Debt by David Graeber
127. Iliad 9:342–44.
128. Evans-Pritchard 1948:36; cf., Sahlins 1981. For a good example of identification of kings and slaves, Feeley-Harnik 1982. Obviously, everyone is well aware that kings do have families, friends, lovers, etc—the point is that this is always seen as something of a problem, since he should be king to all his subjects equally.
129. Regarding the influence of Roman law on the liberal tradition, it is fascinating to note that the very earliest author we have on record who laid out something like Smith’s model, where money, and ultimately coinage, is invented as an aid to commerce, was another Roman jurist, Paulus: Digest 18.1.1.
130. But it has by no means been eliminated. (If anyone is inclined to doubt this, I recommend they take a stroll through their neighborhood ignoring all property rights, and see just how long it takes for the weapons to come out.)
Chapter Eight
1. “Debt, n. An ingenious substitute for the chain and whip of the slavedriver,” wrote the notorious cynic Ambrose Bierce (The Devil’s Dictionary, 1911:49). Certainly for those Thai women who appeared at Neil Bush’s door, the difference between having been sold by one’s parents, and working off one’s parents’ debt contract, was as much a technicality as it would have been two thousand years ago.
2. One of the few authors I know who’s confronted the question head-on is Pierre Dockés (1979), who makes a convincing statement that it has to do with the power of the state: at least, slavery as an institution was briefly revived under the Carolingian empire and then vanished again afterward. It is certainly interesting that since the nineteenth century at least, the “transition from feudalism to capitalism” has become our historical paradigm for epochal social change, and no one much addresses the transition from ancient slavery to feudalism, even though there is reason to believe that whatever is happening now may much more closely resemble it.
3. Robin Blackburn makes this argument quite convincingly in The Making of New World Slavery (1997). There were some exceptions, notably the Italian city-states. The story is of course more complicated than I’m representing it: one reason for the hostility was that during much of the Middle Ages, Europeans were largely victims of slave-raiders rather than their beneficiaries, with many captives marketed in North Africa and the Middle East.
4. The Aegean coins were stamped; the Indian, punched; and the Chinese, cast. This suggests that we are not talking about diffusion here. Speaking of Indian coins, for instance, one historian remarks: “If there is one thing that seems clear from a punch-marked coin, it is that the person who thought it up had never seen a Greek coin—or if he had seen one, it had not impressed him. The punch-marked coin is made by an entirely different metallurgical process” (Schaps 2006:9).
5. Pruessner (1928) was perhaps the first to point this out.
6. They appear to have been widely used by Old Assyrian merchants operating in Anatolia (Veenhof 1997).
7. Powell (1978, 1979, 1999:14–18) provides an excellent assessment of the evidence, emphasizing that Babylonians did not produce scales accurate enough to measure the tiny amounts of silver they would have had to use to make ordinary household purchases like fried fish or cords of firewood in cash. He concludes that silver was largely used in transactions between merchants. Market vendors therefore presumably acted as they do in small-scale markets in Africa and Central Asia, today, building up lists of trustworthy clients to whom they could extend credit over time (e.g., Hart 1999:201, Nazpary 2001).
8. Hudson 2002:21–23, who hypothesizes that the time element was important as merchants would presumably otherwise delay to employ the funds as long as possible. See Renger 1984, 1994; Meiroop 2005.
9. I’m referring here to Qirad and Mudaraba arrangements, similar to the ancient and Medieval Mediterranean Commenda (Udovitch 1970, Ray 1997).
10. Herodotus 1.138.
11. Herodotus 3.102–5.
12. Mieroop 2002:63, 2005:29. He notes that Enmetena’s total grain income in any one year was roughly 37 million liters, making the sum he claims to be owed more than one thousand times his own palace’s annual revenue.
13. Lambert 1971; Lemche 1979:16.
14. Hudson 1993 provides the most detailed overview of this literature.
15. Hudson 1993:20.
16. Grierson 1977:17, citing Cerny 1954:907.
17. Bleiberg 2002
18. One authority states categorically: “I do not know of debt-annulment decrees issued by any Pharaoh” (Jasnow 2001:42), and adds that there is no evidence for debt-bondage until the very late Demotic period. This is the same period when Greek sources begin to speak of both.
