumulative heat exposure, give vaccines a material record of their own journey, so that a health worker at the end of a cold chain can see whether the vial is still the vaccine it was.
Mobility intensifies symbolisation rather than freeing matter from it. A white powder by itself cannot tell distant actors reliably what it is. Names, strengths, batch numbers, expiry dates, storage instructions and package inserts become part of what makes mobility possible, and the medicine travels as a coupled material and symbolic object whose two components must stay aligned. When they come apart, the consequences can be lethal. In 1937 a liquid formulation of sulfanilamide was marketed in the United States using diethylene glycol as its solvent; the drug was correctly named and its active ingredient was real, but the untested solvent killed more than a hundred people, and the disaster led directly to the 1938 federal requirement that new drugs be shown to be safe before marketing. Diethylene glycol contamination of medicines has recurred repeatedly since, with children its principal victims, wherever a cheaper solvent has been substituted for a pharmaceutical-grade one somewhere along the supply chain. Successful portable sameness, by contrast, leaves little trace. The patient swallows a tablet that is what its label says, and nothing about the encounter becomes salient beyond the relief it brings. Misfit is articulated, investigated and legislated, and the history of pharmaceutical regulation has been written largely through its disasters, from the sulfanilamide elixir to thalidomide, whose effects on fetal development led in 1962 to the American requirement that new drugs be shown to be effective as well as safe.
Anthropologists of pharmaceuticals have long followed medicines as travelling objects. Van der Geest, Whyte and Hardon (1996) proposed a biographical approach that traces medicines through production, marketing, prescription, distribution and consumption, and Hardon and Sanabria (2017) argue that pharmaceuticals are fluid, their effects and meanings shifting with the settings through which they pass. Latour (1986) located the power of scientific inscriptions in their capacity to be immutable and mobile at once, while de Laet and Mol (2000) showed that a technology can travel well precisely because it is fluid and adapts to each setting. Medication sits between these poles. Its uses, meanings and effects are fluid in exactly the ways the anthropological literature documents, while its specified material identity must be held as close to immutable as manufacturing and regulation can make it. Portable sameness names the narrow band of invariance on which the fluidity of everything else depends.
A general proposition follows. The greater the spatial mobility of a multimaterial remediation, the greater the burden of stabilising what must remain invariant while everything around it changes.
VI. Medication Remains Fully Multimediated
A medication is multimaterial, but medicinal remediation remains fully multimediated. The five mediations can be followed through any therapeutic event, and none of them is optional.
Multisensorial embodiment is where every medicine finally acts. Absorption, distribution, metabolism and excretion, therapeutic effect and toxicity, nausea, drowsiness, relief from pain and the side effects that lead people to stop taking a drug all occur in living bodies, and they vary with age, weight, sex, genotype, pregnancy, comorbidity and the other substances the body is already handling. Pharmacogenomics has made part of this variability explicit, as with the enzyme variants that cause codeine to be converted into morphine too slowly in some people and dangerously fast in others. Embodiment is also where a medicine is felt: taste, texture, the size of a tablet and the sting of an injection shape whether it will be taken again.
Being-with is where medicines circulate. Someone notices that another person is unwell, recommends, prescribes, dispenses, administers, encourages, warns, monitors or refuses. Trust in doctors, pharmacists, healers, relatives and manufacturers can determine use as decisively as a tablet's chemistry. The literature on placebo and meaning responses shows how far this reaches into efficacy itself, since the colour of a tablet, its number and the manner of the person who gives it can alter measured outcomes (de Craen et al. 1996; Moerman 2002). Household routines of reminding, supervising and sharing medicines belong to being-with as well, coordinating living beings around a medicine's schedule.
Multiversal dwelling comprises the non-human-made conditions within which medicines must remain stable and bodies must be reached. Heat, humidity and light degrade active substances and excipients. Seasons shape when fevers and diarrhoeal diseases arrive and how passable the roads are. Altitude, distance and terrain decide how long a journey takes, and the ecologies of mosquitoes and parasites decide where antimalarials are needed and where resistance evolves. The climatic zones of stability testing are an explicit acknowledgement that a medicine's portable sameness has to be achieved against dwelling conditions no manufacturer controls.
