Abstract
Comparing a tonne of metal with an option on the Nasdaq-100 looks like a category mistake, since one is an identifiable lot of metal on a warehouse floor and the other is a right whose value derives from an index of share prices. This article treats the difference as a difference of derivational distance: the number and kind of materially and symbolically stabilised relations that separate a financial object from the worldly processes to which its value remains answerable. It reconstructs a stratigraphy of financialisation, from found conditions and stored resources through standardised units, transferable, divisible and combinable claims, claims on claims and projectively fabricated contingent claims, and it follows copper from ore body to warrant to future, and the Nasdaq option from corporation to share, price, index, fund and derivative. Three concepts emerge: derivational distance, mobility substitution, in which a claim moves while matter stays still, and cascading recursive reopening, in which crises spread as relations that had been closed become askable layer by layer. Because valuation is constitutively interrecursive and finance is projective toward futures that remain open, the article also treats markets as conventions whose stability is fragile without being illusory. A multimaterial classification of markets replaces the opposition between the real and the financial with a gradient. Finance repeatedly demonstrates downward answerability. Physical delivery, the stones in the nickel sacks, the run on a bank and stranded carbon assets show that symbolic elaboration enlarges the space of possible claims without abolishing the conditions under which claims can be honoured.
I. What Exactly Is Traded in a Financial Market?
Comparing a tonne of copper with an option on the Nasdaq-100 index looks at first like a category mistake. The two seem to belong to different ontological worlds. Somewhere in a warehouse licensed by the London Metal Exchange lies an identifiable lot of copper cathode, twenty-five tonnes of metal meeting a specified standard of purity, stacked on the floor of a building in Rotterdam, Busan or New Orleans. An index option, by contrast, is a right whose value derives from an index, which is calculated from the prices of shares, which are themselves legally constituted claims on corporations. Financial economics nonetheless calls both financial products, prices both on screens, and hedges one against the other in the same portfolios.
The usual way of dealing with this difference is to register it and move on. Copper is a commodity and the option is an equity derivative; one belongs to the metals desk and the other to the index desk; each has its own regulator, its own exchange and its own conventions. These classifications are indispensable for institutional purposes. They are also silent on the question an anthropologist would ask first: what kind of thing is each of these products, and how did such a thing become possible?
That question requires a methodological decision. A history of medication should begin before pharmaceuticals, with the recruitment of matter to remediate a body in trouble. An analysis of made things should begin before the object is already differentiated as the kind of thing it is. In the same way, an analysis of finance should begin before financial products, with the relations between living beings and materially consequential things from which financial claims were eventually built. The deeper question is how living beings came to make present claims on such things, to make those claims transferable, to separate them from immediate possession, to attach them to future events, to divide and combine them, to standardise them, and finally to construct claims whose immediate referents are other claims.
Posed in this way, finance takes its place within the much longer evolutionary history of multimateriality, the history through which living beings learned to recruit, displace, transform, combine and finally project matter. The framework used here treats multimateriality as one of five mediations through which living beings remain coordinated with a world, alongside multisensorial embodiment, being-with, multiversal dwelling and multisymbolization. Multimateriality rests on two preconditions, material dependence and metabolic externalisation, and unfolds through six partially ordered layers: recruited self-produced matter, recruited found matter, selected and displaced matter, transformed matter, combined matter and projectively fabricated matter. Its criterion is recursive uptake, since a material difference becomes multimaterial only when it enters a living being's further coordination. Multisymbolization, the coupled poietic mediation, deposits distinctions into durable symbolic form, and claims, contracts, prices and indices are multisymbolic deposits. They operate at different levels of recursivity. At L1, coordination proceeds smoothly and unreflectively; at L2, a misalignment is felt; at L3, it is articulated; and at L4, it is stabilised in classifications, rules and organisations.
Every made thing works only if it fits the conditions in which it must work, and fit comes in three kinds. Cosmic fit is answerability to conditions indifferent to every living being, such as gravity and the strength of stone. Mesocosmic fit is answerability to the lived world of bodies, other beings, places and practices. Intersymbolic fit is the consistency of symbols with one another, such as the agreement of measurements, specifications, ledgers and labels. The three do not reduce to one another, and success at one level cannot purchase success at another. The principle of downward answerability holds that later layers of made things remain answerable to the older conditions on which their viability depends. The word downward is genealogical and not hierarchical: it names dependence and claims no causal priority for matter in every event.
The argument of this article is that the difference between copper and the index option is best described as a difference of derivational distance. Both are real, both are material and symbolic at once, and both remain answerable to worlds they do not control. They differ in how many stabilised relations separate them from those worlds, and in how those relations are built.
The comparison serves as a stress test of the multimaterial framework. If finance could be shown to have escaped materiality, the framework's central principle, that poietic elaboration never abolishes the conditions of possible fit, would fail at its most ambitious case. The article argues that the principle holds, and that finance forces the framework to grow in the process.
II. Before Finance: Found Worlds and Materially Consequential Uncertainty
Finance cannot manufacture rain, copper deposits, seasons, mortality, the nutritional requirements of human bodies, harvests or the passage of time. These are found conditions, and every financial architecture is built on top of them. Financial theory easily begins after a vast amount of worldly differentiation has already occurred. Copper, wheat, land, labour and mortality risk appear in its models as variables, prices and exposures. Before they become variables, they are aspects of living worlds, and their consequentiality for living beings is what eventually makes claims on them worth having.
