SANSKRIT PITA AND SAIKYA/SAIKYA TWO TERMS OF IRON AND STEEL TECHNOLOGY IN THE MAHABHARATA.

FITZGERALD, JAMES L.

Two weapons-terms of Mah[bar{a}]bh[bar{a}]rata battle accounts, p[bar{i}]ta and [acute{s}]aikya, are poorly understood. This paper examines the use of both words in the epic closely and concludes: 1) p[bar{i}]ta is the past participle of the verb [surd]p[bar{a}], "drink," and refers to the treatment of "iron" with a liquid bath, i.e., the quenching of carburized iron (effectively a low-carbon steel). 2) [acute{s}]aikya is an allomorph of saikya, which is an adjective based on the noun seka ("pour, cast") from the root [surd]sic, "pour, cast (molten metal)"; it is an adjective meaning "metal that has been fused, metal ready for casting, (previously) molten metal." The word saikya/[acute{s}]aikya must refer to India's ancient steel, famous in the classical Mediterranean world, made by a process essentially the same as that of the famous crucible-fused wootz of South India, long the basic steel of "damascene" blades.

ONE DOES NOT EXPECT TO FIND serious descriptions of battles in the Mah[bar{a}]bh[bar{a}]rata's [acute{S}][bar{a}]nti Parvan, [1] which follows the account of the great Bh[bar{a}]rata war and is principally concerned to cool and calm (pra[acute{s}]amana) the dangerous excess of heat generated by the war. [2] But MBh 12.99 is an interestingly ambitious description of war-making that forms part of the R[bar{a}]jadharma Parvan. It contains two extended analogies describing warfare and the battlefield, analogies that employ terms particular to the epic's battle episodes. The meanings of two of the words used to describe weapons, p[bar{i}]ta and [acute{s}]aikya, are highly problematic, and the purpose of this article is to propose resolutions for both of them. In his comprehensive review of the early use of iron in India, Dilip K. Chakrabarti concludes a review of ancient literary references to iron technology with this comment on literary depictions of "Metallurgical Details": "In the whole range of early Indian literature there is not a single reference to any metallurgical process which can be specifically associated with iron." And, ".... it needs to be emphasized that nowhere in early Indian literature is there a specific reference to steel-making." [3] If my arguments here are correct, the evidence of the Mah[bar{a}]bh[bar{a}]rata requires that both these statements be modified.

According to MBh 12.99, when the illustrious ancient r[bar{a}]jan Ambar[bar{i}]sa went to heaven he was surprised to find that his former general (sen[bar{a}]pati) Sudeva occupied a more exalted position there than he himself did. Sudeva had been a man whose inner self was thoroughly calm (he was pra[acute{s}][bar{a}]nt[bar{a}]tm[bar{a}]) and he had not lived a life of rites and pious deeds as Ambar[bar{i}]sa had. Indra explained to Ambar[bar{i}]sa that Sudeva had been a true sacrificer nonetheless:

Indra uv[bar{a}]ca / etasya vitatas t[bar{a}]ta sudevasya babh[bar{u}]va ha samgr[bar{a}]mayaj[tilde{n}]ah sumah[bar{a}]n ya[acute{s}] c[bar{a}]nyo yudhyate narah [4] 12

samnaddho d[bar{i}]ksitah sarvo yodhah pr[bar{a}]pya cam[bar{u}]mukham

yuddhayaj[tilde{n}][bar{a}]dhik[bar{a}]rastho bhavatiti vini[acute{s}]cayah 13

Ambar[bar{i}]sa uv[bar{a}]ca /

k[bar{a}]ni yaj[tilde{n}]e hav[bar{i}]msy atra kim [bar{a}]jyam k[bar{a}] ca daksin[bar{a}]

rtvija[acute{s}] c[bar{a}]tra ke prokt[bar{a}]s tan me br[bar{u}]hi [acute{s}]atakrato 14

