Primitive Money

Some of the problems that culminated in today’s deafening silence were presciently articulated by A. Hingston Quiggin in a mid-twentieth-century monograph. Quiggin was a Cambridge anthropologist, an associate of A. C. Haddon’s. Her book, perhaps the most comprehensive catalog of forms of money around the world ever written, appeared in 1949. In those days, they called it “primitive money” (fig. 2).

Figure 2.

Quiggin 1949, title page

Quiggin’s book begins with a lovely sentence— “everyone, except an economist, knows what ‘money’ means”—a sentence that a great many writers in the anthropological literature on money love to quote. It’s a lovely sentence, because it exhibits some awareness of the voicing structure of forms of money talk; and although what she says about “everybody” is not true, as we have just seen, it does contain an embryo of truth embedded in it, namely, what she says about economists, a point to which I return in later discussion.

Even more interesting is the struggle in which she is engaged in the opening pages, a struggle to get beyond a thing-focal perspective on what the book itself is about:

Quiggin observes that the kinds of “things” her book is about are used in many different ways by the people whose things they are, which raises many questions: What exactly are these things? (Jewelry? Tea? Condiments? Money? Which one?) And some “things” are especially troublesome: “It is still more difficult to classify the varied ornaments. … Is a string of shell-money no longer currency when you wear it round your neck? Is a sovereign no longer money when dangled on your watch chain?” (3). In other words: When is something money? It gets worse: “Shells are merely shells on one island, but are used in trade exchange with another, where they form the currency” (3). In other words: Where is something money? But there’s more: “For the two parties in a transaction may themselves stand in different categories. The trader may consider that he is paying current money when he buys a fowl for ten lengths of brass wire; while the seller regards the exchange as ‘mere barter’” (2). In short: For whom is something money? All of these questions can be given taxonomy-based answers, and Quiggin does give some, but the taxonomic answers of yesterday lead to the deafening silences of today.

The more basic questions that remain unanswered in this entire literature are: How do you know? What are your observables? What criteria guide your observations? How can you use words like “money” to describe any such cases? And what exactly are you describing? Things? Or activity routines in which things play a part? And how do you distinguish these from all other activities involving them?

But questions like “How does the anthropologist know?” are almost irrelevant unless they are linked to a different question: How do the users know? In other words, if people do different things with salt on different days, if folks in neighboring tribes do different things with shells, the question really becomes: How does anyone on any given day in any given social interaction know what is being done with something at that time? And how do they know what response is appropriate to such behavior when it occurs? Whether you’re worrying about beeswax or woodpecker scalps, or about the stuff that is in your pockets right now, the real question is: For anything that anyone has ever called money, how do its users know what it is? What typifications of objects render (what we call) “salt” or “shell” or “beeswax” appropriate to the range of activities in which they play a part (for them), and where do such typifications come from? What are the indexicals of person, place, and time that differentiate criterial activities from others, specify the participation frameworks in which such activities appropriately occur and, if these activities are symmetrically grasped by parties to such conduct, allow social interactions to unfold among them? Such questions remain unanswerable as long as talk of “things” obscures activities involving them. Since these questions are central to what follows, let us first consider what it means to shift our perspective from things to forms of interpersonal conduct involving things.

The important point to note regarding figure 3 is that we are looking not at things but at activity routines involving things. Each of these images depicts a small stretch of social history, organized by sociocentric, mutually coordinated and interpersonally recurrent activity routines. That is to say, before it can be used as money, the tobacco in figure 3,A has to be grown each year in seasonal agricultural cycles through a coordination of the activities of many people, and, once grown, always already acquires many other uses long before it becomes a money token anywhere. The cowries that are lying in a neat little pile in figure 3,B have to be brought here and piled first, which means that, for the eighteenth- and nineteenth-century cases we know best, they have to be harvested from the sea in places like the Maldives islands, year after year, buried in the ground to kill the mollusks that live inside them, sun-dried on mats and then transported by ship (whether to Bengal or Europe or Africa or elsewhere), where they acquire many uses that have nothing to do with money, and some that do (Hogendorn and Johnson Reference Hogendorn and Johnson1986, 82–83, 88–100, passim).

Figure 3.

From “things” to activity routines. Sources: A, Creative Commons license 4.0, © Derek Ramsey, derekramsey.com; D, Creative Commons license 2.5, Classical Numismatic Group, cngcoins.com; E, Wikimedia Commons, US public domain.

As for the ore (fig. 3,C), where did it come from? It has first to be extracted from the ground, which means that people must learn to differentiate it from the rock in which it is embedded, learn to dig mine shafts, and, by acquiring a symmetric grasp of forms of tightly coordinated multiparty routines, learn to extract it from the ground and to melt it into the form that lies before you here. And it has to be available in sufficient supply—which means it must already have acquired other uses—before any type of money-token conduct can become associated with it.

In figure 3,D, we have the first European precious metal coin, manufactured in Lydia, Central Asia, present-day Turkey, around the sixth century BCE. It is associated with a new set of activity routines involving metallic ore, which include making discs like this from it and using them as money tokens. But if these are new activity routines, how do people come to know what they are? How is awareness of this issue symmetrically grasped by sufficiently many people that conduct involving it can effectively count as money-token conduct in some society?

The issue of symmetric grasp is vital, of course, since no mutually calibrated social interaction is possible unless the parties to this interaction can typify the exponents of conduct—the things perceivably deployed as money tokens in conduct—in relatively symmetric ways, or assign them comparable significance. More generally, the question is one of the differentiability of such exponents of conduct from other things, on their identifiabiltiy and characterizability as objects of some specific kind, which require criteria that differentiate them from all other entities in the universe, and specify some among them as money tokens, and others as not money tokens. After all, the money token is often just a handshake. How do you learn that what’s crucial is not the hand but the thing in someone’s hand? And once you know which one it is, how do you know the what-where-when of its deployment in conduct? The way in which the human animal achieves this is through the denotational machinery of discursive semiosis, whereby all such questions can be posed and in principle settled—questions of the characterizability of entity-types through denotation, of the identifiability of specific tokens through selective deixis, of when-to-use and when-not-to-use such stuff through modalized predication, of symmetric interpersonal grasp through talk with others about it (Agha Reference Agha2007a, 37–55, 84–144, passim)—which is why humans tend to have artifacts and activity routines of such kinds, while gophers, marmosets, and porcupines apparently do not. But such denotational machinery is merely a fragment of the forms of discursive semiosis involved in knowing the what-where-when of money-token conduct, as we shall soon see.

What do you do when you don’t know? Well, you can always ask a question. Since trade often involves travel, Georg Knapp, writing in the 1920s, describes this as the traveler’s problem:

Knapp is of course writing two and a half millennia after coinage has been introduced into Europe, at a time when many forms of money talk and money-token conduct are well known to the common sense of Europeans, so that the early twentieth-century traveler can already presuppose that every country has a distinct coinage. Under these conditions, the traveler can simply ask questions like: “What is yours called? What does it look like? What’s it worth?”

But if we go back two and a half thousand years, to a stretch of social history like figure 3,E, the situation is quite different. At this time metal coins are fairly new. The Greeks have adopted the Lydian practice of minting coins. Different city states have different coins. It’s also a heteroglossic situation. We know from Homer that many dialects of Greek—including dialects of the Ionic, Aeolic, and Doric groups—are probably spoken in the circuits in which trade occurs, and some degree of multilingualism is needed in order to do trade. Not everyone uses coins. Not everyone knows what they are. Any coin presupposes a metasemiotic framework for its own identifiability and characterizability as a coin of some specific kind as a condition on its usability as a coin. When the presupposition is not satisfied, when someone is not familiar with it, explicit metasemiotic descriptions are needed to identify and characterize it. Suppose you meet someone who has heard of coins but doesn’t know where the one in figure 3,E is from. He asks you about it. What do you do?

There’s no getting around proper name deixis at this point. You might say something like “This ain’t just a head, it’s Athena’s head!” using the personal proper name of the goddess; and, flipping it over, “This ain’t just an owl, it’s Athena’s Owl,” the owl of wisdom associated with the goddess. And then you might point out that the goddess has a namesake, in a city name: “It’s from Athens, don’t you see?” If your interlocutor knows something about Greek religion (even if he doesn’t know much about coins), he “gets it.” If not, there goes the evening. In such a scenario, speech (or verbal signs) and metal discs (pictorial signs) are deployed together as perceivable exponents of conduct in multichannel semiotic arrays (cf. n. 2). The discursive exponents of conduct (the utterances you’ve just produced) are metasigns that typify their object signs (metallic discs), identifying and characterizing them as coins produced by the mint at Athens, thus giving them a very specific object formulation.

In cases where no script occurs on the money token, deictically anchored spoken utterances—which may contain not only speech event deictics or “shifters” but also speech-chain deictics such as proper names and titles, definite descriptions that combine these with other expressions, and the usual variety of other social indexicals (Agha Reference Agha2007a)—can clarify the What-Who-Where-Etc. attributes of the thing and its uses.