Copper shows the depth of this history. A copper-bearing ore body is about as close as finance comes to the found end of the multimaterial spectrum. Porphyry deposits in the Andes formed over millions of years from hydrothermal fluids circulating through cooling magma, and no human being made them. As long as such a deposit lies unrecognized in rock, it belongs to dwelling, the non-human-made conditions of a place. A long sequence of relations then follows its recruitment. The deposit is prospected and identified, extracted and displaced, crushed and concentrated, smelted and refined, cast into cathodes of commercially recognised purity, transported, warehoused, certified, represented by a document of title, specified in a contract and traded as a future.
Each step in this sequence belongs to a recognisable layer of multimaterial history. Extraction and transport are selection and displacement. Crushing, smelting and electrolytic refining are transformation. Casting into standardised cathodes, bundling and certification move toward the projective, since the cathode is made to answer a specification that existed before it. The warrant, the contract and the future belong to the coupling of multimateriality with multisymbolization. At no point in the sequence does a magical transition occur from material to financial. New mediations are progressively deposited on earlier ones, and the financial object accumulates above a multimaterial history that remains active beneath it.
Materially consequential uncertainty is what makes this accumulation worthwhile. A farmer does not know whether the harvest will be good, a merchant whether the ship will arrive, a smelter what the price of concentrate will be in six months, a household whether its breadwinner will live. Found conditions vary, and their variation matters to living beings with vital stakes. Much of the history of finance can be read as the elaboration of ways to redistribute that uncertainty among those who bear it, through storage, credit, insurance, partnership and eventually markets in claims. Uncertainty about found conditions is the ground from which financial claims draw their relevance, since nobody needs a claim on something whose future is guaranteed and whose consequence for life is nil.
The history is old. In Mesopotamia, small clay tokens representing quantities of grain, oil and livestock were sealed inside clay envelopes, and marks impressed on the outside of the envelopes recorded what was inside, a practice that has been argued to lie at the origin of writing itself (Schmandt-Besserat 1992). In Ptolemaic Egypt, grain deposited in state granaries could be transferred between accounts by written order, so that ownership of grain moved while the grain did not. Such practices show symbolic claims on stored matter preceding coined money by millennia.
A principle follows that will govern the rest of the argument. Later symbolic layers reorganise earlier material relations without abolishing the conditions on which their viability depends. A copper future reorganises how copper is produced, stored and priced. It cannot make copper that does not exist, and it cannot make a copper market that no longer answers to mines, smelters and users of the metal.
III. A Stratigraphy of Financialisation
The six layers of multimaterial history cannot be mapped mechanically onto finance, but a provisional financial stratigraphy can be derived from them. Like the stratigraphy of multimateriality, it is partially ordered and accretive: later relations presuppose some earlier ones and incorporate them without abolishing them, and different financial traditions have entered it at different points.
The sequence begins with materially consequential differences in found worlds: fertile and infertile land, ore and waste rock, good years and bad. It continues with possession or control of materially relevant things, a field, a herd, a store of grain. Then come stored and displaced resources, grain carried to a granary and held against a later need, the first separation of a resource from the time and place of its production. Then come standardised material units, weights of silver, measures of grain, bars of metal of known fineness, which make heterogeneous quantities commensurable. Then come symbolically stabilised claims on those units: tallies, receipts, bills and titles. Then come transferable claims that can change hands separately from the thing they concern. Then come divisible and combinable claims, shares in a voyage, fractions of a loan, portfolios of receipts. Then come claims on other claims, such as a share in a company that itself holds claims, or a note secured by a pool of mortgages. Finally come projectively fabricated contingent claims, whose present form is specified in relation to future states that have not yet occurred: futures, options and the families of derivatives built from them.
Material and symbolic mediation remain intertwined at every level of this sequence. The tally stick was notched wood and the receipt was clay or paper; the electronic record of a derivative position is held on servers in data centres that draw megawatts of power, and trading itself depends on cables, microwave towers and matching engines whose physical location is a source of competitive advantage (MacKenzie 2021). The transition the stratigraphy traces is accordingly a series of specific new possibilities and not a movement from matter to symbol.
First, the claim can travel farther than the thing. A bill of exchange drawn in Florence can be presented in Bruges while the goods it concerns never leave Italy. Second, the claim can change hands while the thing stays still. Grain in an elevator or copper in a warehouse can be sold many times without moving. Third, the claim can be divided without dividing the thing. A ship owned in sixty-fourth shares sails as one ship. Fourth, claims can be combined without combining their underlying things. A fund holds claims on companies across continents that have never had any relation to one another. Fifth, a claim can be written on another claim. An option on a share is a claim on a claim on a corporation. Eventually financial objects become possible whose proximate underlying is itself a symbolically constituted financial object, and whose relation to any found condition runs through several such objects in succession.
This is what the concept of derivational distance is meant to capture. Derivational distance is the number and kind of materially and symbolically stabilised relations separating a financial object from the worldly processes to which its value ultimately remains answerable. Two features of the definition matter. It counts kinds as well as numbers, because a warehouse warrant and an index calculation are relations of different kinds, and the kind determines how answerability is transmitted. And it refers to value that remains answerable, because distance never severs the connection, however long the chain becomes.