19. VerSteeg 2002:199; see Lorton 1977:42–44.
20. This in certain ways resembles the legal loopholes created in both the Medieval Christian and Islamic worlds, where interest was formally banned: see chapter 10 below.
21. Diodorus Siculus 1.79. See Westermann 1955:50–51 for a comparison of Greek and Egyptian sources on the subject.
22. The history of the dissemination of interest-bearing debt is only beginning to be reconstructed. It does not yet appear in Ebla (c. 2500 bc), in Old or Middle Kingdom Egypt, or in Mycenaean Greece, but it eventually becomes common in the Levant in the late Bronze Age, and also in Hittite Anatolia. As we’ll see, it came quite late to Classical Greece, and even later to places like Germany.
23. In Chinese historiography, in fact, this whole epoch is known as “the feudal period.”)
24. The Guanzi, cited in Schaps 2006:20.
25. Yung-Ti (2006) has recently argued that they weren’t, though we wouldn’t really know. Thierry (1992:39–41) simply assumes they were, providing much evidence of their use both as units of account and means of payment, but none of their use for buying and selling.
26. At any rate, cowries were definitely being used as the equivalent of coins in later periods, and the government periodically either suppressed their use or reintroduced them (Quiggin 1949, Swann 1950, Thierry 1992:39–41, Peng 1994.) Cowrie money survived, alongside tally sticks, as a common form of currency in Yunnan province in the far south until relatively recent times (B. Yang 2002), and detailed studies exist, but—as far as I can tell—only in Chinese.
27. Scheidel 2004:5.
28. Kan 1978:92, Martzloff 2002:178. I note in passing that a study of the Inca khipu system itself would itself be quite fascinating in this regard; the strings were used to record both obligations we would consider financial, and others we would consider ritual, since as in so many Eurasian languages, the words “debt” and “sin” were the same in Quechua as well (Quitter & Urton 2002:270).
29. L. Yang (1971:5) finds the first reliable literary reference to loans at interest in the fourth century bc. Peng (1994:98–101) notes that the earliest surviving records (the oracle bones and inscriptions) do not mention loans, but there’s no reason they would; he also assembles most of the available literary references, finds many references to loans in early periods, and concludes that there’s no way to know whether to take them seriously. By the Warring States period, however, there is abundant evidence for local usurers, and all the usual abuses.
30. Yan tie lun I 2/4b2–6, in Gale (1967):12.
31. Guanzi (73 12), Rickett (1998:397)
32. So around 100 bc, “when flood and drought come upon them … those who have grain sell at half value, while those who have not borrow at exorbitant usury. Then paternal acres change hands; sons and grandsons are sold to pay debts; merchants make vast profits, and even petty tradesmen set up business and realize unheard of gains” (in Duyvendak 1928:32). Loans at interest are first documented in the fourth century bc in China but may have existed before that (Yang 1971:5). For a parallel case of child-selling for debt in early India, Rhys Davids 1922:218.
Chapter Nine
1. Jaspers 1949.
2. Parkes 1959:71.
3. Or, if one must be even more precise, we should probably end it in 632 ad, with the death of the Prophet.
4. Ob
5. The date used to be set much earlier, at 650 or even 700 bc, but recent archaeology has called this into question. Lydian coins still seem to be the earliest, though, as most of the others have been seem to be the earliest though.
6. Prakash & Singh 1968, Dhavalikar 1974, Kosambi 1981, Gupta & Hardaker 1995. The latest accepted dates for the appearance of coinage in India, based on radiocarbon analysis, is circa 400 bc (Erdosy 1988:115, 1995:113).