What happens when script does occur? How are these attributes typified in those cases? They tend to be typified by the script artifact itself. Figure 4 shows a Mesopotamian money token and script artifact, a clay tablet inscribed with cuneiform script by hand and stylus. It says that “Amil-mirra will pay 330 measures of barley to the bearer of the tablet at harvest time.” The activity to be conducted by it is metapragmatically typified by speech that occurs as script on the money token itself. What is to be done? Paying 330 measures of barley. Who will do this? Amil-Mirra. When? At harvest time. To Whom? The bearer of the tablet.

Figure 4.

Script artifacts. Source: Creative Commons license 4.0, © The Trustees of the British Museum.

So some features of the interactional schematization of money-token conduct—its What-When-Who-To Whom aspects—are explicitly denoted by the script written on it. But the Where? is not specified through denotational content. So where will this transaction happen? The Where? of conduct is typified not by the denotational content of cuneiform script but by the ability to read the script. If you can read the script, then you are a resident of this locale (and not of Europe or of some other place). The readability of the script is implicitly indexical of the “where” of reading and transaction, of a stretch of social history of which this tablet is a fragment.

In a third type of case, an artifact neither is dedicated to an exclusively money-token use nor contains script, but discursive semiosis nonetheless mediates its uses as a money token (fig. 5). Due to the shortage of coin available to the American colonists in the seventeenth century, corn and tobacco were stipulated to be money tokens—corn in the North, tobacco in the South. The Massachusetts law of 1630, in figure 5,A, stipulates that corn is legal tender in that colony. And in the South, as indicated in figure 5,B, Virginia made tobacco legal tender in 1642, as did other southern colonies at various times. So corn is declared legal tender in Massachusetts in 1630, and tobacco is declared legal tender in Virginia in 1642, though each has other uses too.

Figure 5.

Comestibles as money tokens. Sources: A, Wikimedia Commons Public Domain, Photo courtesy of USDA NRCS; B, Creative Commons license 4.0, © Derek Ramsey, derekramsey.com.

What are the difference in the forms of social interaction possible through corn and tobacco before and after the law? Let us bear in mind that before the law is passed, corn and tobacco are not just things, but things that play a part in already established sociocentrically organized activity routines in these places: They are cultivated in recurrent seasonal agricultural cycles each year through the discursive coordination of a carefully organized sequence of agricultural activities involving many people, farm animals, tools and implements, and are harvested each year and used as comestibles, activities that occur both before and after these laws are passed.

The passage of these laws—in 1630 in Massachusetts, in 1642 in Virginia—induces a segmentation of activity routines. It introduces a new use in each locale:

The new usage (in italics) is specific to participation frameworks and locales of social interaction. That is, you can’t use tobacco as money token in Massachusetts. You can’t use corn as a money token in Virginia. In other words, the appropriateness conditions on their money-token usage are different. One is appropriate in the North, the other in the South. We have two different registers of money-token conduct. Two new ones. Each new to a different locale, its social domain.

But the physical objects—bundles of corn or tobacco—themselves contain no instructions on the What-How-When-Where-With Whom-Etc. of the new usage. The law is not written on the corn or on the tobacco. Only those who are acquainted with criterial forms of discursive metasemiosis (namely, the new legislation in each locale)—and independently of their familiarity with corn as such, or with tobacco as such—can use their knowledge of the law to segment their own activity routines into money token “types” of usage versus other “types” of usage. And of course in order for such money-token usage to occur, many people in the North must be acquainted with the 1630 law, and many people in the South with the 1642 law. If only one person knows about them, corn and tobacco can’t function as money tokens. A criterial population can learn about the new activity routines involving corn or tobacco only by learning about the new law.

We know that literacy is not widespread at this time. So the law can become widely known only through some form of verbal retelling. The legal discourse has to be recycled into vernacular idioms. In order for it effectively to organize most people’s activities, it must—since most people are illiterate—become part of an oral tradition through which the activity segmentations induced by the law become known as common custom. Now, recall Quiggin’s discussion of similar segmentations of activity routines elsewhere:

This is much like Massachusetts corn: before 1630, you could just eat it; after 1630 you could both eat it and use it as currency, at least in Massachusetts. And it is much like Virginia tobacco: before 1642 you could chew or inhale it; after 1642, you could chew or inhale it, and you could use it as legal tender, at least in Virginia.

All differentiations of sociocentric activity routines involving money-tokens rely on forms of discursive semiosis to become intersubjecively ratified social realities, whether these be written statutes or matters of so-called custom or common law. Quiggin’s own data (museum collections of money-tokens from around the world) were gathered by various people at various times, mainly by people who paid little or no attention to the forms of discursive semiosis through which such differentiations occurred in places where they gathered them. And she is aware of the difficulty: “Unfortunately it is too late now to discover actual or ‘ceremonial’ use of many of these objects. All that can be done is to collect what information is still available” (Reference Quiggin1949, 4–5).

Script Artifacts

Let us turn to the case—all too familiar today—of money tokens that are script artifacts made of paper, and to the locale where their issuance first proliferated in the Western hemisphere, namely early America. Yet some of the early American script artifacts that came to be used as money tokens were originally designed for other uses, like the playing cards in figure 6. These cards are much like cowrie shells, or corn or tobacco in one sense: They have use differentiations, but they do not describe their own uses. The forms of discursive semiosis that specify their use characteristics are “off the page.” Sometimes they do have brief stretches of scrawled script, as on the card at bottom left. But even these leave out much of the verbal agreement that enables their (otherwise irregular) money-token use. These cards were given as money tokens to workmen who built a cathedral in St. Louis around 1790 or so. The forms of discursive semiosis that enabled people to use them in this way (including the terms of the verbal contract that allowed them to be treated as money tokens) are highly underspecified by the scrawl itself; the scrawl is merely the persistent (transcribed) fragment of a larger process of (spoken) discursive semiosis that no longer survives.

Figure 6.

Playing cards as money tokens

Since there’s a severe shortage of coin in colonial America at this time, laws are passed in various provinces that enable the use of pieces of paper as currency. Here’s an early example from Massachusetts:

We can see that this law gives pieces of paper a specific object formulation—that they are “Bills of publick credit,” and that they are “equivalent to money,” that is to say, are to be treated as equivalent to precious metal coinage. The government has explicitly to deal with the novelty of the money token itself, to persuade its citizens that paper money is as good as “stampt silver.” And it gives activities of using paper money a specific interactional organization: the colonial government first issues bills to the public, encourages them to take these seriously as “equivalent to money” in their own private affairs, and then accepts them back from citizens in “all publick payments,” such as taxes. It offers an inducement for usage by offering a 5 percent discount on tax payments made through them. Similar laws are passed in the other colonies.

Dedicated Money Tokens

After such laws are passed, we begin to see a proliferation of varied types of pecuniary media, all of which are initially new and unfamiliar to the relevant population of persons for whom they are designed, and by whom they are to be used. In order to enable such persons to be able to imagine what, in particular, such objects might possibly be for, pieces of paper that have acquired some sort of dedicated use as money tokens must contain explicit descriptions of their own characteristics, of the kinds of activities possible through them, and of the time, place and social-category membership of both users and issuers, and thus of the types of persons and statutory organizations linked to each other through them.

The script artifact in figure 7,A, for instance, begins its narrative by deictically individuating itself as “THIS” (its referent: the thing in your hand at the moment of looking), characterizing itself as an “Indented bill” for “One Shilling,” characterizing the Where of its usage as “the Province of Pennsylvania,” and the When of its issuance relative to two distinct indexical origos or zero points of time reckoning, namely, the ascension of the king to the throne (the origo for: “the 17^th Year of the Reign of King George II”) and the birth of Christ (the origo for: “Aug. 1, 1744 [Anno Domini]”), thus locating all actors and activities involving itself not merely within a geographic locale, but, given the divine right of kings, within a religious and political cosmic polity. Says Who? The seal and signatures at its bottom attest to and formulate its validity, vouch for its authenticity.

Figure 7.

A, Dual indexical origo (Pennsylvania, 1744); B, Less detail (North Carolina, 1776); C, More detail (Massachusetts Bay, 1775).

The discourse genre exhibits variation: sometimes the text is more elliptical, sometimes far more detailed. The document in figure 7,B is laconic: A stacked series of noun phrases specify the where (“N Carolina”) and when (“April 1776”) of its issuance, what it is (“Currency”), its amount (“one eighth of a dollar”), while nested prepositional phrases specify who issues it (“By Authority of Congress at Halifax”), and a culminating seal and signature formulate its authenticity. But, whereas the document in figure 7,B contains no complete sentences, the one in figure 7,C has an extended epistolary form and an elaborate contractual content that provides detailed instructions on the sociocentric activity routines to be carried out through it, including who is to carry them out (“The Possessor of this Note”; the note is transferrable), where he can do so (the “Colony of the Massachusetts Bay”), what he is licensed to do (“to receive … the sum of Fourteen Shillings” of “lawfull money”; implying thereby that it is not itself “money,” on which more below), from whom (“the publick Treasury of this Colony”), when the note is issued (“May 25, 1775”), and specifying, in addition, two transactional options for its bearer thereafter, namely, first, the ability to receive fourteen shillings plus an increment of 6 percent interest in “lawfull money” exactly one year later (on “the twenty fifth day of May AD 1776”), or, second, the ability to make payments to the treasury, presumably of taxes, at any time during that intervening year (but in an amount lacking interest accrual).