Derivational distance is fully compatible with reality. A derivative binds parties, moves money, appears on balance sheets and can ruin those who misjudge it. An index is real, and so is a corporation, and so is a share. They differ in their position within a chain of derivation. The vocabulary of abstraction, virtuality and fictitious capital tends to obscure this, since it describes the top of the chain as though it had lost contact with the bottom. The stratigraphy describes the same phenomenon as a lengthening of the chain, in which every link is real and every link can fail.
The concept also clarifies a familiar dispute. Critics of finance often describe it as detached from the real economy; its defenders reply that it serves the real economy by allocating capital and managing risk. Both positions presuppose a boundary between a real and a financial sphere. On the stratigraphic account there is no such boundary, only a gradient of derivational distance along which financial objects are distributed, and along which answerability to found conditions is transmitted more or less directly, more or less quickly, and with more or less distortion.
IV. Copper: The Short Derivational Chain
The London Metal Exchange, founded in 1877, trades copper in lots of twenty-five tonnes. The metal is physical in every respect. It has weight, chemical composition, shape, purity and location. It was extracted somewhere, refined somewhere and transported somewhere. It occupies space, and storing it costs rent. Contamination, damage, theft and the time needed to load it onto trucks all matter. None of this disappears when the metal becomes the underlying of a futures contract.
A commodity exchange nonetheless requires variation among pieces of metal to be reduced far enough that units become substitutable. Here standardisation becomes central. The exchange accepts for delivery only copper cathodes of a specified grade, produced by refiners whose brands it has approved after testing, in specified shapes, weights and bundles. The effect is remarkable. The exchange does not eliminate copper's material differences, which persist in every cathode. It specifies which differences are permitted to matter. Two cathodes from different refineries on different continents, differing in trace elements, surface finish and history, become equivalent for the purposes of a contract because both fall within the specification.
This is a problem Living Value Theory is well placed to describe. Actual pieces of copper are never cosmically identical, since at the level of atoms and impurities no two cathodes are the same. What the market needs is mesocosmically sufficient sameness: equivalence adequate to the uses to which buyers will put the metal and to the coordination the contract is meant to achieve. Financial fungibility is accordingly achieved equivalence rather than discovered identity. The market creates rules under which materially nonidentical things can be treated as sufficiently equivalent, and it maintains those rules through brand approval, sampling, assaying and the policing of warehouses.
The history of American grain markets shows the same achievement in its making. Before the 1850s, grain moved through Chicago in sacks that remained the property of the farmer who had filled them. The introduction of steam-powered elevators and of official grading made it possible to pour grain from many farms into the same bins, and to issue receipts for quantities of a given grade rather than for particular sacks. A farmer's wheat lost its identity in the elevator, and a receipt for number two spring wheat could be traded without reference to any particular grain (Cronon 1991). Grading was the institutional act that made grain fungible, and fungibility was the condition for the futures market that the Chicago Board of Trade developed in the following decades.
Warehousing makes the interface between material and symbolic stabilisation almost diagrammatically visible. An LME warrant is a document of title to a specific lot of metal in a specific licensed warehouse, now held electronically, and it passes from owner to owner while the metal stays where it is. The copper remains stationary while control over it becomes mobile. A symbolic representation circulates through the exchange's systems while tonnes of metal sit on a warehouse floor.
This is the reverse of the pharmaceutical problem. With medicines, enormous effort goes into making the material object itself mobile while preserving its identity, a problem of portable sameness solved through stability testing, packaging and cold chains. With warehoused commodities, one solution is to make the claim mobile while allowing the matter to remain immobile. This can be called mobility substitution: the replacement of the movement of matter by the movement of a claim on matter. Mobility substitution is one of finance's oldest and most consequential achievements, from Ptolemaic granaries to Chicago elevators to LME warehouses, and it adds a dimension to the comparative analysis of made things. For any configuration, it becomes possible to ask whether the matter moves, whether the claim moves, or whether both move, and at what cost.
Mobility substitution is never costless. The warehouse must be licensed, inspected and insured, and the metal in it must actually be there. It also creates its own frictions. Between about 2010 and 2014, queues to withdraw aluminium from some LME-licensed warehouses in Detroit and Vlissingen stretched to well over a year, while warehouse owners continued to earn rent on metal that buyers had paid for and could not collect. The physical rate at which metal could be loaded out of a building, a matter of forklifts and truck bays, became an issue of market structure, regulatory investigation and litigation. The stationary matter beneath the mobile claim reasserted itself as soon as anyone tried to make the claim collapse back into matter at scale.
V. Physical Delivery: When the Symbolic Chain Can Collapse Back into Matter
A physically settled futures contract contains, as a standing possibility, the termination of a symbolic relation in a material transfer. At the end of all the contracts, prices, screens, clearing arrangements and calculations, a party holding a position at expiry may become entitled to receive, or obliged to deliver, physical copper meeting the specified conditions. Most positions are closed out before that point, and only a small fraction of contracts ends in delivery. The possibility of delivery nonetheless shapes the whole market, because it ties the futures price to the price at which physical metal can actually be bought, stored and delivered. If futures drift too far from what physical metal costs, traders can profit by buying metal and delivering it, or by taking delivery and selling it, and their doing so pulls the prices back together.