Indra uv[bar{a}]ca /

rtvijah ku[tilde{n}]jar[bar{a}]s tatra v[bar{a}]jino 'dhvaryavas tath[bar{a}]

hav[bar{i}]ms[bar{i}] param[bar{a}]ms[bar{a}]ni rudhiram tv [bar{a}]jyam eva ca 15

srg[bar{a}]lagrdhrak[bar{a}]kol[bar{a}]h sadasy[bar{a}]s tatra satrinah

[bar{a}]jya[acute{s}]esam pibanty ete havih pr[bar{a}][acute{s}]nanti c[bar{a}]dhvare 16

Indra said:

Son, this Sudeva performed the tremendous sacrifice of battle, and so does any other man who wages war: Every warrior equipped for battle is ritually consecrated, and when he goes to the front of the army he gains the right to perform the sacrifice of battle--that's a settled conclusion.

Ambar[bar{i}]sa said:

What are the offerings in this sacrifice? What the clarified butter? What are the presents for the priests in it? And who are supposed to be the priests? Tell me this, O [acute{S}]atakratu.

Indra said:

The elephants there are the invoking priests, and the horses are the officiating priests. The chunks of the enemy's flesh are the offerings and their blood is the clarified butter. Jackals, vultures, and ravens sit in the ritual assembly and are participants in the solemn rite. They drink what remains of the clarified butter and they eat the offerings of the rite.

Indra's metaphor moves from participants to implements and we encounter the problematic word p[bar{i}]ta:

pr[bar{a}]satomarasamgh[bar{a}]t[bar{a}]h khadga[acute{s}]aktipara[acute{s}]vadh[bar{a}]h

jvalanto ni[acute{s}]it[bar{a}]h p[bar{i}]t[bar{a}]h srucas tasy[bar{a}]tha satrinah 17

c[bar{a}]paveg[bar{a}]yatas tiksnah parak[bar{a}]y[bar{a}]vad[bar{a}]ranah

rjuh suni[acute{s}]itah p[bar{i}]tah s[bar{a}]yako 'sya sruvo mah[bar{a}]n 18

dv[bar{i}]picarm[bar{a}]vanaddh[acute{s}] ca n[bar{a}]gadantakrtatsaruh

hastihastagarah khadhah sphyo bhavet tasya samyuge 19

The masses of darts, lances, swords, spears, and battle-axes--blazing, whetted, and pita--are the sruc ladles of each of the rite's partakers. The keen arrow--straight, whetted, and pita--racing away from the bow's violent thrust and splitting the body of an enemy is his great sruva ladle. And the sword wielded in battle by arms big as elephants' trunks--its grip made of elephant's tusk and its sheath from tiger skin--would be its sphya stirrer.

The other problem word, [acute{s}]aikya, turns up in the next [acute{s}]loka:

jvalitair ni[acute{s}]itiaih p[bar{i}]taih pr[bar{a}]sa[acute{s}]aktipara[acute{s}]vadhaih

[acute{s}]aiky[bar{a}]yasamayais t[bar{i}]ksnair abhigh[bar{a}]to bhaved vasu 20

[bar{a}]veg[bar{a}]d yat tu rudhiram samgr[bar{a}]me syandate bhuvi

s[bar{a}]sya p[bar{u}]rn[bar{a}]hutir hotre samrddh[bar{a}] sarvak[bar{a}]madhuk 21

The blows landed with the keen darts, spears, and battle-axes made of [acute{s}]aikya iron--gleaming, whetted, and pita-- would be its riches. The blood which runs upon the earth from the violence of the battle is its full libation, the rich cow from which all wishes flow.

The episode develops the analogy further and adds to it another powerfully constructed passage that describes the field after battle by means of a detailed river analogy. But in the midst of this elaborate and studied praise of war, what do the words p[bar{i}]ta and [acute{s}]aikya mean? I take them each in turn.