Multimateriality comprises everything made that carries the remediation: active ingredients, excipients, capsule shells, coatings, bottles, blister packs, syringes, inhalers, refrigerators, cold boxes, warehouses, pharmacies, clinics, roads and factories. The blister pack is a small masterpiece of the fifth and sixth layers combined, sealing each dose individually against moisture, marking days or counts, and turning a course of treatment into a visible, depletable material sequence.
Multisymbolisation stabilises what the material is supposed to be and do: the drug's name, the diagnosis and indication, dose, contraindications, prescription, package insert, trial results, clinical guidelines, patent, marketing authorisation and expiry date. These symbols operate at every recursivity level, from the half-read label glanced at in the morning to the institutional stabilisations of pharmacopoeias and regulatory agencies at L4.
Pharmaceutical remediation takes place inside a multimediated mesocosm, and a pharmaceutical acts upon embodiment only through the other mediations with which it is coupled. A chemically effective medicine can fail because a person does not trust the prescriber. A correctly prescribed medicine can fail because it spent a week at forty degrees. A perfectly manufactured drug can fail because the diagnosis was wrong. A useful medicine can remain out of reach because price, law or distance blocks its circulation, and a correctly named medicine can harm through a dispensing error. Regulatory frameworks reflect this in their own idiom, extending oversight from the molecule to production, packaging, labelling, storage, distribution, prescription and use.
The mediations also carry consequences between bodies. Antimicrobial resistance shows remediation in one body altering the conditions of remediation for others: each course of antibiotics selects among the bacteria it meets, and the resistant populations it leaves behind circulate through households, hospitals, farms and wider microbial ecologies. Resistance is older than the clinic. Genes conferring resistance to antibiotics have been recovered from bacterial DNA some thirty thousand years old (D'Costa et al. 2011), because the antibiotics that medicine recruited were already weapons in a long microbial competition. Human remediation enters a contest that predates it and alters its terms. A remediation that succeeds for one patient can thus erode the effectiveness of the same medicine for everyone else, which makes the medicinal object answerable to being-with and dwelling far beyond the single therapeutic encounter.
Remediation can also reach beings that nobody set out to treat. Diclofenac, an anti-inflammatory long used in human medicine, came into wide veterinary use in South Asia in the 1990s. Vultures that fed on the carcasses of treated livestock died of kidney failure, and the populations of the resident Gyps vultures collapsed (Oaks et al. 2004). India banned the veterinary use of the drug in 2006. The vultures had no part in the therapeutic relation between herder and cow, yet the remediation of one body ran through the metabolic externalisation of its remains into the bodies of scavengers, and from there into the way a region disposes of its dead. A medicine's consequences do not stop at the bodies it was recruited to alter.
Success has a characteristic form in this setting. A remediation succeeds when the disturbed process settles back into coordination that needs no attention, so that the work of the medicine disappears from view. The consequence is visible in the long-term treatment of conditions that produce no felt disturbance. A person taking a drug to lower blood pressure feels nothing when it works, while the nausea, the cost and the daily inconvenience of taking it are all salient, and the felt balance of the treatment runs against it. The World Health Organisation's review of adherence estimated that in developed countries about half of the patients with chronic illness take their medicines as prescribed (Sabaté 2003), and among the many reasons for this, the invisibility of success is a familiar one. Coordination that has been remediated successfully has been muted, and what has been muted gives its beneficiary little reason to keep supporting it.
A definition of medicinal success follows. Medicinal success is the fit of a material remediation within the multimediated situation in which a living process is being altered, and pharmacological efficacy is one necessary component of that fit among several.
VII. Law, Property and the Problem of the Mobile Copy
The Shard in London is a strongly projective multimaterial object, specified in drawings, models and calculations before a single pile was driven. Its design may be protected: United Kingdom law treats a building as an artistic work of architecture, and copyright arises automatically without registration (Copyright, Designs and Patents Act 1988, s. 4). The decisive difference from a medicine is physical. Nobody can carry the Shard away, substitute another Shard for it, dilute it, counterfeit a batch of it, or reproduce millions of materially interchangeable Shards for sale across borders. A building's material singularity does much of the work of securing its identity.