Since paper-based pecuniary media are relatively new to the majority of potential users, they often describe what they may be transacted for in considerable detail, even as they appear to assert (or, more often, elliptically to imply) that they are not themselves “money.” Although England did not permit its early American colonies to mint coins, transactional activities were mediated by a great variety of pecuniary media in eighteenth-century America, including Spanish milled dollars (the term English speakers gave to Spanish 8 reales, minted in Spain’s rich colonies in Mexico from Peruvian bullion) as well as various other European precious metal coins. Thus the document in figure 8 appears to assuage the one whose hands are holding it that even though, as paper, it is worthless scrip, it can nonetheless be converted into one or another of the more familiar pecuniary media. Since taxes of varied kinds are a perennial source of anxiety for many then (as they are now), and since taxes are paid in varied pecuniary media in different early American locales, the document in figure 9 reassures its reader that its bearer is entitled to coined money, and additionally, that it belongs to the class of pecuniary media “which will be received for taxes.” Since the notion that pieces of paper could be used to discharge tax burdens would have seemed improbable to many, the document’s explicit claim to be usable as such appears designed to persuade people to use it, and, of course, not only for taxes. When the Connecticut bill of 1735 (fig. 10,A) reassures its reader that it has a value of two shillings “equal to money,” it in effect implies that it is not itself “money,” though exchangeable for an equal measure of whatever is called “money” in that locale (as did similar pecuniary media in the other colonies). Not everyone was so persuaded:

The document in figure 10,B which is also from Connecticut (but from 1776 not 1735) is both like and unlike its precursor: It describes itself as a “bill” worth two shillings and sixpence of “lawful Money,” thus clarifying that it is not itself lawful money, but transactable for such, as does the bill of 1735. But the dual-origo indexical anchoring of the 1735 bill and the politico-religious cosmic polity it sustained are gone; time reference is now anchored solely to the Gregorian calendar, as is the promise that the bill’s bearer will acquire lawful money by “the first Day of January, A[nno] D[omini] 1781”; and the promise is guaranteed not by “his majesty” and his “General Court” (as in the 1735 bill), but “By order of assembly,” thus bringing its issuer, its user and its warrant together in 1776 into a cosmic polity of an altogether new and distinct kind.

Figure 8.

You can get gold or silver for me (Congress issue, 1779)

Figure 9.

You can pay taxes with me (Congress issue, 1785)

Figure 10.

A, I am not money (Connecticut issue, 1735); B, I am not money but you’ll get some on January 1, 1781 (Connecticut issue,1776).

As the Revolutionary War unfolded around such acts of issuance, acceptance, transaction and belonging, many types of pecuniary media directly sought forms of political or military allegiance among members of a divided nation, continuing to maintain a distinction between themselves (and what they stood for) and all the other already legally sanctioned and more familiar pecuniary media with which they contrasted themselves, promising nonetheless their own redeemability in some among these, whether in “eight dollars in continental Currency” (fig. 11,A) or in “one thousand pounds of nett inspected tobacco” (fig. 11,B), or in some other.

Figure 11.

A, I am no dollar but I support the troops (Georgia, 1777); B, You can get tobacco legal tender for me (Virginia, 1780).

Although the forms of pecuniary variation across locales were considerably more complex, the few examples I have chosen here do clarify some general issues: that activity routines involving pecuniary media were vastly differentiated across geosocial locales in eighteenth-century America; that such differences were demarcated and mediated by specific forms of discursive semiosis; that the activities such artifacts enabled included the ones they described; that such descriptions located their users in a variety of social category memberships, communities, political formations, and allegiances as preconditions on their capacity to engage in such activities at all; and that, when the noun money does occur in them (e.g., figs. 7,C and 10), their self-descriptions reflect facts of then current usage whereby the noun money properly describes a class of pecuniary media much narrower than the class to which such documents belong. It is not merely the case today, as I noted earlier, that the noun money denotes a small subset of the pecuniary media in use, it was the case in early America as well, as I show in the next section, and, indeed, is the case in every period of every known society, as I show later on.

Money Talk and Its Folk Taxonomies

It will come as no surprise that the forms of variably anchored and taxonomically diverse money talk that were produced and circulated through such documents would yield varieties of confusion in forms of money talk in everyday life too, as well as in varied published documents, a process that continued well into the next century. For instance,

In pronouncements such as these, the noun money may be glossed roughly as ‘precious metal coin,’ the noun debt as ‘paper bills.’ The self-descriptions and self-and-other positioning formulated by then extant pecuniary media are recycled here into the opinion that the denotata of such nouns are “natural antagonists,” a doctrine that episodically finds its way into varied descriptions of, and varied proposals for overcoming, the Manichean conflict between such forms of “fire and water,” and, in the light of such pronouncements, into proposals for how to underwrite colonial skirmishes and wars, including some against “primitive” peoples, as well as dreams of empire or revolutionary upheaval. By the late nineteenth century, it had become fashionable to use an entire cluster of nouns—barter, money, currency, and the like—to produce classifications of human conduct across stages of an imaginary “evolutionary” timescale, even though such taxonomies were based entirely on observations of then extant people (when they were based on any observations at all); and talk about those who were called “primitives” was deployed chiefly in efforts to identify what was distinctive about those who called them that. A painlessly brief sample looks like this:

This particular sample of this vintage of money talk is of interest here chiefly because Colonel Temple is using it to classify and characterize nonbullion currencies among “savages,” and Quiggin’s book on “primitive money” (fig. 2) begins with a discussion of Temple’s taxonomy and struggles against it, dismissing and refuting it in passages like the ones cited earlier, and producing, in turn, a far more elaborate (380 pages long) evolutionist taxonomy of her own. But let us return to the case at hand.

We have already seen that many of the script artifacts that function as pecuniary media in early America cite the statute or legislation that issues them. They contain inscriptions such as these: “according to an Act of Assembly of the State of Maryland, made the 8th Day of May, 1781” (fig. 12,A); “pursuant to a Law of the State [of New Jersey] … passed in the Year 1786” (fig. 12,B); “pursuant to an ordinance of the General Assembly [of South Carolina] passed the 8th day of July, 1779” (fig. 12,C); “According to a resolution of Congress passed at Philadelphia, February 17, 1776” (fig. 12,D). Similar citations occur in most of the preceding examples and in many that follow. This particular feature of the discourse genre is worth noting at this point, but will acquire a particular signficance later on, especially when we have seen sufficiently many samples of the genre so as to be able to identify variation within it, and thus to discern how variably distributed forms of indexical anchoring to forms of legitimacy produce surprisingly distinct fates and fortunes among those anchored by them.

Figure 12.

A, Assembly of Maryland, May 1781; B, Law of New Jersey, 1786; C, Assembly of South Carolina, July 1779; D, Resolution of Congress, February 1776. Source: Wikimedia Commons, US public domain.

Printers and Issuers

Meanwhile, almost anyone can print money at this time. The power to manufacture and mass produce paper currency is not yet centralized as the perogrative of a government or state agency. Many different types of pecuniary media are manufactured by private printers, whose personal names appear on these documents, such as “John Holt” (in fig. 13,A) or “Hall and Sellers” (in fig. 13,B) or “B. Franklin and D. Hall” (in fig. 13,C). The public use-user characteristics of these pecuniary media are specified in various ways, whether by specifying the domicile of issuance-use-and-user as “New York Currency” (fig. 13,A) or simply as “Philadelphia” (fig. 13,B), or by specifying the legitimacy of issuance by naming a legislative authority, whose warrant is itself grounded by dual calendric origo in an appropriate cosmographic scheme (fig. 13,C), or both, much as in previous cases. And yet these are merely the simplest and perhaps the least interesting cases.

Figure 13.

A, Printer: John Holt; B, Printers: Hall & Sellers; C, Printers: Franklin & Hall.

Many issuers and printers formulate themselves in figurements of (in)corporation. Businesses of every description and size are printing their own currency bills at this time. Many of these bills are used mainly in transactions with the businesses that issue them, while some acquire other uses too. Each such bill formulates a specific type of indexical selectivity for a propsective clientele of currency users. Thus if issuers like the “The New Hope Delaware Bridge Comp[an]y” (fig. 14,A) employ images of Lady Liberty and cherubim that suggest grandiose “hopes,” currency issuers like “The original Fringe and worsted yarn warehouse” (fig. 14,B) depict representative moments of manufacture itself—such as the harvesting of cotton and wool, and their weaving on looms—thus appearing to address the working class clientele for which the bill is designed, and to whom it proposes to pay 50 cents “in Goods” or in “Philad^a Bank notes.” s

Figure 14.