Physical settlement is therefore theoretically remarkable. The whole symbolic tower contains a route back down. Physical delivery is an institutionalized mechanism of downward answerability: it guarantees that the financial contract cannot indefinitely redefine what counts as acceptable copper, or how much of it exists, without consequences for its relation to the physical market. Downward is meant genealogically here, naming the dependence of later layers on earlier ones without implying that the metal determines everything the market does.
The mechanism is most visible when it is tested. In 2014 it emerged that metal stored in bonded warehouses at the port of Qingdao had been pledged as collateral for multiple loans at once, through duplicated or falsified warehouse receipts. The claims had multiplied while the matter had not. In 2023 the LME found that some sacks held against warrants for nickel in a licensed warehouse in Rotterdam contained stones instead of metal. In both cases a symbolic system of receipts, warrants and ledgers had remained internally consistent while ceasing to correspond to what lay on the warehouse floor. The failure became visible only when someone went to look, and the consequences followed the chain upward from the floor.
These cases show the difference between intersymbolic fit and mesocosmic fit with unusual clarity. A ledger of warrants can be perfectly consistent with itself, every warrant accounted for and every transfer recorded, and still fail to fit the world, because the world contains stones where the ledger says nickel. Physical delivery exists because intersymbolic fit alone is not enough. The analogy with medication is exact. Regulation cannot legislate an ineffective molecule into efficacy, and a commodity exchange cannot legislate nonexistent copper into abundance. The symbolic system has considerable autonomy, since it can set grades, prices, margins and delivery rules, but its autonomy has limits that are enforced, sooner or later, by the matter it concerns.
VI. Cash Settlement: Lengthening the Tether
A comparison internal to commodity markets supplies something close to a natural experiment. In physical settlement, a financial claim becomes an entitlement that can terminate in material delivery. In cash settlement, a financial claim is resolved by comparing a contract price with a reference price and paying the difference in money. Nobody needs to exchange the underlying thing. Copper can be the nominal underlying of a cash-settled future or of a swap priced against the LME's official settlement price, and the contract can run to expiry without a single cathode changing hands.
Cash settlement is no modern invention. At the Dojima rice exchange in Osaka, officially licensed by the Tokugawa shogunate in 1730, merchants traded rice futures that were settled by paying differences rather than by delivering rice, alongside a spot market in rice tickets that did represent stored grain. The two markets coexisted, and the relation between them had to be managed by the exchange, since a futures market that no longer referred to the price of actual rice would have lost the connection that gave it purpose.
In cash settlement the material tether changes form without disappearing. The mechanism of answerability now runs through a symbolically stabilised representation of what the underlying material thing is worth. The reference price must itself be produced, whether by trading in a physically settled market, by a panel of assessors, or by a price-reporting agency surveying physical transactions, and each method has its own vulnerabilities to thin trading, manipulation and error. Cash settlement thus depends on a second derivational relation that physical settlement does not need: the relation between the reference price and the physical market it is supposed to summarize.
This distinguishes material delivery from referential settlement. Under material delivery, answerability is transmitted through the possibility of transferring the thing. Under referential settlement, it is transmitted through the fidelity of a representation. The difference is consequential. When the representation fails, as it did when the benchmark interest rate LIBOR was shown to have been manipulated by the banks whose submissions produced it, contracts worth hundreds of trillions of dollars were found to have been settled against a figure that did not mean what it claimed. No comparable failure is possible in a market where the buyer can demand the thing itself, though other failures, such as the stones in the nickel sacks, remain possible.
The comparison demonstrates that derivational distance can vary even where the nominal underlying remains identical. A physically settled copper future and a cash-settled copper swap concern the same metal and move with the same prices most of the time, yet they occupy different positions in the multimaterial and symbolic architecture, and they transmit answerability to copper by different routes. In a physically settled market, the futures price must converge on the price of the physical metal as expiry approaches, because at expiry the two become the same thing. In a cash-settled market, convergence is guaranteed by definition, since the contract is settled at the reference price, and the burden of keeping the reference price honest is transferred to whatever process produces it. This alone shows that commodity market is too coarse a classification for the questions asked here.
VII. Nasdaq: When the Underlying Is Already Made
At the opposite end of the comparison stands the Nasdaq-100, an index of about one hundred of the largest non-financial companies listed on the Nasdaq stock market, first calculated in 1985. It would be easy to describe products written on this index as having escaped materiality altogether. That description is wrong at every link of the chain, and following the chain link by link is the best way to see why.
The chain begins with a corporation. A corporation is already an extraordinary stabilisation. It consists of people, contracts, property, buildings, machines, data centres, stocks of goods, intellectual property, bank accounts, obligations, productive capacities and symbolic recognition as a legal person able to own, owe, sue and be sued. The largest companies in the index design semiconductors, build and operate the data centres of the internet, sell devices assembled from minerals mined on several continents, and depend on electricity grids, ports, fabrication plants and the labour of millions. Their corporate form is a multisymbolic achievement resting on a vast multimaterial base.
What holds the corporate form in place is convention. Godfrey-Smith (2024) argues, against Harari's (2014) account of companies and money as shared fictions, that such institutions rest on interlocking behaviour and expectations within a community, that a convention of this kind is sustained by the reasonable expectation that others will keep behaving as they have, and that conventions are fragile without being illusory. If enough of the interlocking behaviour is withdrawn, the arrangement comes apart. The corporation, the share and the currency are conventions of this kind, and the account of crisis below concerns what happens when the expectations that sustain them are tested.