7. Kosambi (1981) notes that there seems to be a direct connection between the first of these and Bronze Age Harappan cities: “even after the destruction of Mohenjo Daro, which is entirely a trade city as shown by its fine weights and poor weapons, the traders persisted, and continued to use the very accurate weights of that period.” (ibid:91). Given what we know of Mesopotamia, with which the Harappan civilization was in close contact, it also seems reasonable to assume that they continued to employ older commercial techniques, and, indeed, “promissory notes” do appear as familiar practices in our earliest literary sources, such as the Jakatas (Rhys Davids 1901:16, Thapar 1995:125, Fiser 2004:194), even if these are many centuries later. Of course, in this case, the marks were presumably meant to confirm the accuracy of the weight, to show that it hadn’t been further trimmed, but the inspiration of earlier credit practices seems likely. Kosambi later confirms this: “The marks would correspond to modern countersignatures on bills or cheques cleared through business houses.” (1996:178–79)
8. Our first literary record of coinage in China is of a kingdom that reformed its currency system in 524 bc—which means that it already had a currency system, and presumably had for some time (Li 1985:372).
9. Schaps 2006:34. For a similar recent argument, Schoenberger 2008.
10.Of course the very first coins were of fairly high denominations and quite possibly used for paying taxes and fees, and for buying houses and cattle more than for everyday purchases (Kraay 1964, Price 1983, Schaps 2004, Vickers 1985). A real market society in Greece, for instance, could only be said to exist when, as in the fifth century, ordinary citizens went to the market carrying minuscule coins of stamped silver or copper in their cheeks.
11. First proposed by Cook (1958), the explanation has since lost favor (Price 1983, Kraay 1964, Wallace 1987, Schaps 2004:96–101; though cf. Ingham 2004:100)—largely, on the argument that one cannot pay soldiers with coins unless there are already markets with people willing to accept the coins. This strikes me as a weak objection, since the absence of coinage does not imply the absence of either money or markets; almost all parties to the debate (e.g., Balmuth [1967, 1971, 1975, 2001) who argues that irregular pieces of silver were already in wide use as currency, and Le Rider (2001), Seaford (2004:318–37) or for that matter Schaps (2004:222–35), who argue that they were not numerous enough to be a viable everyday currency, seem to give much consideration to the possibility that most market trade took place on credit. Anyway, as I’ve noted earlier, it would be easy enough for the state to ensure that the coins became acceptable currency simply by insisting that they were the only acceptable means of payment for obligations to the state itself.
12. Most of the earliest known Greek bankers were of Phoenician descent, and it’s quite possible that they first introduced the concept of interest there (Hudson 1992).
13. Elayi & Elayi 1992.
14. Starr 1977:113; see Lee 2000.
15. It’s interesting to note that, to our knowledge, the great trading nations did not produce much in the way of great art or philosophy.
16. The great exception was of course Sparta, which refused to issue its own coinage but developed a system whereby aristocrats adopted a strict military lifestyle and trained permanently for war.
17. Aristotle himself noted the connection when he emphasized that the constitution of a Greek state could be predicted by the main army of its military: aristocracies if they relied on cavalry (since horses were very expensive), oligarchies in the case of heavy infantry (since armor was not cheap), democracy in the case of light infantry or navies (since anyone could wield a sling or row a boat) (Politics 4.3.1289b33–44, 13.1297b16–24, 6.7.1321a6–14).
18. Keyt (1997:103) summarizing Politics 1304b27–31.
19. Thucydides (6.97.7) claimed that 20,000 escaped from the mines in 421 bc, which is probably exaggerated, but most sources estimate at least 10,000 for most of that century, generally working shackled and under atrocious conditions (Robinson 1973).
20. Ingham 2004:99–100.
21. MacDonald 2006:43.
22. On Alexander’s armies monetary needs, Davies 1996:80 in turn, 83; on his logistics more generally, Engels 1978. The figure 120,000 includes not only actual troops but servants, camp-followers, and so forth.
23. Green 1993:366.
24. The Roman institution was called nexum, and we don’t know entirely how it worked: i.e., whether it was a form of labor contract, whereby one worked off the debt for a fixed term, or something more like African pawn systems, where the debtor—and his or her children—served in conditions roughly like those of a slave until redemption (see Testart 2002 for the possibilities). See Buckler 1895, Brunt 1974, Cornell 1994:266–67, 330–32.
25. Hence, most of the scandalous stories that sparked uprisings against debt bondage centered on dramatic cases of physical or sexual abuse; of course, once debt bondage was abolished and household labor was instead supplied by slaves, such abuse was considered normal and acceptable.