A, The New Hope Delaware Bridge Company (one dollar, Lambertville, NJ, 1840s). Source: Wikimedia Commons, US public domain. B, The original Fringe & worsted yarn warehouse (fifty cents, Philadelphia, 1814).

Some pecuniary media were minted and issued by banks, though the noun bank denotes several disparate things in early America:

By the early nineteenth century, “the monetary funds that [private] banks provided were commonly in the form of their own circulating notes, handed over the counter to the borrower” (viii–ix), and a growing number of private banks, which were sometimes simply households or families, were “organized without corporate charter” (11), often only for specific business needs, community clienteles, and periods of time. In states like New York and Michigan, where “laisser faire” doctrines of “free banking” linked banking to the view “that all are freely permitted to embark in it,” free banking “meant, in effect, an indefinite and unlimited number of banks” (573). Around the country, banks were “organized under at least as many distinct laws as there were states” and many banks “were little more than legal counterfeiting shops” (Unger Reference Unger1964, 17).

Although the number of banks did remain “indefinite,” in the sense of being largely unknown, it was not in practice “unlimited,” of course. As one estimator observes, “In this country the measure of value is committed to some two thousand banks,”Footnote ⁴ a figure most useful perhaps only in giving a rough idea of the scale and locale-specific diversity of what counted as “banks” by midcentury, at a time when their continued lack of effective regulation (in several senses) had become a topic of national debate.

The currency bills produced by such private “banks,” as in figure 15, are organized by principles of addressivity that are much the same for the businesses discussed earlier: they are indexically selective for specific social categories of clientele and associated participation frameworks of commercial interaction. The Mechanics Bank of Georgia (fig. 15,A) emblazoned inter alia with trains, steamships, and arm and hammer, reaches out to an emerging industrial working class. The Somerset and Worcester Savings Bank of Maryland (fig. 15,B), with images of a man petting his dog, a woman milking her cows, a little girl cuddling her puppies, presents a tableau of savings-based idyllic moments of prosperity for an agrarian household. The Piscataqua Exchange Bank of Portsmouth, New Hampshire (fig. 15,C), depicts an array of milkmaids, cows, and pails in an anonymous tranquil glen (while a galleon approaches the port, on the right hand side), while images of agrarian & mercantile endeavor are flanked by images of William Penn, Benjamin Franklin and George Washington.

Figure 15.

Private banks: A, Mechanics Bank (fifty dollars, Augusta, CA 1854); B, Somerset & Worcester Savings Bank (two dollars, Maryland, 1862); C, Piscataqua Exchange Bank (twenty dollars, Portsmouth, NH, 1800s, updated).

As the statutes that define “banks” vary across locales and periods of the nineteenth century, so do the banks themselves. For instance, the National Banking act of 1863 “provided that any group of five persons possessing a specified amount of capital could organize themselves into a banking corporation” (Unger Reference Unger1964, 18). The range, variety and sources of currency issue is also very large. At least “6,178 different types” of bank note are attested as issued by various private banks in New England alone, but it is likely that more than “12,000 different varieties had been issued,” and “that even this estimate is conservative” (Bowers Reference Bowers1979, 46).

And yet issuers of notes are even more varied than “banks” (under any definition), and those that use print technology and bill design reminiscent of paper currency in use today are merely one type of case. The scarcity of specie in the early nineteenth century had the result that many banks suspended specie payments for paper bills, which, in turn, “stripped the country of small change; not a sixpence, not a shilling, not a pistareen, was anywhere to be seen,” so that “cities, in their corporate capacity, printed thousands of dollars’ worth of cents, two-cent, and six-cent notes” (McMaster Reference McMaster1927, 297). Around 1810–15 many diverse social categories of private persons followed suit:

The practice of producing locale-business-and-community-specific script artifacts continued through much of the nineteenth century (see figs. 16 and 17), and whether the issuers of these bills used print or woodblock or handwriting, the script artifacts they issued faced challenges similar to those evident in the machine-printed bills (cf. figs. 14 and 15), namely, the problem of how to create social regularities of transactional and other social interaction with just some among the totality of persons who might encounter them as readers.

Figure 16.

Transactional equivalence classes, John Thompson, 1814

Figure 17.

Transactional equivalence classes, S. P. Cocke, 1861

Quite apart from sketching the social categories of persons that are selectively addressed within, and potentially recruited to, participation frameworks of issuer-user interactions, as discussed above, many pecuniary media specify sortal kinds of objects that appropriately belong to, or have a place within, such interactions, thereby differentiating these from all other entities in the universe, and often do so by linking specific pecuniary media to specific objects of transaction, and, indeed, to transactional equivalence classes that organize activities involving such objects, thus differentiating such activities from all others. In the case of a scrip issued in 1814 (fig. 16), the issuer, Mr. John Thompson, the proprietor of a store called the “Chest of Tea and Hogshead” at 130 North Water Street, Philadelphia, reassures its reader that it is worth “six-and-a-quarter cents,” an amount in which it is redeemable for two sortal kinds of entities in the universe, the “groceries” at his store, and “Philadelphia bank-notes.”

A similar interactional schematization of its own use characteristics is formulated by the scrip in figure 17 nearly half a century later. The issuer, a Mr. S. P. Cocke, declares that the thing in the bearer’s hands is worth five cents, that it can be transacted for two sortal kinds of entities, either for things like “flour, baled hay, oats, wood, &c.” at the store (where the et cetera clause covers whatever else Mr. Cocke may happen to have in his store at the time of transaction) or, alternatively, “for bankable funds” if the bearer comes to “my office” (and has one hundred such scrips).

Such script artifacts preserve, in conveniently inscribed form, those metasemiotic formulations of transactables and transactors on which their usage relies. But the presence of script is neither necessary nor sufficient. The most omnipresent form of discursive semiosis that enables the effective use of such pecuniary media is the audible speech that accompanies and informs the social interactions in which they play a part. The following case from Massachusetts, 1693, is recounted by a lady traveler, one Madame Knight, who describes four sortal kinds of pecuniary media then current in this locale—namely: (i) pay, (ii) money, (iii) pay as money, (iv) trusting—and the transactional equivalence relations customary among them, as well as the role of conversation in enabling transactions to occur at all:

Several issues are worth noting. First, “grain, pork and beef, &c.” can be eaten—separately and severally—or can be grouped together as a unitary sortal kind of pecuniary media termed pay. Second, these edibles can also serve as pay as money, a class of pecuniary media that are “one third cheaper.” Items of a third sortal kind, “pieces-of-eight, ryals, Boston or Bay shillings … silver coin … also wampum,” are grouped together as money. The fourth sortal kind consists of utterances—not any and all utterances, of course, just the ones that yield an “agree[ment] for the time”—which are said to facilitate transactions as trust or trusting. In her last three sentences, Madame Knight describes the typical form of the conversation that specifies which sortal kinds are relevant to any current interaction, and, once it takes place, enables transactions to occur.

Analogous equivalence classes have been reported in many societies that lack script. Paul Bohannan (Reference Bohannan1955, Reference Bohannan1959) identifies three classes into which transactables were sorted (allegedly more elaborately so in the nineteenth century) among the Tiv of Northern Nigeria, a population of about 800,000 people at the time of his research (the mid-twentieth century): items of class I—which includes chickens, goats, yams, corn, pepper, other local foodstuffs, household utensils, some tools, raw materials for tools—can appropriately be transacted most readily for each other; items of class II—which includes cattle, slaves, ritual offices, tugudu cloth, magic, medicine, metal bars—are also transacted for each other; and class III, special because it has only one member, women as spouses, is treated in rather distinct ways. Apart from other difficulties with Bohannan’s work (see Guyer Reference Guyer2004), the limitations of his thing-focal and exchange-centric approach (Agha Reference Agha2011) and inadequate attention to discursive data create insuperable difficulties: Bohannan reifies discursively mediated sortal kinds as inscrutable mental models (“ideas” about “things”), transforms contextually appropriate registers of transactional conduct into something called “spheres of exchange,” and, lacking any concept of the social domain segmentation of enregisterment, essentializes “The Tiv” by implying that all of them (all 800,000 of them?) shared the same mental models (“ideas”) of “exchange.” And although his account is often read as differentiating pre-colonial Nigeria from something called “the West,” such appraisals are not sufficiently well informed about what was happening in America around the same time, as noted above, nor what happens there today, as I show later on.

When Not to Do It

Competence in any register of conduct depends on knowing when its use is effective and appropriate, and when it is not. In the case of speech registers, elites frequently downshift when addressing nonelites (e.g., from pure Received Pronunciation to Cockney blends in Britain, Agha 2007a, 224–27); and nonelites, such as users of youth slang, link effective competence to knowing “when not to use it,” which they assess through co-textual construals of who they are interacting with and in what type of setting (Agha Reference Agha2015b: 311). Much the same is true of registers of money-token conduct everywhere. Not to know when to engage in money-token conduct, and with whom, is to be a misfit. For instance,

For Malay fishermen, transactional conduct is appropriate only in male-male/nonkin participation frameworks of social interaction. And such do’s and don’ts of money-token conduct differ across societies, differentiating locale-specific registers.