A share is a standardised, transferable relation to that corporation, carrying rights to residual income, to vote and to a portion of assets on dissolution. The share is already at considerable derivational distance from the found conditions on which the corporation depends. It is also, in contemporary markets, itself held through a chain of intermediaries. Most shares in American companies are registered in the name of a single nominee of the central securities depository, and the investor's ownership exists as entries in the books of a broker, which holds its own position through the depository. Even the share, the apparently simple unit of ownership, is several derivational steps from the certificate that once represented it.
Shares acquire continuously changing market prices through trading. Those prices are produced by the interaction of orders matched by computers in a data centre in Carteret, New Jersey, where the exchange's matching engine runs and where trading firms pay to place their own servers as physically close to it as possible, since microseconds of transmission time are worth money (MacKenzie 2021). The material infrastructure of price formation is thus intensely site-dependent, a striking inversion of the apparent placelessness of the prices it produces.
Selected shares then become constituents of an index. The index is a rule-governed symbolic construction rather than a pile of companies: its methodology specifies which companies are eligible, how their weights are determined, how often the list is reviewed, how corporate actions are handled and how the level is calculated. The index level is a number produced continuously by applying these rules to share prices. It owns nothing and can be owned by no one, yet vast sums are committed to its movements.
Financial products can then refer to the index. A futures contract on the Nasdaq-100 is settled in cash against a special opening quotation of the index. Options on the index are settled in cash as well. An exchange-traded fund tracking the index holds the constituent shares and issues units that trade like shares, adding another material and symbolic vehicle to the chain, and options on the fund's units can be exercised into the units themselves. At the top of this chain, physical delivery reappears in a curious form: what is delivered is a claim on a fund, which holds claims on companies. Delivery collapses one symbolic relation into another rather than into matter. The fund is held to its index by a further mechanism that resembles physical delivery one level up. Large institutions known as authorized participants can hand the fund a basket of the constituent shares in exchange for new units, or return units in exchange for shares. If the units trade above the value of the shares they represent, it pays to create new units; if below, it pays to redeem them. This arbitrage ties the fund's price to its holdings much as the possibility of delivery ties a copper future to physical metal, with the difference that what is delivered is a basket of claims on companies rather than a lot of metal.
The chain now runs approximately from found conditions, through the infrastructures and resources on which production depends, through the productive activities of firms, to the corporation, the share, the market price, the index, the index-linked product and the derivative written on that product. Every link matters, and every link is real. The top of the chain is deeply derived rather than unreal. A shift in the price of the chips a company designs, a flood in a fabrication plant, a change in the cost of electricity for its data centres or a collapse in consumer demand passes through each link, in attenuated and transformed form, to the option.
This yields the article's central contrast. What distinguishes copper futures from Nasdaq derivatives is derivational architecture rather than the materiality of the one and the immateriality of the other: in the number of links between product and found conditions, in the kinds of relation those links are, and in whether any link offers a route back down to matter. The copper future's chain is short and ends, if a holder insists, in metal on a warehouse floor. The index option's chain is long, passes through legal persons, share registers, prices and rules of calculation, and ends in money.
VIII. The Recursivity of Financial Value
Finance contributes something to the multimaterial framework that the pharmaceutical and the building do not. A pharmaceutical's identity is ideally stabilised against other people's opinions. A tablet should not become chemically different because traders think poorly of it, and much of the regulatory apparatus of medicine exists to ensure that it does not. A financial object's valuation, by contrast, is constitutively interrecursive.
Keynes (1936) described the stock market as a beauty contest in which competitors must pick the faces that other competitors will find prettiest, and in which the sophisticated player anticipates what average opinion expects average opinion to be. I value a share partly in anticipation of how others will value it. Others anticipate how still others will value it. Those anticipations affect present prices, and present prices then become information to which participants respond. Valuation, anticipation of others' valuation, price, altered valuation and altered expectation succeed one another in a loop that has no natural resting point.
This goes beyond symbolic derivation into interrecursive valuation, in which the counterparts whose responses matter are other valuers whose responses respond in turn. Copper's price is interrecursively formed as well, since traders anticipate one another there too, but the loop is disciplined more tightly by the physical market, because the metal will eventually be used by someone for something. The longer the derivational chain, the more room the loop has to run before it meets such discipline.
The recursion can also travel downward through the chain. A company's prospects inform the valuation of its share; the share price determines how cheaply the company can raise capital, pay employees in stock, acquire other firms or reassure suppliers; those possibilities alter the company's decisions; and the decisions change its prospects. A falling share price can change the material possibilities of the company whose expected possibilities contributed to the fall. Soros (1987) called this reflexivity, and a substantial literature in the social studies of finance has shown that financial models themselves can alter the markets they describe, as when the Black-Scholes-Merton option pricing model became more accurate after traders began using it (MacKenzie 2006).
The performance of models has a further significance. A model that treats price movements as a process indifferent to being modelled applies nonrecursive logic to interrecursive coordination, and the mismatch tends to show itself as persistent coordination failure when the model is relied upon too widely. A model that is used by enough traders, however, changes the behaviour it describes, so that the market comes to resemble the model for a time. Both effects follow from valuation being a coordination among counterparts who respond to one another and to the instruments they share.