A pharmaceutical is almost the opposite kind of poietic object. It is deliberately small, transportable, reproducible, divisible and substitutable, and it is manufactured in enormous numbers of units meant to be indistinguishable. Its very success as a mobile remediation creates a property problem. If a formulation can be copied easily enough, ownership cannot reside primarily in the possession of a singular physical object, because no singular object matters. Property accordingly attaches to symbolic descriptions of reproducible relations: a molecular structure, a synthetic process, a formulation, a therapeutic use, a brand, a regulatory dossier, a body of clinical data. What is owned is the authorised right to reproduce a specified class of tablets, or to exploit a specified medicinal relation, rather than any particular tablet.
Property of this kind is a convention. Godfrey-Smith (2024) argues that institutions such as money rest on interlocking behaviour and expectations within a community, and that such conventions are fragile without being illusory. A patent holds while offices, courts, customs authorities and competitors behave as if it holds, and it can be reopened when someone with standing and evidence challenges it, as Section VIII shows.
The legal instruments differ, and the differences matter. A patent protects a qualifying invention, must be applied for and granted, and lasts for a limited term, usually twenty years from filing. A trademark protects a brand name and its associated signs. Regulatory systems add further forms of exclusivity, such as periods during which a competitor may not rely on the originator's clinical data to obtain approval for a copy. The broader theoretical point matters more than this taxonomy. Mobility and reproducibility push property upward toward symbolic specification, and the more easily matter can be copied, the more value comes to reside in the right to specify, authorise and exclude.
The Indian case makes the point historically sharp. Between the Patents Act of 1970 and its amendment in 2005, India granted patents on processes for making medicines but not on medicines as products, so that the same molecule could be made legally by a different route. An immense generic industry grew within that opening and became a supplier of affordable medicines to much of the world. Compliance with the World Trade Organisation's agreement on Trade-Related Aspects of Intellectual Property Rights restored product patents in 2005, and the amended Act's section 3(d) excluded new forms of known substances unless they showed enhanced efficacy. In Novartis AG v. Union of India (2013), the Supreme Court upheld the rejection of a patent on a crystalline salt form of imatinib on this ground. What was contested, in each phase, was where along the chain from molecule to process to form to use the ownable specification should be located.
The Indian market also shows the reverse movement, symbolic differentiation laid over material sameness. Much of it consists of branded generics, the same off-patent molecule sold by many manufacturers under many brand names, often at different prices and with different reputations (Ecks 2022). The material invariance required for portable sameness is shared across these products, while brand names, packaging and the relationships between companies, doctors and pharmacists carry the differences that matter commercially. Portable sameness at the level of the molecule coexists with deliberate symbolic difference at the level of the product.
Law also participates in stabilisation. Law determines who may manufacture. Regulation defines acceptable identity, purity and performance. Licensing defines legitimate distribution, prescription law defines legitimate access and labelling rules stabilise intended use. Pharmacopoeias define standardised material identities, and the World Health Organisation's International Pharmacopoeia seeks international harmonisation of those specifications. The roots of this are old. Venice regulated the public preparation of theriac for centuries, requiring that its ingredients be inspected and its compounding witnessed, because a compound of dozens of substances could not be verified by looking at it (Griffin 2004).
Legal symbolisation is part of what produces the globally transportable identity of the medicine. A tablet that cannot be connected to an authorised manufacturer, batch, dose and label becomes epistemically unstable even if its molecules happen to be correct. The World Health Organisation estimated in 2017 that roughly one in ten medical products in low- and middle-income countries is substandard or falsified (WHO 2017). A falsified medicine is a symbolic identity detached from the material relation it claims, and a substandard one is a material relation that has drifted from its symbolic identity. Both are failures of portable sameness, and both can be detected only through the coupling of matter and symbol that portable sameness requires.
VIII. Turmeric, Traditional Knowledge and Competing Symbolic Stabilisations
In December 1993 two researchers at the University of Mississippi Medical Center, both of Indian origin, filed a United States patent application on the use of turmeric in wound healing, and in 1995 the patent was granted, covering oral and topical administration of turmeric powder. In 1996 India's Council of Scientific and Industrial Research requested re-examination, submitting evidence of prior use that included classical Sanskrit texts and a 1953 article in the Journal of the Indian Medical Association. In August 1997 the patent office rejected all six claims for lack of novelty (Jayaraman 1997), and a re-examination certificate was issued the following year. Those involved in the challenge later noted a striking difficulty: although turmeric's use on wounds was known in countless Indian households, written documentation of exactly that use, in a form a patent examiner could recognise, was hard to find.