In print cultures, etiquette guides describe such do’s and don’ts for those unfamiliar with them. The following example, from Miss Leslie’s Behavior Book, shows that nineteenth-century Philadelphia ladies face cotextual constraints much like Malay fishermen from Langkawi (and people elsewhere), and like them, the ones they face are distinctive to their own social milieu:

Miss Leslie is describing the rules of etiquette for a “lady” in mid-nineteenth-century Philadelphia, where, in a changing milieu of class relations, young women (particularly those not born in the genteel urban locales where they come to live) must learn appropriate emblems of ladylike behavior. The etiquette guide tells its reader how to dress, travel and take a cruise, how to borrow things or accept a gift, how to raise children, how to behave in “society” and on “the street,” and much else besides. In the passage just cited, it describes how to raise children into adult norms of money-token conduct in order not to shame one’s own household.

Any register of “money-token conduct” (of whatever kind) invariably models criterial activities as phase-segments of a social interaction, and current interactants are either familiar with how to engage in (the otherwise unrelated) activities that must precede and follow it (like the ability to construe an interlocutor’s social role, or to perform appropriate roles in response) or are incompetent in it. In short, money-token conduct, wherever it occurs, whatever its forms, effectively occurs only as a phase segment of forms of conduct that have little to do with money.

The Voicing Structure of “Money” Talk

Many people produced money tokens at home in early America. For instance,

Although such practices were outlawed by legislatures, they remained commonplace in many communities all over colonial America. Here are examples from Virginia:

The direct contradiction between facts of Virginia coinage and Jefferson’s assertion in no way impugns his veracity. Rather, it reveals something far more interesting: All answers to the question “Are these coins?” have a voicing structure: “Yes,” according to those who issue or use them as coins, and “No,” according to a state that does not license their issuance. It is only when state-origo voicing has become a naturalized feature of common sense, as it has for many today, that appeals to voicing structure appear odd or remarkable.

The Virginia coins of 1773 and 1774, both struck in copper, are shown in figure 18,A. They pass muster as coins to visual inspection: They bear a bust of the king with the legend “Georgivs . III . Rex” on the obverse, and the legend “Virginia 1773” divided by a shield & crown on the reverse. The shield is emblazoned into four quarters, showing the conventional emblems of “1, England impaling Scotland; 2. France; 3, Ireland; 4, the Electoral dominions” (Crosby Reference Crosby1875, 339). A user, mindful of such text-internal metasemiotic formulations of its own characteristics, sees coins; the state, mindful of the absence of an entirely distinct type of metasemiotic framing, that no license to mint them has been issued, sees non-coins. Since the practice of minting coins at home remains common, and the use of such coins remains widespread and routine in many communities, items of this kind remain both coins and non-coins for much of the eighteenth and nineteenth centuries, though from different points of view.

Figure 18.

Are these coins? Yes, and No: A, Virginia coppers, 1773 and 1774; B, Connecticut coppers, 1737–39.

The varieties of both internal and external forms of metasemiotic framing differ vastly across the corpus of early American coins. A single example, the variation in coinage struck by a single person (allegedly a Mr. John Higley of Granby, Connecticut) between 1737 and 1739 may prove illustrative. Sketches of two of his coins are shown in figure 18,B. The first one, on the left, has three hammers, each with a crown, and the legend “I am [a] good copper. 1737” on one side, and, on the other, the image of a deer and the legend “value me as you please” and a Roman numeral “III” within scroll work below the deer, which suggests three cents as its value. The one on the right has a single broad axe and the legend “i cut my way through” on one side, and the deer and legend, “value me as you please” with the scroll containing “III” below it, on the obverse. A story about Mr. Higley has it that he mined the copper for these coins from a mine near his residence in Granby, and paid for his liquor at the public house with the coins he minted. This practice, which continued for a while, possibly with coins he had minted earlier that explicitly described themselves as worth “three cents” (not shown in fig. 18), eventually became subject to complaint, at which time “he presented coppers bearing the words, ‘Value me as you please’ ‘I am a good copper’” instead (Crosby Reference Crosby1875, 326–27). It is unclear if this story is true, but it is clearly part of the money talk or lore of the period.

In postcolonial times, trends of local currency manufacture continued unabated, and even grew. By the middle third of the nineteenth century, a profusion of locale-specific mints, coins, and coin inscriptions had emerged. Gold coins were being “minted by Templeton Reid, in Georgia, in 1830,” and, “In 1831, only one year after Reid began coining, a better known and more prolific mint was set up at Rutherfordton, North Carolina, by Christopher Bechtler … [who was noted for] … honesty in making his coins the same value as the coins of the United States.” And yet, soon afterward, “In California … private mints sprang up as much to supply the people with what they demanded as to fatten the coffers of their owners. … The various establishments rendered this service with varying degrees of honesty” (Barnard Reference Barnard1917, 617–20). Meanwhile, impeccably honest, the Mormon Church was minting its own coins: on a twenty dollar piece of 1849, one side bears the legend “holiness to the lord,” and on the other appear the letters “G. S. L. C. P. G.,” which stand for “Great Salt Lake City Pure Gold” (Barnard Reference Barnard1917, 624).

As we consider the range and variety of such coinage we can begin to see that along with their commercial functions, such metallic script artifacts also consitute a type of folk literature that has a complex internal voicing structure that partakes of genres like the political cartoon, the broadside and the pamphlet, albeit in miniaturized form. The following examples involve pecuniary media issued by common people in the 1830s, a time when President Andrew Jackson’s fiscal policies had created curency shortages:

These pecuniary media therefore function not only as “coins” in commercial transactions, and hence not only as “pecuniary” media, but also as devices that communicate the opinions of their issuers on the policies of the state and the politics of the day. Just as the pecuniary media in figure 14 are not just “bills” but also advertisements for businesses (and the ones in figure 15 advertisements for businesses that call themselves “banks”), the items now at issue are not just “coins” but also political pamphlets, which, although much smaller than conventional ones, have a much larger circulation.

An even more politically charged example (and a sharper contrast) may be seen in figure 19, which shows plates engraved by the Boston silversmith Paul Revere. The item in figure 19,A depicts “The Boston Massacre” of March 5, 1770, an incident in which British soldiers killed civilians who faced them as a mob, an event publicized by Paul Revere in the plate shown in figure 19,A to encourage rebellion against Britain. The plate carries the inscription “Engrav’d Printed & sold by Paul Revere Boston,” which formulates it as a commodity manufactured for sale, and not, in any sense, a money token. By contrast, in 1775 Paul Revere engraved and printed the “Bill” shown in figure 19,B, whose self-description (recto) formulates it as a canonical sample of the money-token genre discussed above. But the bill also depicts (verso) a man with a sword in one hand, the “Magna Carta” in the other, with the legend “Issued in defense of American Liberty” and the Latin inscription “Ense petit placidam, sub Libertate, Quietem” (By arms he seeks quiet peace under liberty), adopted as revolutionary motto by the Provincial Congress in 1775, all of which appear beneath its denominated value, two shillings. Since Britain’s Currency Acts of 1751 and 1764 prohibited colonists from issuing paper money (Allen Reference Allen2009, 96–98), it is plain that the item in figure 19,B is not money in 1775 from the standpoint of the Crown. At the same time, and from the standpoint of the colonists, it is both legal tender (recto) and a call to arms (verso).

Figure 19.

Plates engraved by Paul Revere: A, Is this money? No; B, Is this money? Yes, and No. Source: A, Wikimedia Commons, US public domain.

As we pay attention to the role of discursive semiosis in enabling the existence and use of money and its relatives, it becomes evident that most taxonomies of “money” (including the ones in the numismatic literature that I cite here) and all efforts to give histories of “money” that rely on such taxonomies, tend unproblematically to assume a state-origo voicing structure in the very labels they assign to pecuniary media (thus extending the state’s present-day sortal categories back into the past), tend also to overlook non-state-issued pecuniary media (or their hybrid voicing and non-pecuniary genre characteristics), and so tend, overall, to produce accounts that are predictably skewed through the “erasure” (Irvine and Gal Reference Irvine, Gal and Krokrity2000) of features of the very past they seek to analyze. But where do the intuitions that enable such erasure come from?

Legal Struggles

In eighteenth- and nineteenth-century America, various legislatures and governments attempted to regulate the production of—and the use of money as a term for describing—pecuniary media of the kind just described, and attempted also to regulate the rights of various social actors (individuals, communities, provinces, colonies, and other subsidiaries) in activities of issuance or use. We have seen that many bills in the preceding examples specify the statute or law that issues them, and hence the geopolitical boundaries of their legitimate use. These specifications indexically locate their issuers in legal and political struggles whose details these bills do not describe. Although the kinds of legal struggles are quite varied, two types of cases are fairly ubiquitous.