I. P[bar{I}]TA

The Petersburg dictionary registers three words of the form "p[bar{i}]ta." The first is the past participle p[bar{i}]t[acute{a}] of the root [surd]p[bar{a}] (p[bar{a}]ti, p[acute{i}]bati), "drink," a word known from the RV onward. The second, an adjective meaning "yellow," appears first in the Ch[bar{a}]ndogya Upanisad and is common in the epics. Third is a rare past participle of the root [surd]{p[bar{i}] (also listed variously as pi, py[bar{a}], py[bar{a}]y, and pyai), p[acute{a}]yate, py[bar{a}]yate, "puff out; be or become full" (the normal past participle of this root is p[bar{i}]na, a word unattested before the MBh). This third form has been attested previously only in [bar{a}]p[bar{i}]ta (at RV 8.9.19) and pr[acute{a}]p[bar{i}]ta (at RV 7.41.7 and 7.80.3), but it seems to occur, without any modifying prefix, at MBh 3.186.66, as I will point out below.

There are over one hundred instances of one or another of the words p[bar{i}]ta in the MBh, and more than forty of these occur in connection with weapons, as in the passages cited above. In 12.99.17, darts (pr[bar{a}]sa), lances (tomara), swords (khadga), spears ([acute{s}]akti), and battle-axes (para[acute{s}]vadha) were all referred to as ni[acute{s}]ita (which, along with its cognates, means [acute{s}]il[bar{a}][acute{s}]ita, "whetted on a grindstone," "ground, i.e., shaped or sharpened on a stone") and p[bar{i}]ta. And so too the arrow (s[bar{a}]yaka) next in 99.18. Similarly, 4.32.6, an upaj[bar{a}]ti stanza, tells us that the armies of the Trigarlas and the Matsyas attacked each other "gad[bar{a}]sikhajgai[actue{s}] ca para[acute{s}]vadhai[acute{s}] ca pr[bar{a}]sai[acute{s}] ca t[bar{i}]ksn[bar{a}]grasup[bar{i}]tadh[bar{a}]raih," that is, with "clubs, swords, short swords, battle-axes, and darts with sup[bar{i}]ta blades and sharp points." In other weapon-contexts of the MBh, other iron ([bar{a}]yasa, ayomaya) weapons or parts of weapons are also described as p[bar{i}]ta: iron clubs ([bar{a}]yasa parigha, 1.17.16), arrows ([acute{s}]ara, 3.24.3; bhalla, 6.43.16; ksurapra and a[tilde{n}]jalika, 6.58.38; vi[acute{s}]iskha, 11.16.35), another dart or hurled knife (rsti, 8.68.27), another sword (nistrim([acute{s}]a, 4.38.34), knives (ksura, 6.58.38), and sharp-bladed weapons in general ([acute{s}]astra, 7.138.17). So, which word p[bar{i}]ta is being used in this connection and what does it mean?