The case is often told as a simple story of Western appropriation, and the inventors' Indian origin already complicates that story. A more precise question is more revealing. What happens when an existing medicinal relation between matter and embodiment is redescribed within another symbolic system as a potentially novel invention? Turmeric had long existed simultaneously as spice, dye, cosmetic and medicine, and its medicinality illustrates the opening argument of this article, that substances do not belong intrinsically to exclusive categories and that their medicinal significance is relational. The patent claimed no new material, transformation or combination. It claimed a use, and so attempted to convert a relation into property by redescribing it in the idiom of a different symbolic system.
A granted patent is a closure. It settles, for the term of the patent, the question of who may exploit the action of turmeric on wounds, and it does so without asking the households that had long used turmeric in this way. Closures of this kind are provisional, and the burden of reopening one is rarely symmetrical. The party that benefits from a closure needs only to let it stand, whereas the party disadvantaged by it must find a forum, standing and evidence. The Indian agency had to identify the patent, retain counsel in a foreign jurisdiction at a reported cost of about fifteen thousand dollars (Jayaraman 1997), assemble documents in several languages and, above all, find a form of evidence the examiner could recognise. The difficulty noted by those involved is the difficulty of reopening a closure with material that the closure's own symbolic system does not count as evidence. The asymmetry in the burden of reopening is a dimension of power, and the turmeric case shows it operating between jurisdictions and between kinds of knowledge.
India's response was institutional. The Traditional Knowledge Digital Library, begun in 2001 as a collaboration between the Council of Scientific and Industrial Research and the government department responsible for Indian systems of medicine, transcribed formulations from classical Ayurvedic, Unani and Siddha texts, and later Yoga texts, into several international languages, and organised them through a Traditional Knowledge Resource Classification designed to map onto the International Patent Classification used by examiners. Patent offices, including the European Patent Office, were given access under non-disclosure agreements, and the library has since been used to challenge many applications before grant. The Library redistributes the burden of reopening in advance. Instead of leaving each challenge to be organised after a patent has issued, it places the evidence where examiners look before grant, so that what would have been a costly induced opening becomes a routine step of examination.
The arrangement carries its own tension. To protect traditional knowledge against patenting, the Library had to make it legible to the very institutions from which protection was sought, while restricting access so that the translated formulations would not themselves become a free resource for product development. Symbolic portability was granted to examiners and withheld from others, a selective invariance of access that mirrors the selective invariance of the medicine itself.
The Library is best understood as an act of defensive intersymbolic stabilisation. The knowledge it contains was already symbolic, often textual and sometimes very old. Its difficulty was inadequate portability between symbolic systems rather than any absence of writing or classification, since knowledge stabilised in Sanskrit verse, in Unani pharmacopoeias and in household practice could not travel into the search tools of a patent examiner in Washington or Munich without being reformatted. The route runs from classical text, through translation and classification, to searchable prior art and an examiner's decision. The parallel with pharmaceutical mobility is exact. A medicine must be materially stabilised so that it can cross climates and supply chains without ceasing to be the same medicine. Traditional medicinal knowledge must sometimes be symbolically reformatted so that it can cross legal regimes without losing the evidence of its prior existence.
This reframes the contrast between traditional and modern medicine. The oppositions between natural and manufactured, or between cultural belief and scientific medicine, obscure what is at stake. The productive questions are how many multimaterial transformations have occurred, how the remediative relation is stabilised, which symbolic system defines efficacy, which institution authorises circulation, which system can establish prior use, who bears the burden of reopening a closure, and who acquires the right to reproduce the relation or to profit from it. Intellectual property conflicts over medicinal knowledge then appear as contests over which symbolic stabilisation will govern a medicinal relation that is already multimaterial, already mobile and often already very old.