Finance therefore recursively alters the realities it represents. This prevents downward answerability from becoming a crude model of base and superstructure, in which prices merely reflect material facts. Causation runs in both directions. Higher symbolic formations can reorganise lower material conditions, financing mines that would otherwise not be dug and closing factories that would otherwise have continued, and they can do so through their own interrecursive dynamics. What they cannot do is abolish the conditions of viable fit. A company whose share price is sustained by expectations its operations cannot meet will eventually have to meet them or fail, however long the loop sustains it in the meantime.
IX. Futurity: Finance as Projective Multisymbolization
Finance adds a further dimension absent or less extreme in other domains of made things. Many financial products concern things that do not yet exist as events. Will the wheat be delivered? Will the borrower repay? What will a share price be in three months? Will interest rates change? Will a company default? A bond is a claim on future payments. A future is a present agreement about a future exchange. An option goes further, since its present identity depends on formally specified relations among possible future conditions: the right to buy at a given price if, and only if, the market price turns out to be higher.
Finance is accordingly an extraordinary form of projective multisymbolization. Projective multimateriality specifies a bridge or a molecule before it is built. Financial instruments specify relations to futures that may never arrive, and they price the specification now. The future itself is not known, and it is open in the strong sense that living coordination leaves it undetermined; Knight (1921) distinguished measurable risk from uncertainty that cannot be reduced to probabilities, and financial crises repeatedly expose the difference. Financial instruments do not bring future states into existence by describing them. They stabilise present relations to uncertain futures. Esposito (2011) has argued that finance trades in the future's own uncertainty, using present expectations about future expectations as its material.
A precise formulation follows. A financial product is a presently stabilised relation to specified future contingencies, rather than a piece of the future brought into the present. The distinction matters because it locates the product's reality where it belongs. The option exists now, as a binding relation among parties with present consequences for their balance sheets and their behaviour. What it refers to does not yet exist and may never exist. Its value is a present estimate of the weight of possibilities that remain open, and the estimate changes as those possibilities narrow.
Godfrey-Smith (2024), following Wright (1976), offers a way to place this practice among the ways in which things come about because of their effects. In one route, feedback from past cases does the work. In the other, a future outcome is represented and pursued even where nothing has yet produced it. Financial practice interleaves the two. Models are fitted to past prices, and past defaults and volatilities inform the estimate of what may come, so that the past supplies the feedback. The instruments themselves, however, are built to pay out under conditions that may never have occurred, and they are priced by a representation of the weight of possibilities that remain open. Finance is a practice of pursuing represented futures with the help of an archive of past ones, and the crises it suffers tend to occur where the archive fails to anticipate the represented case.
This formulation also explains why futurity and derivational distance compound one another. A copper future is a short chain projected forward in time. An index option is a long chain projected forward. Each link in a long chain has its own futures, since each company's prospects, each share's price and the index methodology itself can change before expiry, and the option's value integrates all of them at once.
The copper future shows how intimately futurity and matter are joined even in finance. The difference between the price of copper for delivery today and for delivery in three months is shaped by the cost of carrying the metal through that interval: warehouse rent, insurance and the interest forgone on the money tied up in it. When metal is plentiful, prices for later delivery usually exceed prices for immediate delivery by roughly this cost of carry, a structure traders call contango. When metal is scarce, users will pay a premium to have it now, and prices for immediate delivery rise above later ones, a structure called backwardation. The shape of the curve of prices across delivery dates is therefore a symbolic expression of material conditions: of how much metal sits in warehouses, of what it costs to store, and of how urgently it is needed. Time in the copper market is priced partly in the rent of warehouse floors. The index option has no comparable anchor, since an index cannot be stored, and its relation to time runs through interest rates, expected dividends and the anticipated volatility of prices alone.
X. Toward a Multimaterial Classification of Financial Markets
Conventional classifications of financial markets, by asset class, venue, settlement mechanism and regulatory category, remain useful and answer institutional questions. The multimaterial approach asks different ones, and it yields a set of comparative dimensions that can be applied to any market.
Markets differ first in derivational distance, the number and kind of stabilised relations that separate the traded product from its underlying worldly conditions. They differ in material deliverability, whether settlement can terminate in the transfer of a material thing, as with a physically settled copper future, or only in money or in another claim. They differ in fungibility burden, the amount of standardisation required to make heterogeneous instances exchangeable. Copper requires brand approval, grading and warehouse policing; shares of the same class in the same company are fungible by legal definition; bespoke derivatives are not fungible at all and must be valued one by one. Fungibility burden is required sameness, the tightness with which instances must conform for them to count as the same thing, achieved through standardisation for the purpose of exchange. Markets differ in site dependence, whether the underlying must remain somewhere specific, as warehoused metal must and an index need not. They differ in mobility substitution, whether the matter moves, the claim moves or both, and at what cost. They differ in futurity, whether the product concerns present possession, a future obligation or a contingent future possibility. They differ in the intensity of interrecursive valuation, how strongly present value depends on anticipation of other valuers. They differ in symbolic depth, whether the immediate referent is a material thing, an organisation, another claim, an aggregate of claims or another derivative. And they differ in downward settlement, in what ultimately happens when the symbolic relation has to be resolved, and in whether that resolution reaches matter, money or another claim.