The basic approach was correctly indicated by N[bar{i}]lakantha, who comments on this word at least six times. He interprets p[bar{i}]ta as the past participle of [surd]p[bar{a}] "drink," and though he usually understands this to signify some kind of treatment of metal with a liquid, doubtless water, making the metal better suited to its use as weaponry, his explanations of exactly what this treatment was are not detailed and are not consistent. The Petersburg dictionary did not initially have access to a complete edition of N[bar{i}]lakantha's commentary (which was not published until 1860-62), and the commentator's glosses of p[bar{i}]ta seem never to have become available to its authors. [5] B[ddot{o}]htlingk and Roth do take brief note of this usage of p[bar{i}]ta (in connection with arrows only) and, like N[bar{i}]tlakantha, they see the past participle of [surd]p[bar{a}] at work. They gloss p[bar{i}]ta as meaning here "soaked" (getr[ddot{a}]nkt) and "dipped, or immersed in oil" (eingetaucht in Oel), or "having imbibed," "full of" (imbibirt, voll von), influenced in part by compounds such as tailap[bar{a}]yita (based on the causative participle of [surd]p[bar{a}]), describing the immersion of arrows in oil (see below). They did not arrive at an understanding of a treatment of metal that enhanced its basic qualities. Ganguli, the translator of the Vulgate MBh that was published by P. C. Roy, [6] follows Nilakantha's general lead and consistently renders p[bar{i}]ta with "(well-)tempered," evidently using "tempered" broadly to include the quenching of the metal. Curiously, E. W. Hopkins does not mention the word p[bar{i}]ta at all in his survey of the war technology of the MBh, though he takes note of the frequent descriptor [acute{s}]ita. [7] Because the word p[bar{i}]ta occurs several times in the critical edition of the MBh before it occurs in corresponding passages of the Vulgate, van Buitenen, thus unfamiliar with N[bar{i}]lakantha's and Ganguli's interpretations, took p[bar{i}]ta to be the adjective mea ning "yellow," which he understood to refer to "yellow metal," which he took to be copper. [8] Most recently, in the one R[bar{a}]m[bar{a}]yana passage where p[bar{i}]ta occurs with reference to weapons (arrow-heads), [9] Goldman and Sutherland Goldman, following the majority of the R[bar{a}]m[bar{a}]yana commentators, translate pa[tilde{n}]c[tilde{a}]yas[tilde{a}]s tiksn[bar{a}]h sit[bar{a}]h pitamukh[bar{a}]h [acute{s}]ar[bar{a}]h as "... five bright iron arrows. Sharp and yellow-tipped...." According to the translators' note to this passage, four of the commentators here mention gold as a source of the yellow color, and another sees it as an attribute of the quality of the metal; but copper is not offered as an explanation. Satyatirtha's commentary (alone) interprets p[bar{i}]ta as a past participle of [surd]p[bar{a}] ("p[bar{a}]yitap[bar{a}]n[bar{i}]yamukha"), which the Goldmans render and explain as "having mouths [i.e., tips] that have been made to drink what they were supposed to drink [presumably the blood of enemies]."

Van Buitenen's Claim of Copper Weapons

Before presenting N[bar{i}]lakanta's brief comments and discussing the meaning of p[bar{i}]ta, I must deal with van Buitenen's interpretation, "yellow," referring to copper. There are nine occurrences of p[bar{i}]ta in the context of weapons in the first five books of the epic--those van Buitenen translated before his untimely death.

In the first such occurrence, [10] van Buitenen translates the phrase parighai[acute{s}] c[acute{a}]yasaih p[bar{i}]taih as "with copper-spiked bludgeons," and he adds in a note, "the text has p[bar{i}]ta, 'yellow,' which I take to refer to copper." [11] The word ayas and its derivatives in the MBh generally refer to iron, [12] and he only basis for the idea that copper, or bronze, weapons are involved here is the presence of the word p[bar{i}]ta, its interpretation as "yellow," and the postulate that "yellow metal" here means copper. But, as the Vedic descriptive terms for copper (loha, lohita "red") suggest, unalloyed copper is a reddish metal. And bronze, the principal copper-alloy in the ancient world, [13] is not likely to be described as yellow either, as it normally has "a rich golden brown color." [14] Brasses exhibit a range of colors from golden to yellow to whitish, but these alloys of copper and zinc seem not to have been used in India until well after iron was in common use and ayas had come to signify iron, more or less exclusively. By the time of the MBh, loha and lohita mean "metal" generically, though they actually refer to iron much of the time, particularly when referring to the utilitarian metal of weapons. [15] N[bar{i}]lakantha glosses parigha at 1.17.16 with "sarvatah kantakito lohadandah," and the metal he has in mind can only be iron ("an iron rod with spikes on every side"). [16] It is possible that when van Buitenen reflected upon N[bar{i}]lakantha's gloss of [bar{a}]yasa with a word that once meant copper (and still signifies copper at least occasionally in the MBh) and when he considered that fact in conjunction with the enigmatic reading p[bar{i}]ta in the Poona text (N[bar{i}]lakantha's text reads t[bar{i}]ksna [17]), he might have surmised that the MBh here preserved some vestige of a Vedic bronze age.