In one type of case, a state legislature imposes legal sanctions on its citizens. But such efforts often produced mixed results, at least in the short run. For instance,

However,

Although such legislative efforts were variably effective in regulating the forms that money-token conduct took in specific times and places, it is perfectly clear that by the late nineteenth or early twentieth century, the federal government had, through a sustained series of legal actions, managed effectively to regulate forms of currency issuance across the United States, and to establish itself as the sole issuer of a legitimate national currency.

A second type of case involves battles not between a government and its citizens but between a central government and its subsidiary governments. I offer two examples here, the first from the eighteenth-century colonial period:

Although the ensuing policies were sometimes quite effective, the issue of paper currency never ceased in colonial America, as we have seen. The second example is from the nineteenth century, from the concluding period of the American Civil War, circa 1865:

These measures (and others) led to the elimination of the great diversity of the locale-and-issuer-specific American currencies discussed above, and eventually to the federally regulated single national standard currency (and to the state-origo criteria on what counts as “money”) that we take so much for granted today. And yet the production and use of other kinds of pecuniary media has never ceased, as we shall presently see.

Performative Locutions

We have been considering a variety of forms of deixis and indexicality in the preceding examples, the manner in which the values of indexical variables are discursively specified in these cases, and the ways in which distinct specifications are laminated together to yield composite effects. In the case of contemporary currencies, to which we now turn, the specific type of indexical array that Austin termed a “performative” acquires a distinctive genred form. I have argued elsewhere (Agha Reference Agha2007a, 55–64) that an Austinian performative is a very specific type of indexically composite sign, in which a number of distinct kinds of semiotic partials—including specific values of speech-event deictics, verba dicendi, proper names and titles, and forms of enregisterment—come together to yield composite indexical effects, and that such effects are defeasible in as many ways as there are categorial dimensions that comprise them. (I return to these issues in subsequent sections.) Here is Austin’s description of the basic issues:

Contrasting the written and spoken alternatives, Austin adds, by way of example, that the spoken form, “I, John Jones, warn you that the bull is dangerous” is performatively equivalent to the written form, “This bull is dangerous. (Signed) John Jones” (62).

It is readily seen (fig. 20) that the British pound sterling employs the “utterable” version of the performative locution, and that the United States dollar relies on the inscriptional form. In the case of the British five pound note, we see at the top left (below “Bank of England”) the performative formula “I promise to pay the bearer on demand the sum of five pounds,” an utterable version personalized below by the picture of the queen, who is now understood as the referent of “I.” On the US twenty dollar bill, the statement at the bottom left carries the inscriptional form of the performative formula “This note is legal tender for all debts, public and private. (Signed) Treasurer of the United States.” If all “felicity conditions” are met (e.g., the bill is authentic, the signatory is indeed the treasurer, and so on), the performative formula appears to constitute the social fact that it denotes, namely that it indeed “is legal tender for all debts, public or private” (but see below). Additional signatures occur on both bills, which have complementary or additional legislative functions, but are visually separated from the performative locutions that occur on them.

Figure 20.

Performative locutions, British pound and United States dollar. Source: B, Wikimedia Commons, US public domain.

We have already seen that American currency did not always take its present-day form, the form taken by the Federal Reserve note in figure 20. Earlier forms of American currency contained performative locutions too, but these were more deictically selective than present-day ones, and correspondingly restricted the activities to be carried out through them. A comparison with currency notes from the more recent past will make the distinction clear.

Whereas the dollar note in figure 20 is deictically nonselective to a maximal degree in describing its uses as “legal tender” through a universal quantifier, “for all debts, public and private” and thereby underspecifies the limits of its own uses to a high degree, the first United States federal note (1862), in figure 21,A—with a picture of Salmon P. Chase, then treasurer of the United States—has a more restrictive self-description because it cites limitations on its own use (verso) : “This note is legal tender for all debts public and private except duties on imports and interest on the public debt” (emphasis added). It tells its bearer that it can’t be used for customs duties, nor for interest payments on the national debt. However, the US Silver Dollar Certificate (1886), depicted in figure 21,B—which bears a picture of Martha Washington—fills in these gaps, and describes itself as receivable for “customs, taxes, and all public debts.” So each kind of dollar bill is restricted, by criteria of transactional equivalence class (and others), to a distinctive kind of use, thereby segmenting the kinds of sociocentrically coordinated, interpersonally recurrent and cotextually appropriate activity routines performable by bearers through each.

Figure 21.

Transactional equivalence classes, late nineteenth-century United States: A, federal note, 1862; B, silver dollar certificate, 1886. Source: Creative Commons license 4.0, National Numismatic Collection, National Museum of American History.

This genre of self-formulation is nowadays found on many (but not all) national currency bills, and, even when it occurs, exhibits some variation. For instance, all three bills in figure 22 use some version of the inscriptional form of the performative formula, whether by using the full sentence locution, as in figure 22,A and C; or just the nominal predicate phrases, as in figure 22,B; and in all three cases, the performative locution is followed by the signature of a requisite official, whether a governor (New Zealand), or a treasury secretary and a governor (Australia, Papua New Guinea). The serial numbers in these (and other) bills specify the batch in which they are issued, and are thus indexical of time of issuance (though this is opaque to the general public). Some explicitly specify locale of use: The item in figure 22,B describes itself as “legal tender throughout Papua New Guinea,” the item in figure 22,C as “legal tender throughout Australia and its territories,” both using proper name deixis to delimit locales—whether formulated as country locales, or as country locales plus dominions—as geopolitical boundaries of appropriate use. Thus, although these documents share common genre characteristics of national currencies, they differ among themselves in deictic selectivity across one or more dimension of use-user-place formulation, thus contrastively delimiting the appropriateness conditions under which transactional conduct can occur through them.

Figure 22.

Deictic selectivity, geopolitical locale of use: A, New Zealand dollar; B, Papua New Guinea kina; C, Australian dollar.

Describing Your Own Felicity Conditions

I noted earlier that any performative locution is defeasible in as many ways as there are categorial dimensions that formulate its own effectiveness. It will be obvious that any attempt to use the Kina note, figure 22,B—which formulates itself as “legal tender throughout Papua New Guinea”—to try to buy chewing gum in France or Switzerland, say, is an infelicitous use of it, or fails to meet the felicity conditions it specifies for its own use, and is unlikely to succeed. Such cases of ineffective infelicity can be multiplied without limit.

The most pervasive kind of potentially effective defeasibility that any paper currency faces is usually expressed in worries about counterfeit bills. Early American notes call attention to this particular felicity condition through genre characteristics that present-day notes lack.

The examples in figure 23 are all authentic notes, issued by various communities and printers as described on the notes themselves. They also carry inscriptions that flag worries about authenticity: The example in figure 23,A says “death to Counterfeit” both at the top and the bottom of the bill. The example in figure 23,B says “Tis death to counterfeit” at the bottom. The example in figure 23,C, “To counterfeit is death” at the top.

Figure 23.

Describing your own felicity conditions: genuine bills

The genre of flagging the felicity condition of a bill’s authenticity is so well known at this time that even counterfeit bills use it, as in figure 24. The first example (fig. 24,A) is counterfeit, but says that it is issued by “a resolution of CONGRESS passed at PHILADELPHIA, February 17, 1776”. The second one (fig. 24,B) is also counterfeit, but says that it is issued by the “Colony of new-york,” and includes all the genre-specific features of phraseology, seal and signature that we have seen in authentic bills above, and, even though it is counterfeit, it adds “‘Tis Death to Counterfeit” below the seal. When counterfeit bills deploy such features, they are canceling, or rendering defeasible, the state’s own performative efforts, while appearing to uphold them.

Figure 24.

Describing your own felicity conditions: counterfeit bills

Emergence of a Standard Register of Money-Token Conduct

We have considered a number of kinds of pecuniary media in the above discussion and have noted a number of dimensions of variation across them. In every case, items function as specific types of pecuniary media only for those acquainted with discursively mediated metasemiotic formulations of their characteristics as objects or entities of some specific kind, of the kinds of activities possible through them, of the time, place and social category membership of their users, and of the communities, allegiances, or political formations to which they belong.

For those that contain no inscriptions, a variety of artifact-external discursively mediated metasemiotic formulations of their thing-use-user characteristics (whether formulated by law, by custom, or by appeal to such in conversation; see the discussion of fig. 3) serve as preconditions on the symmetric grasp by current interactants of their usability as money tokens. Even for the peculiar case of script artifacts, where some features of these formulations are contained on the artifact itself, additional metasemiotic formulations of their thing-use-user characteristics always play a part too, as we have seen, so that forms of discursive semiosis that are “off the page” readily supervene upon ones that are on it. Although the range and variety of issues discussed above is quite complex, and so cannot be summarized here without distortion, it will be useful now to consider a few selected features of these artifacts in order to see how changing object formulations of their thing-use-user characteristics have altered conceptions of what the folk term money supposedly describes, and what such views obscure about the role of pecuniary media in social life.