These dimensions should not yet be hardened into a taxonomy. They form a multidimensional comparative method, one that allows comparison among commodity spot markets, physically settled futures, cash-settled futures, equity markets, bond markets, currency markets, index markets, options markets and highly derived structured products on a common footing. A government bond, for instance, has short derivational distance to a legal obligation of a state but long distance to the taxable activity that must eventually service it; it is fungible by design, settles in money, concerns a future obligation and is valued with moderate interrecursive intensity except in crises, when intensity rises sharply. A currency is a claim whose value rests almost entirely on interrecursive acceptance, backed by a state's capacity to tax and by the goods and services it can command.
The purpose of the method is to show that different markets solve different coordination problems through different combinations of multimaterial and multisymbolic stabilisation, and it involves no ranking of markets from primitive to advanced. A physically settled commodity market solves the problem of coordinating producers and users of a material thing across time and space. An equity market solves the problem of dividing and transferring claims on productive organisations. An options market solves the problem of distributing exposure to specified contingencies. None of these problems is more advanced than the others. They are different problems, and the dimensions of comparison show how each solution positions its products relative to the found conditions they ultimately concern.
The method also connects finance to the comparative analysis of made things more generally. Mobility substitution, fungibility burden and site dependence are properties that pharmaceuticals, buildings and banknotes possess as well, each in its own degree. Finance makes some of them extreme, and in doing so it reveals dimensions, such as derivational distance and interrecursive valuation, that the other domains possess only faintly.
XI. Crisis as Recursive Reopening
Under ordinary conditions, enormous parts of financial architecture operate at L1, in smooth, unreflective coordination. Nobody asks every morning whether a share really is a share. Nobody reconstructs the legal ontology of the corporation before buying one, and nobody reopens the foundations of money before accepting payment. Stabilisation of this kind is what makes finance possible, since a market in which every relation had to be re-examined before every trade could not function.
Crisis changes this. Questions proliferate: what is this asset actually worth; does the borrower have the money; what exactly backs this security; can this institution meet its obligations; is this collateral worth its recorded value; can this claim be converted into money; will someone accept this money tomorrow? Relations that had been closed become recursively open. What had operated at L1 is felt at L2 as misalignment, articulated at L3 as questions and doubts, and eventually addressed at L4 through emergency lending, guarantees, suspensions of trading and new regulation.
Financial crises can therefore be analysed as cascading recursive reopenings across derivational layers. In 2007 and 2008, doubts about American subprime mortgages, claims on the incomes of households and ultimately on houses, travelled up through mortgage-backed securities, through collateralised debt obligations built from tranches of those securities, through further obligations built from tranches of those, and into the market for short-term secured lending on which banks depended, where lenders began demanding more collateral or refusing to lend at all (Gorton and Metrick 2012). At each layer, a relation that had been taken for granted became askable, and the asking spread faster than anyone could answer it, because the information needed to answer it lay several layers down.
The bank run is the elementary form of this process. A bank deposit is a claim that the depositor treats, at L1, as money. The bank holds only a fraction of its deposits as cash, lending the rest, and this arrangement works as long as depositors do not all ask for their money at once. Once doubt arises, it becomes rational for each depositor to withdraw before others do, whatever the bank's underlying soundness, and the run can destroy a bank that would otherwise have survived (Diamond and Dybvig 1983). The run is a convention failing in the way conventions fail, by the withdrawal of the expectations that hold it together. Deposit insurance, lending of last resort and suspension of convertibility are L4 institutions whose purpose is to keep this relation closed, preventing the question of whether a deposit is really money from becoming consequentially askable. Their success leaves little trace, because a run that does not happen is not recorded.
Openings of financial closures differ in kind. A crisis is a disruptive opening, forced by a breakdown that the closure did not anticipate. Markets also contain scheduled openings, built into the closure from the start: daily marking to market, margin calls, audits, expiry dates and settlement cycles return valuations to question at fixed intervals, so that doubts arrive in small amounts and not all at once. And there are induced openings, brought about by parties with standing to bring them about, such as a regulator's examination, a short seller's report or a court. The burden of reopening is seldom symmetrical. Those who benefit from a closure need only let it stand, whereas those who doubt it must find evidence and standing, and the Qingdao and Rotterdam cases surfaced only when someone went to the warehouse floor and looked.
Derivational distance may matter here, though not mechanically. The more layers a financial architecture contains, the more relations it contains whose fit can become consequentially askable, and the more places at which reopening can begin and from which it can spread. It does not follow that greater derivational depth necessarily causes instability. Institutional redundancy, deep collateral pools, central clearing and lenders of last resort can make highly derived systems extremely stable for long periods, and short chains can fail catastrophically, as the stones in the nickel sacks showed. The proposition is more precise: derivational depth creates additional relations whose fit may become consequentially askable. Whether they become so depends on what happens at the bottom of the chain and on the institutions that stand between the bottom and the top.
The nickel market supplied a vivid instance of reopening at short derivational distance in March 2022, when prices on the LME more than doubled within hours amid a squeeze on a large short position, and the exchange suspended trading and cancelled a day's transactions. The episode showed that even a short chain can reopen violently when the interrecursive valuation loop runs faster than the physical market can respond, and that the exchange itself, as an L4 institution, can close a reopening by decree when its own viability is at stake.