In the next instance, [bar{A}]ranyaka Parvan 3.24.3, van Buitenen renders the upaj[bar{a}]ti p[bar{a}]da dhan[bar{u}]msi varm[bar{a}]ni [acute{s}]ar[bar{a}]ms ca p[bar{i}]t[bar{a}]n, "And bows and shields and copper arrows." [18] The next is Vir[bar{a}]ta Parvan 4.5.24, where van Buitenen renders khadg[bar{a}]m[acute{s}] ca p[bar{i}]it[bar{a}]n d[bar{i}]rgh[bar{a}]m[acute{s}] ca "long yellow swords." [19] But he has second thoughts here, as he asks in a note: "Does this imply they were actually copper or brass, or is it simply an old epithet that has stuck?" [20] One of his concerns is the fact that the Virata Parvan is relatively younger than most of the rest of the written Sanskrit text (at least in its language and composition), and was thus certainly composed well after the close of chalcolithic times. [21] Van Buitenen again shows himself less than fully convinced when, thirty odd chapters later, he renders the phrase p[bar{i}]t[bar{a}]h sarv[bar{a}]yas[bar{a}]h [acute{s}]ar[bar{a}]h with "arrows ... ma de of copper and solid iron," translating [bar{a}]yasa as "iron." [22] This is the standard sense of the word in the MBh, but if one believes that p[bar{i}]ta here means "yellow (metal)," that is, that the passage refers to copper-based weapons, then sarv[bar{a}]yasa should have been rendered with "solid copper (or bronze)"; or, minimally, "solid metal." [23] There are five other instances of the word p[bar{i}]ta used in the context of weapons in the Vir[bar{a}]ta Parvan, and van Buitenen renders them in the same way as those noted. [24]

Van Buitenen's doubts about copper weapons were well placed. His claim that the MBh knows copper weapons is based only on his bafflement over the word p[bar{i}]ta, and it is not at all likely that copper, bronze, or brass weapons are the "p[bar{i}]ta" arrows, axes, swords, etc., referred to in the MBh. Wilhelm Rau's study of the terminology of metals and metal-working in Vedic texts suggests a very plausible evolution of the use of the word ayas in pre-Mah[bar{a}]bh[bar{a}]rata Sanskrit literature, an evolution that conforms well to the general archaeological record in the Panjab and the upper Ganges valley. That record dates the first major appearances of iron against the chalcolithic background in northern India to the first century or two before 1000 B.C. and shows its proliferation in the subsequent centuries down to 500 B.C. [25] Rau hypothesizes that in the earlier strata of Vedic texts the unmarked word ayas ("metal") and its derivative [bar{a}]yasa meant copper. When iron was meant in these texts, th e marked expressions [acute{s}]y[bar{a}]ma- ("dark") or krsna- ("black") ayas was used. But from late Vedic texts onward the unmarked words ayas and [bar{a}]yasa normally refer to iron, and copper is designated by the marked expression loha-ayas ("red metal"). [26] By the time of the MBh not only ayas, but also loha had come to refer to iron, [27] and copper (probably including bronze) is designated unambiguously by t[bar{a}]mra ("red") and brass (and bronze) is k[bar{a}]msya (from kamsa, "metal cup"). Under these particular names these two metals are distinguished from "plain metal," ayas unmarked, i.e., iron. Copper, or bronze, t[bar{a}]mra, almost never occurs in a military context--I know of only one such instance and there it occurs alongside silver, and probably iron, and refers to body armor. [28] Brass, or bronze, k[bar{a}]msya, occurs in military contexts in connection with body armor only. [29]