Table 2 highlights selected features of a few of the currencies used in America over the last three centuries, separating out some of the dimensions specified within them—here listed in separate rows—through which pecuniary media are given distinct types of object formulations. Row a indicates differing formulations of what these objects are. When comestibles are decreed to be pecuniary media, we find various terms for characterizing what they are, so that corn or tobacco are legal tender (in different use locales) but are not paper money, “pig, grain, beef, etc.” are grouped together as pay or as pay as money in some locales, but none are money, a term then reserved for coinage. When coins are minted at home they are noncoins for the State. A similar voicing structure influences the self-descriptions of the earliest State-issued paper media, which call themselves notes or bills (figs. 7,A, 7,C, 10,A, etc.), because they are issues of a colony (and thus of subsidiaries of a larger empire), and England has issued the colonies no right to issue money at this time, retaining the right to do so for itself. And when they are issued by private businesses or individuals (figs. 14–18), they often avoid all such labels, but are called tokens by those who look back at them with fully naturalized nation-state-origo intuitions, or are entirely overlooked. And finally, today, we use the term money for “State Issued Money Tokens” (figs. 20, 22), or SIMTs, but not for all pecuniary media that organize contemporary lives (hedge funds, derivatives, etc.; and see following sections).

Table 2.

American currencies: Decreasing deictic selectivity of object formulations

Row b indicates Where items are used. Some of the privately issued notes (figs. 16–17) are used only at stores that issue them, and only by a small subset of the population, and so registers of conduct that treat them as money tokens have a small social domain, and locales of use are indexically differentiated as one store vs. other stores. Others explicitly state that they are designed for use all over a specific colony or province or State of the Union (figs. 7, 10,A, 12,B, 12,C) and are thus in principle usable in any of the stores or businesses within such political boundaries, though not necessarily outside them. In later versions of the genre (figs. 20, 22), a sovereign nation-state becomes the relevant political boundary for acts of everyday purchase by its citizens, although such pecuniary media serve other uses across such boundaries too, as when currencies of another country are used as media of investment or trade by businesses or nation-states, most typically as aggregated ledger entries than as individual paper bills.

The issue of Who the one using it happens to be, row c, amounts to issues of role inhabitance and of persona manifest at the moment of use, as is the case for all performed arrays of signs that function as emblems (Agha Reference Agha2007a, chap. 5). Someone who uses the item in figure 16, for instance, is thereby formulated as a customer of the “Chest of Tea and Hogshead” at the moment of transaction, an indexical positioning that differentiates its user from other customers at other stores, such as those who frequent “the original Fringe and worsted yarn warehouse” (fig. 14,A) in the same year and city. Formulations of role inhabitance may be more generic or inclusive. The act of using the 1735 bill in figure 10,A formulates its user as a resident of Connecticut (and hence also a subject of “his majesty” in England). Since the text-internal self-descriptions of these bills are readily supervened by forms of discursive semiosis that are “off the page,” such as regional trade alliances or royal decrees, the emblematic positioning of their users readily gets incorporated into more inclusive forms of belonging, or, in responses to attempts at imposing these, readily fractionate into more fiercely locale-specific forms of loyalty:

For present-day national currencies, sociocentrically recurrent usage formulates their users as “citizens of a country,” as a text-default of construal (Agha Reference Agha2007a, 39–64), though nondefault construals are readily formulated through forms of co-textual semiosis in specific interactions.

As for issuance By Whom, row d, we have seen that many pecuniary media are issued by a miscellany of “regular folks” (whether by merchants, stage owners, tavern keepers or ferrymen, or by Mr. S. P. Cocke, Mr. Thompson, Mr. Higley, or by countless others) or by businesses such as the The New Hope Delaware Bridge Company, which we would call “a firm” today, or by a miscellany of organizations that once called themselves “banks,” or by various cities, colonies or provinces, where each issuer defines itself against other social categories of issuer in attempting to anchor its recurrent user in specific participation frameworks of community belonging, customer loyalty, political allegiance, and others. Only after a prolonged series of struggles spread out over historical time—of store against store, firm against firm, colony against colony, the state against its citizens, central governments against subsidiary ones, and so on—do we find the emergence of the historically peculiar (but now dominant) case that is so familiar today, that of a federal government having secured the right to be the sole issuer of a particular kind of SIMT, a national currency, even though any such government continues to struggle against other governments in matters of international trade, or against the effects of pecuniary media that are produced by its citizens but are outside its own control (such as derivatives and hedge funds), which yield periodic crises of national governance and international trade.

Finally, considering the variables in rows e, f, and g, we find that earlier pecuniary media specify their own term of use, or the legislation or statute that issues them, or the specific uses to which they can be put, but contemporary national currency bills—such as the US dollar (fig. 20)—no longer do so. Each of these features of modern SIMTs readily yield specific construals, or default uptake formulations, on the part of users: The absence of any calendric specification of term of use readily implies that the bill is sempiternal, always and forever able to be pressed into use with the value denominated. The absence of any specific statute or warrant suggests its timeless and placeless authority or autonomy. And the absence of specification of what, in particular, it may be transactable-for supports the ubiquitous but absurd folk-theory that nowadays “money buys everything.”

Such forms of money talk—which we might caption as always-autonomous-everything views of money—readily emerge as nomic construals of the deictically nonselective texts (Agha Reference Agha2007a, 42–45) that constitute paper SIMTs today and are readily recycled into countless genres of print publication and everyday talk in which users come together as participants. Yet the seemingly settled form that the genre now takes is a relatively recent achievement. It is as much a signature of the victory of the State against its subsidiaries and citizens as it is a promise of a State-centric “always-autonomous-everything” future to come. And insofar as it is a genre of metapragmatic discourse—one which appears to model forms of conduct—the increasing nonselectivity of the genre appears to calibrate activities of an apparently unbounded sort, and of actors of an apparently undifferentiated sort, increasingly to each other. And so a widely enregistered genre of money talk and conduct becomes a gaze-narrowing device.

Much as in a standard language community, where the elevation of one speech register to the status of a national standard, creates a baseline of metapragmatic evaluation, whereby all other speech registers appear somehow abnormal, defective, parochial, and so on (Agha Reference Agha2015b), the emergence of nationally centralized registers of money talk and conduct, where a nation-state becomes at once its animator-author-issuer and arbiter-judge-enforcer, has the consequence that many users of SIMTs readily become axiologically grounded in—and treat as their origo of reckoning—a scheme of evaluation that conflate SIMTs with money itself, not only because they call it that, but also because, insofar as they come equipped with “always-autonomous-everything” outlooks and metrics, other pecuniary media appear not to be “real” money at all, since they appear not to satisfy such metrics; and since the metasemiotic bases of such outlooks and metrics are highly non-transparent to discussion or debate, this fashion of speaking readily persists through time as a gaze-narrowing device, often surviving each business cycle and periodic crash, each period of inflation and of the incapacity to self-provision, sometimes making us moderns at home in a regime of modern money, a phrase which does not, of course, denote what money is today, but, instead, denotes what we moderns have come to call money.

Right around the corner of what is visible at all lie pecuniary media of other kinds, which are not SIMTs at all, but have entirely different issuer-issuance and use-user characteristics. But most people who have a firm axiological grounding in evaluative frameworks like the one just described are already equipped with naturalized diagrams of sortal kinds, so that when they are presented with samples of such pecuniary media, they are likely to produce evaluations of the kind that forms the subject header of the next section.

It’s Not Real Money

The first case I’d like to consider is the now ubiquitous case of digital game tokens. There are so many varieties of these, and there is so much to be said about each, that my discussion here is, of necessity, very brief.

Figure 25 depicts images of digital game money,Footnote ⁵ or digital game tokens as they are properly called in modern money genres of money talk and in associated taxonomies of sortal kinds. This Korean game, Dragon Heroes, has six pecuniary media, wings, gold, rubies, keys, medals and emeralds, such terms comprising a game-internal lexical register for denoting criterial pixellations (and not such entities in the universe as their ordinary English homonyms denote). Among them, keys, medals, and emeralds are somewhat subsidiary and have limited uses. Wings are used for the chance to play the game (one per game) and are automatically issued by the server every ten minutes. Gold is earned in-game by completing quests or missions. Rubies may be earned in-game, but since doing so is difficult, they can also be purchased through national currencies or SIMTs, and thus episodically become part of transactions that are indexically anchored to the world of “real money” and to the fates and fortunes of its users.

Figure 25.

Digital game money: Dragon Heroes

In the case of MMORPGs, we see more complex forms of money-token conduct because these games have “societies” embedded within them. In contrast to Dragon Heroes, which has relatively simple transactional roles (the game company is “seller/provider,” the players are “buyers/gamers”), these MMORPGs have more differentiated role categories of users, including players, intermediaries, brokers, and even part-timers, who are hired to play games and to participate in both in-game and out-of-game transactions. For example, the Korean MMORPG game Lineage (fig. 26), whose users are often middle-aged men, is notorious because of lawsuits regarding SIMT transactions involving game items—for example, when rare game items, worth millions of dollars, are traded with “real money.” It is sometimes said that even Korean gangsters intervene in these deals, though whether or how often they do so is largely unknown.