IX. Conclusion: Medicine as Mobile Multimediated Remediation
Between the chimpanzee chewing bitter pith and the blister pack on a pharmacy shelf, one relation remains recognizable: a felt or discerned bodily misalignment, a recruitment of matter and an anticipated remediation. Everything else is evolutionary and historical layering. Found medicinal matter became collected and stored matter, stored matter became processed matter, processed matter became compounded matter, compounded matter became purified and standardised matter, and standardised matter became projectively fabricated matter. Each step added to the earlier ones without removing them, and symbolic stabilisation came to specify ever more exactly what is being made, for whom, at what dose, for which condition, under which conditions of storage, by whom and with what legal authority.
None of this makes medication less zoetic. The most elaborately synthesised molecule still has to meet a metabolising body. The globally standardised tablet still has to be swallowed, absorbed and tolerated. A flawless regulatory dossier cannot make an ineffective molecule effective, and a patented drug cannot produce health merely because its property rights are clear. Downward answerability applies to medicines with particular force, because they are designed to leave the conditions in which their fit was tested, and because their consequences reach beyond the patient: to the bacteria that adapt to them, to the scavengers that eat what treated bodies leave behind and to the animals through whose bodies a candidate must pass before it reaches a human one. Medicines became progressively more mobile, reproducible and symbolically stabilised, but they never escaped the embodied processes they were made to remediate. The pharmaceutical industry can manufacture extraordinary material sameness across the globe; it cannot manufacture away the living variability to which every medicine must finally answer.
Medication brings into view a variable that runs through every made form: how far it can travel while remaining the same in the respects that matter. Bridges, medicines, food, clothing, money, weapons and digital hardware could be compared along this axis of mobility, and each would show a different distribution of the burden of stabilisation between matter and symbol. The history of medication is accordingly the cumulative history of making remediation portable while keeping mobile matter aligned with living bodies and the multimediated worlds in which those bodies live, rather than a movement from nature to chemistry or from traditional remedies to modern drugs.
Selected References
Copyright, Designs and Patents Act 1988. c. 48. London: HMSO.
D'Costa, Vanessa M., Christine E. King, Lindsay Kalan, et al. 2011. "Antibiotic Resistance Is Ancient." Nature 477: 457–461.
Davies, Julian, and Dorothy Davies. 2010. "Origins and Evolution of Antibiotic Resistance." Microbiology and Molecular Biology Reviews 74 (3): 417–433.
de Craen, Anton J. M., Ted J. Roos, A. Leonard de Vries, and Jos Kleijnen. 1996. "Effect of Colour of Drugs: Systematic Review of Perceived Effect of Drugs and of Their Effectiveness." BMJ 313: 1624–1626.
de Laet, Marianne, and Annemarie Mol. 2000. "The Zimbabwe Bush Pump: Mechanics of a Fluid Technology." Social Studies of Science 30 (2): 225–263.
De Vos, Paul. 2010. "European Materia Medica in Historical Texts: Longevity of a Tradition and Implications for Future Use." Journal of Ethnopharmacology 132 (1): 28–47.
Ecks, Stefan. 2013. Eating Drugs: Psychopharmaceutical Pluralism in India. New York: New York University Press.
Ecks, Stefan. 2022. Living Worth: Value and Values in Global Pharmaceutical Markets. Durham, NC: Duke University Press.
Freymann, Elodie, Susana Carvalho, Leif-Alexander Garbe, et al. 2024. "Pharmacological and Behavioral Investigation of Putative Self-Medicative Plants in Budongo Chimpanzee Diets." PLOS ONE 19 (6): e0305219.
Godfrey-Smith, Peter. 2024. Living on Earth: Life, Consciousness and the Making of the Natural World. London: William Collins.
Griffin, J. P. 2004. "Venetian Treacle and the Foundation of Medicines Regulation." British Journal of Clinical Pharmacology 58 (3): 317–325.
Hardon, Anita, and Emilia Sanabria. 2017. "Fluid Drugs: Revisiting the Anthropology of Pharmaceuticals." Annual Review of Anthropology 46: 117–132.
Henrich, Joseph. 2016. The Secret of Our Success: How Culture Is Driving Human Evolution, Domesticating Our Species, and Making Us Smarter. Princeton: Princeton University Press.
Huffman, Michael A. 1997. "Current Evidence for Self-Medication in Primates: A Multidisciplinary Perspective." Yearbook of Physical Anthropology 40: 171–200.