Beyond van Buitenen's p[bar{i}]ta-based conjecture there is no evidence of copper or bronze weapons in the MBh, a text that, for all the vagueness of its dating overall, was entirely composed well after the use of iron had generally displaced copper as the main utilitarian metal in north India. In fact, as the second part of this article will show, iron is not only commonplace in the MBh, it was being superseded in some situations by crucible steel. The chalcolithic age was well over when most of the words of the epic narratives (eventually set down in writing as a single text in the fourth century A.D.) were actually under composition in oral and written traditions (probably during the time of the Mauryas, 320 B.C. to 185 B.C., and even more intensively during the century after the [acute{S}]u[dot{n}]ga overthrow of the Mauryas in 185 B.C.).

There is, however, one passage that might appear at first glance to offer support to the idea of special, copper-based weapons. In the remarkable thirty-eighth chapter of the Vir[bar{a}]ta Parvan, the Matsya prince Uttara wonders at the weapons-cache of the P[bar{a}]ndavas, which Arjuna (disguised as a eunuch and serving, with richly developed irony, as Uttara's charioteer in battle) has just had him fetch from its hiding place. Uttara sees five marvelous bows, five sets of arrows, and five wondrous swords, and he asks Arjuna about each of the fifteen items. Of the second set of arrows he asks: vip[bar{a}]th[bar{a}]h prthavah kasya g[bar{a}]rdhrapatr[bar{a}] [acute{s}]il[bar{a}][acute{s}]it[bar{a}]h / h[bar{a}]ridravarn[bar{a}]h sunas[bar{a}]h . pit[bar{a}]h sarv[bar{a}]yas[bar{a}]h [acute{s}]ar[bar{a}]h. Van Buitenen takes the epithet h[bar{a}]ridra-varna ("turmeric-colored") as the equivalent of p[bar{i}]ta, "yellow," and renders pita here "made of copper": "And [whose are] these wide arrows with vulture f eathers, whetted on stone, yellow like turmeric, finely tipped, made of copper and solid iron?" With or without the word p[bar{i}]ta, the epithet h[bar{a}]ridravarna does suggest, in conjunction with sarv[bar{a}]yasa (solid metal, if not solid iron), a bright orangish-yellow metal that well might be a species of bronze or brass, or which could even be gold. Such expensive and lavish arrows are not at all implausible in connection with this special demonstration of the P[bar{a}]ndavas' weapons, and it is possible the poet had solid brass or bronze or gold arrows in mind here.

But the text does not oblige us to imagine these arrows only this way, and it gives no other characterization of the metal involved. I would expect there to be some explicit indication of the fact that a precious or special metal were involved; in the absence of any such indication a more mundane explanation of the epithet would be welcome. The possibility that solid iron arrows might display a turmeric hue finds an explanation when we read, a little later in the same chapter, Arjuna's answers to Uttara's awestruck questions. Arjuna tells him at 4.38.49-50 that these arrows are Nakula's: h[bar{a}]ridravarn[bar{a}], ye tv ete hema-pu[dot{n}]kh[bar{a}]h [acute{s}]il[bar{a}][acute{s}]it[bar{a}]h /nakulasya kal[bar{a}]po 'yam pa[tilde{n}]ca[acute{s}][bar{a}]rd[bar{u}]la- laksanah. [30] The epithet h[bar{a}]ridravarna is repeated and the newly added detail that these arrows have golden nocks (hemapu[dot{n}]kh[bar{a}]h) accounts for the arrows' having a turmeric hue, without requiring that the shaft be made of som e unnamed metal rather than iron.