26. The first bronze coins paid to soldiers seem to have been coined around 400 bc (Scheidel 2006), but this was the traditional date according to Roman historians.
27. What I am arguing flies in the face of much of the conventional scholarly wisdom, summed up best perhaps by Moses Finley when he wrote “in Greece and Rome the debtor class rebelled; whereas in the Near East they did not”—and therefore reforms like those of Nehemiah were at least minor, temporary palliatives. Near Eastern rebellion took a different form; moreover, Greek and Roman solutions were both more limited and more temporary than he supposed.
28. Ioannatou 2006 for a good example. Cataline’s conspiracy of 63 bc was an alliance of indebted aristocrats and desperate peasants. On continued Republican debt and land redistribution campaigns: Mitchell 1993.
29. Howgego makes this point: “If less is heard of debt under the Principate it may well be because political stability removed the opportunity for the expression of discontent. This argument is supported by the way in which debt re-emerges as an issue at times of open revolt” (1992:13).
30. Plutarch, Moralia, 828f—831a.
31. There is, needless to say, a vast and conflicting literature, but probably the best source is Banaji (2001). He emphasizes in the late empire, “debt was the essential means by which employers enforced control over the supply of labour, fragmenting the solidarity of workers and ‘personalizing’ relations between owners and employees” (ibid:205), a situation he compares interestingly to India.
32. Kosambi 1966, Sharma 1968, Misra 1976, Altekar 1977:109–38. Contemporary Indian historians, who refer to them as gana-sanghas (“tribal assemblies”), tend to dismiss them as warrior aristocracies supported by populations of helots or slaves, though of course, Greek city-states could be described the same way.
33. In other words, they looked more like Sparta than like Athens. The slaves were also collectively owned (Chakravarti 1985:48–49.) Again, one has to wonder how much this was really the general rule, but I yield to the predominant scholarly opinion on such matters.
34. Arthasastra 2.12.27. See Schaps 2006:18 for a nice comparative commentary.
35. Thapar 2002:34, Dikshitar 1948.
36. There were also taxes, of course, usually ranging from 1/6 to 1/4 of total yield (Kosambi 1996:316; Sihag 2005), but taxes also served as a way to bring goods to the market.
38. And wage labor, two phenomena that, as so often in the ancient world, largely overlapped: the common phrase for workers used in texts from the period was dasa-karmakara, “slave-hireling” with the assumption that slaves and laborers worked together and were barely distinguishable (Chakravarti 1985). On the predominance of slavery, see Sharma 1958, Rai 1981. The extent is contested, but early Buddhist texts do seem to assume that any wealthy family would normally have domestic slaves—which certainly wasn’t true in other periods.
39. Punch-marked coins were also eventually replaced, after Alexander’s brief conquest of the Indus Valley and his establishment of Greek colonists in Afghanistan, by Aegean-style coins, ultimately causing the entire Indian tradition to disappear (Kosambi 1981, Gupta & Hardaker 1985.)
40. It’s referred to as the “Pillar Edict” (Norman 1975:16).
41. There’s a good deal of debate as to when: Schopen (1994) emphasizes there is little evidence for substantial Buddhist monasteries until the first century ad, perhaps three centuries later. This has a great deal of bearing on monetarization too, as we’ll see.
42. “The private trader was regarded as a thorn (kantaka), a public enemy just short of a national calamity, by Arth. 4.2, taxed and fined for malpractices of which many are taken for granted” (Kosambi 1996:243).
43. Those wishing to become monks had to first affirm that they were not themselves debtors (just as they also had to promise they weren’t runaway slaves); but there was no rule saying the monastery itself could not lend money. In China, as we’ll see, providing easy credit terms for peasants came to be seen as a form of charity.
44. Similarly, Buddhist monks are not allowed to see an army, if they can possibly avoid it (Pacittiya, 48–51).
45. Lewis 1990.
46. Wilbur 1943, Yates 2002. The state of Qin, during the Warring States period, not only allowed for army officers to be allocated slaves by rank, but for merchants, craftsmen, and the “poor and idle” to themselves be “confiscated as slaves” (Lewis 1990:61–62).