Huffman, Michael A. 2001. "Self-Medicative Behavior in the African Great Apes: An Evolutionary Perspective into the Origins of Human Traditional Medicine." BioScience 51 (8): 651–661.
Huffman, Michael A., and Mohamedi Seifu. 1989. "Observations on the Illness and Consumption of a Possibly Medicinal Plant Vernonia amygdalina (Del.), by a Wild Chimpanzee in the Mahale Mountains National Park, Tanzania." Primates 30 (1): 51–63.
Jayaraman, K. S. 1997. "US Patent Office Withdraws Patent on Indian Herb." Nature 389: 6.
Latour, Bruno. 1986. "Visualization and Cognition: Thinking with Eyes and Hands." Knowledge and Society 6: 1–40.
Laumer, Isabelle B., Arif Rahman, Tri Rahmaeti, et al. 2024. "Active Self-Treatment of a Facial Wound with a Biologically Active Plant by a Male Sumatran Orangutan." Scientific Reports 14: 8932.
Moerman, Daniel E. 2002. Meaning, Medicine and the "Placebo Effect". Cambridge: Cambridge University Press.
Newman, David J., and Gordon M. Cragg. 2020. "Natural Products as Sources of New Drugs over the Nearly Four Decades from 01/1981 to 09/2019." Journal of Natural Products 83 (3): 770–803.
Novartis AG v. Union of India. 2013. 6 SCC 1 (Supreme Court of India).
Oaks, J. Lindsay, Martin Gilbert, Munir Z. Virani, et al. 2004. "Diclofenac Residues as the Cause of Vulture Population Decline in Pakistan." Nature 427: 630–633.
Oudhoff, Menno J., Jan G. M. Bolscher, Kamran Nazmi, et al. 2008. "Histatins Are the Major Wound-Closure Stimulating Factors in Human Saliva as Identified in a Cell Culture Assay." FASEB Journal 22 (11): 3805–3812.
Paddon, Chris J., Patrick J. Westfall, Douglas J. Pitera, et al. 2013. "High-Level Semi-Synthetic Production of the Potent Antimalarial Artemisinin." Nature 496: 528–532.
Peplow, Mark. 2016. "Synthetic Biology's First Malaria Drug Meets Market Resistance." Nature 530: 389–390.
Riddle, John M. 1985. Dioscorides on Pharmacy and Medicine. Austin: University of Texas Press.
Sabaté, Eduardo, ed. 2003. Adherence to Long-Term Therapies: Evidence for Action. Geneva: World Health Organization.
Sneader, Walter. 2000. "The Discovery of Aspirin: A Reappraisal." BMJ 321: 1591–1594.
Sun, Duxin, Wei Gao, Hongxiang Hu, and Simon Zhou. 2022. "Why 90% of Clinical Drug Development Fails and How to Improve It?" Acta Pharmaceutica Sinica B 12 (7): 3049–3062.
Tu, Youyou. 2011. "The Discovery of Artemisinin (Qinghaosu) and Gifts from Chinese Medicine." Nature Medicine 17: 1217–1220.
United States Congress. 2022. FDA Modernization Act 2.0. Enacted as part of the Consolidated Appropriations Act, 2023, Public Law 117-328.
van der Geest, Sjaak, Susan Reynolds Whyte, and Anita Hardon. 1996. "The Anthropology of Pharmaceuticals: A Biographical Approach." Annual Review of Anthropology 25: 153–178.
World Health Organization. 2014. "WHO Good Manufacturing Practices for Pharmaceutical Products: Main Principles." Annex 2 in WHO Expert Committee on Specifications for Pharmaceutical Preparations: Forty-Eighth Report. WHO Technical Report Series 986. Geneva: World Health Organization.
World Health Organization. 2017. A Study on the Public Health and Socioeconomic Impact of Substandard and Falsified Medical Products. Geneva: World Health Organization.
World Health Organization. 2020. "WHO Good Storage and Distribution Practices for Medical Products." Annex 7 in WHO Expert Committee on Specifications for Pharmaceutical Preparations: Fifty-Fourth Report. WHO Technical Report Series 1025. Geneva: World Health Organization.
Wright, Larry. 1976. Teleological Explanations: An Etiological Analysis of Goals and Functions. Berkeley: University of California Press.