XII. Downward Answerability: Why Finance Never Escapes the World
Finance achieves astonishing symbolic autonomy. Claims circulate without things. Claims are written on claims. Indices aggregate claims, derivatives refer to indices, and algorithms transact these objects at speeds no human trader can follow. None of this is illusory, and all of it has reorganised the material world, financing mines, factories, cities and wars.
Symbolic elaboration cannot, however, abolish the worlds on which its continued relevance depends. The copper still has to exist, the harvest still has to occur and borrowers must sometimes repay. Companies must eventually remain viable, and human beings must continue to treat monetary claims as consequential. All of this returns, finally, to living beings with vital stakes: people who need to eat, to be housed, to be cared for and to plan for futures in which they will still be alive. Finance is relevant because it redistributes the consequences of found conditions among such beings. If those conditions ceased to matter to anyone, finance would have nothing to redistribute.
Finance therefore offers perhaps the strongest demonstration yet of the general principle of multimateriality: poietic elaboration increases the space of possible forms without abolishing the conditions of possible fit. The extraordinary achievement of finance lies in allowing coordination to occur at enormous derivational distances from immediate material possession while preserving enough connection to underlying worlds for the claims to remain consequential, rather than in transcending materiality. Where that connection weakens, through fraud, through representations that no longer represent, or through loops of valuation that run far ahead of what their underlying can sustain, the principle reasserts itself, usually at great cost and usually to those least able to bear it.
Downward answerability does not always arrive through fraud or collapse. It can arrive through the found world itself. The recognition that a large share of known fossil fuel reserves could not be burned without exceeding agreed limits on global warming led analysts to describe such reserves as potentially stranded assets, claims whose valuation assumed extraction that the atmosphere could not accommodate (Carbon Tracker Initiative 2011). Insurers facing rising losses from floods, storms and wildfires have repriced or withdrawn cover in exposed regions, and those decisions travel upward into mortgage markets and property values. In such cases the conditions of possible fit are set by the chemistry of the atmosphere and the behaviour of weather, the most found of all found conditions, and the symbolic system of finance is obliged to register them whether or not its participants would prefer to.
XIII. Conclusion: From the Mine to the Option
At one end of the comparison stands copper-bearing rock in the earth. At the other stands an option on a financial index. Between them lies no clean boundary between material reality and symbolic fiction. There are instead cumulative layers of finding, extracting, transforming, standardising, storing, representing, claiming, transferring, dividing, aggregating, anticipating and deriving, each resting on those beneath it and each able to fail.
The payoff of the comparison is that financialisation is better understood as the construction of increasingly long, mobile, projective and recursively valued chains of claims that remain differently answerable to multimaterial worlds, rather than as dematerialisation. Three concepts emerged from the comparison: derivational distance, mobility substitution and cascading recursive reopening. Each describes something finance does to the architecture of made things, and each can be taken back to other domains to see what it reveals there.
The distance from a copper deposit to an index option is enormous, but it is a distance produced by accumulated mediations rather than the distance from reality to abstraction. Finance makes it possible for a claim to travel where matter does not, to divide what remains materially whole, to exchange what has not yet happened, and to derive new claims from claims already derived. None of these achievements repeals the worlds from which their relevance arose. Mines can be exhausted, harvests can fail, corporations can cease trading, borrowers can default and bodies can die. Financial abstraction is one of the most extraordinary things multimaterial and multisymbolic beings have learned to build upon multimateriality, and it has never been an escape from it.
Selected References
Carbon Tracker Initiative. 2011. Unburnable Carbon: Are the World's Financial Markets Carrying a Carbon Bubble? London: Carbon Tracker Initiative.
Cronon, William. 1991. Nature's Metropolis: Chicago and the Great West. New York: W. W. Norton.
Diamond, Douglas W., and Philip H. Dybvig. 1983. "Bank Runs, Deposit Insurance, and Liquidity." Journal of Political Economy 91 (3): 401–419.
Esposito, Elena. 2011. The Future of Futures: The Time of Money in Financing and Society. Cheltenham: Edward Elgar.
Godfrey-Smith, Peter. 2024. Living on Earth: Life, Consciousness and the Making of the Natural World. London: William Collins.
Gorton, Gary, and Andrew Metrick. 2012. "Securitized Banking and the Run on Repo." Journal of Financial Economics 104 (3): 425–451.
Harari, Yuval Noah. 2014. Sapiens: A Brief History of Humankind. London: Harvill Secker.
Keynes, John Maynard. 1936. The General Theory of Employment, Interest and Money. London: Macmillan.
Knight, Frank H. 1921. Risk, Uncertainty and Profit. Boston: Houghton Mifflin.
MacKenzie, Donald. 2006. An Engine, Not a Camera: How Financial Models Shape Markets. Cambridge, MA: MIT Press.
MacKenzie, Donald. 2021. Trading at the Speed of Light: How Ultrafast Algorithms Are Transforming Financial Markets. Princeton: Princeton University Press.
Schmandt-Besserat, Denise. 1992. Before Writing. Vol. 1, From Counting to Cuneiform. Austin: University of Texas Press.
Soros, George. 1987. The Alchemy of Finance: Reading the Mind of the Market. New York: Simon and Schuster.
Wright, Larry. 1976. Teleological Explanations: An Etiological Analysis of Goals and Functions. Berkeley: University of California Press.