Figure 26.

Lineage

I describe these two cases here not to offer any deep or illuminating insights about them but to call attention to several issues of far more general interest. First, the number and variety of non-SIMT pecuniary media that pervade forms of social life today are very large, a fact readily obscured by idioms of modern “money”-talk and attendant gaze-narrowing devices. Although online game tokens are a rapidly growing (hence handy) example, countless offline cases can also be found in communities around the world. Second, money-tokens are never merely tokens of money: we have already seen that the “types” of which money-tokens are “tokens” extend far beyond “money,” that such artifacts are also emblematic of the place, time, social kinds and communities to which users and issuers belong, of the participation frameworks in which they come together as interactants, and of variably complex social relations, whether among themselves, or between them and others. Third, for every kind of pecuniary media we have considered in the preceding pages —that is, whatever the artifactual characteristics of such pecuniary media may be (whether made by mollusks or people, whether made of resin or metal or paper or sound or pixels), and whatever label they may be given from specific positional points of view (whether called “(not) money” or “(not) coins” or “trusting” or “pay as money” or “real” or “counterfeit” or something else)—all cases comprise discursively mediated systems of social indexicals, where metasemiotic frameworks of diverse kinds (whether their fragments be inscribed on them, or not) formulate their object signs as having specific characteristics as things appropriate to specific activities involving specific social categories of users. Lest anyone suppose that all this was once true but no longer is, my final example concerns crypto-currencies, of which there are many varieties today, and I consider just one.

Bitcoin

Bitcoin formulates itself as a fiscal or monetary alternative to SIMTs. It is not issued by any government. It is not grounded in any central authority. In order to describe bitcoin, even in the minimal way that I propose to do here, it is necessary to describe a lexical register and an algorithmic protocol. Let us take a small step in that direction (table 3).

Table 3.

Bitcoin: Fragments of a Lexical Register

What is a digital money ecosystem? It’s an infrastructure created by algorithms and technologies which enables people to store and transmit bitcoin information-partials from node to node in a digital network. Bitcoins are a virtual currency. There are no physical coins or even digital coins as such. The coins are, in effect, implied by messages that effectuate transactions, which are then verified in a ledger. A bitcoin lacks the identity conditions of pieces of metal and script: a bitcoin only exists as a money token when criterial forms of digital semiosis occur, after which, it exists as a transaction record.

What’s a Bitcoin owner? Bitcoin users own keys, which allow owners to unlock bitcoin-information and transfer it to others. These keys are stored in a digital wallet on the owner’s computer. The wallet doesn’t store bitcoins. It stores keys. You can start mining bitcoin by downloading a bitcoin protocol stack (open source). Mining is a term used to describe the activity of finding solutions to certain mathematical problems by using your computer processor’s computational power. You don’t need to know math. Your stack and processor solve the problems. Using these devices, someone on the network is able, on average, to solve the problem every ten minutes; this is called “mining a bitcoin” on the network. The network is a peer-to-peer system. There’s no central bank and no other central authority that issues or regulates currency. The algorithm does all this in a transparent and effective way.

The bitcoin protocol reduces the question of money-token transactional conduct to two problems: (1) Is the money authentic and not counterfeit? (The trust problem); (2) Can no-one else claim that the money belongs to them and not me (The double spend problem). The bitcoin protocol solves these problems through a secure public transaction ledger, called a blockchain and a distributed computational system called a proof-of-work algorithm, which arrives at a mathematical consensus about the state of transactions over the entire network every ten minutes. The mathematical principles underlying it are entirely transparent, though opaque to nonmathematicians.

Today Bitcoin is used for almost every type of commercial transaction imaginable, from everyday retail transactions at coffee shops, to high-end retail transactions in the art world, to charitable donations, international import/export, and so on. And yet although the system is used for many things, it is not used by the majority of people who use money tokens in the world today. It’s new. It’s unfamiliar. It’s confusing. And it may fail one day. Or it may not. Many people have never heard of it. Many use it quite a lot. Some use it for illegal activities.

What does a bitcoin transaction look like? A user, let’s call her Jill, downloads software, and gets an address, which looks like this: 1efkl3SHAnnprsvdoQmQwwGnpXGTrbG3m5DY. It’s a unique identifier, the equivalent of a personal proper name, her bitcoin name; and when used in transactions, works like a signature. It’s her only identifier on the network, or her address on the network. The address is stored in her digital wallet. When she gets an address, Jill doesn’t have any bitcoin yet. But her friend Adam does. So she gives him fifty dollars and her bitcoin address. He checks the exchange rate with US dollars. Let’s say it’s 100 dollars per bitcoin at the time. So he uses her address to transfers 0.5 bitcoin to her wallet from his own. This transaction is recorded on the network so that everyone can see it. At first it’s “unconfirmed.” Once a bitcoin miner includes it in a block of transactions—which happens every ten minutes or so—it is recorded on the blockchain as “confirmed.” The bitcoin, along with her ownership of it, exists as a social fact. She can spend it now.

The blockchain has multiple copies on multiple servers, so you can’t fake it. It’s also computationally secure. It uses public key cryptographic methods. This type of cryptography relies on certain mathematical functions, like prime number exponentiation and elliptic curve multiplication (bitcoin uses the latter), which are mathematically irreversible in the sense that it is easy to calculate from inputs to outputs, but not the other way. You cannot calculate the inputs by starting from the outputs.

I’ve just given a very brief sketch of the discursive architecture of bitcoin. Most of us don’t fully understand what it’s all about. Bitcoin is therefore a good example of the larger point discussed earlier: people who engage in money-token conduct—in whatever society—tend to be unfamiliar with most registers of money-token conduct in their own societies. You and I are actually like the folks I quoted from the Banker’s Magazine of 1858. Most of us don’t really understand it, and don’t call it money.

Yet conjectures about “most of us” and what they do are distractions. Table 4 describes the ways in which bitcoin usage has expanded in recent years: Mining revenue per day became three and a half thousand times larger during 2010–14, transaction volume per day over thirteen thousand times larger, and market capitalization over nineteen thousand times larger, even if the number of bitcoin in circulation has increased less than fourfold.

Table 4.

Bitcoin’s expansion, 2010–14

Source. Wolfson Reference Wolfson2015, 203.

These trends permit a glimpse of the growth of Bitcoin-mediated money-token conduct during this period, of an increase in the sheer number of users and transactions, and thus of expansion in the social domain of enregisterment of competent use (though these figures convey nothing about the social-demographic attributes of new users). To those outside this social domain, bitcoin is not “real money,” perhaps because they are unfamiliar with the metasemiotic frameworks that enable its use, or wary that no nation-state backs denominational values, features that make it seem like a sample of “slant stuff” to them (table 1), something that doesn’t quite belong to the class of “normal” pecuniary media they know best. A colleague of mine notes that her father uses bitcoin, but she herself doesn’t because she doesn’t “trust” it, she says. In this respect, she is much like the mid-nineteenth-century Americans discussed earlier, who used similar mental state predicates to describe paper currencies in an age when precious metal coin was called money; and verba sentiendi like “trust” may well be the most common kinds of metapragmatic verbs through which money-token users look anxiously across the boundaries of the register range that happens to be theirs.

There are other reasons too. Newspaper coverage of recent events involving bitcoin suggest that bitcoin is known by many today as associated with ups and downs in reputation and in the types of metasemiotic evaluations it receives by authoritative institutions around the world. Table 5 lists a series of recent events involving bitcoin that were widely reported in the public press. Using the informal terms introduced in table 1 earlier on in this article, namely, the contrast between bold stuff versus slant stuff—which are typographic eponyms for judgments about pecuniary media that are viewed as normal for some “us” versus those that are not—I suggest that many of these news stories tilt the public reputation of bitcoin more in the direction of slant stuff (e.g., People’s Bank of China states bitcoin is not “legal tender”), while others tilt its public reputation the other way (e.g., Bank of America report finds Bitcoin viable in money transfer market). I wish to suggest, in other words, that the battle between the bold stuff and the slant stuff is often a battle between publicly widespread metasemiotic discourses about pecuniary media—which organize default forms of identifiabiliy, characterizability, and symmetric grasp that people may have of them and which, through differences in participation frameworks and semiotic chain networks of discursive encounter, produce sociologically asymmetric denotational stereotypes of the very nouns that denote them (Agha Reference Agha2007a, 103–32)—and that the battle is potentially always going on whether or not these media happen to be SIMTs (but not always for all types of pecuniary media, nor for every public), even if it is the case that, for some among them, forms of trust and confidence prevail sufficiently well in some locales and appropriate axiological frameworks are sufficiently well in place that, for persons who live in these locales, their favorite bold stuff may indeed appear to have something like the “always-autonomous-everything” formulations discussed earlier, at least for a while.

Table 5.

The struggle between the slant stuff and the bold stuff never stops!

Source. Wolfson Reference Wolfson2015, 209