But the MBh may allow us to go a bit further than this. As I will point out briefly later, there are a number of places that describe the conditioning of metal weapons for battle and tell us that, apart from being sharpened on stone, weapons were "cleaned on grindstones" ([acute{s}]il[bar{a}]dhauta), "cleaned with oil" (tailadhauta, which N[bar{i}]lakantha glosses with "polished with oil," tailam[bar{a}]rjita), and "polished by smiths" (karm[bar{a}]ram[bar{a}]rjita). We cannot know exactly how these weapons looked, but the MBh's descriptions of them in battle are replete with epithets of brightness and fire: The arrows and swords gleam, blaze, shine like the sun, sparkle, and streak across the darkened skies over the battle like flames of fire. [31] There is a marked emphasis upon the luminous qualities of the weapons as they are used, and this fact makes more than simply poetic sense if we take seriously all these descriptions of the weapons as being ground and polished. We can easily imagine that these sar v[bar{a}]yasa arrows of Bh[bar{i}]ma's have had all the red and black oxides of iron removed and been polished and oiled (protecting them against rapid reoxidation). The resulting silvery-gray shafts would not be bright or highly reflective, but perhaps it is not too far-fetched to wonder if the polished shaft reflects some of the hue of the arrow's golden nocks along its length; or at the least, reinforces the bright impression of those golden nocks in a way that black, oxide-coated iron would not. This last train of thought may be pushing matters too far, but we need more compelling evidence than the epithet h[bar{a}]ridravarna for sarv[bar{a}]yasa here to mean some metal other than iron, and the word p[bar{i}]ta does not provide such evidence, even if we take it to mean yellow.

N[bar{i}]lakantha's Six Unclear Glosses

N[bar{i}]lakantha gives half a dozen glosses of p[bar{i}]ta scattered here and there throughout his commentary on the Vulgate, and though his explanations are not univocal in their details, they do provide us with a general way of understanding the word p[bar{i}]ta in connection with iron weapons that is more than plausible. Here are the glosses in order:

1. MBh 4.32.6, cited above, seems to be the first passage where the word p[bar{i}]ta occurs in a weapons' context in the Vulgate. [32] Both the Poona text and the Vulgate read pr[bar{a}]sai[acute{s}] ca t[bar{i}]ksn[bar{a}]grasup[bar{i}]tadh[bar{a}]raih in p[bar{a}]da d. N[bar{i}]lakantha's gloss of sup[bar{i}]ta is interesting:

sup[bar{i}]t[bar{a}]h: sutar[bar{a}]m p[bar{i}]t[bar{a}]h; [acute{s}]osit[bar{a}]h kr[acute{s}][bar{i}]krt[bar{a}]h an[bar{a}]y[bar{a}]sena

[acute{s}][bar{i}]ghram para[acute{s}]ariraprave[acute{s}]aksam[bar{a}] dh[bar{a}]r[bar{a}][acute{s}] ca yes[bar{a}]m, te ...

sup[bar{i}]ta [means] 'p[bar{i}]ta to a very high degree,' [that is,] dried out, made lean. [The text describes weapons] that have blades that are ready to enter an enemy's body easily and quickly. [33]

N[bar{i}]lakantha understands p[bar{i}]ta to be the past participle of [surd]p[bar{a}], "drink," which here would have to signify either that the metal has drunk some liquid, or that the metal has been "drunk." Both senses are possible, [34] though the latter, passive, sense is more likely and better fits N[bar{i}]lakantha's explanatory glosses, [acute{s}]osit[bar{a}]h kr[acute{s}][bar{i}]krt[bar{a}]h. N[bar{i}]lakantha seems to be describing the smith's putting a thin, hard edge on a blade by repeatedly heating it and hammering it, and he seems to conceive this forging process as one of removing liquid from the metal being worked.

2. At 4.38.26 (the first "turmeric-colored arrows" passage discussed above), N[bar{i}]lakantha glosses, p[bar{i}]t[bar{a}]h: karm[bar{a}]rena tejanajalam p[bar{a}]yit[bar{a}]h ("'p[bar{i}]t[bar{a}]h' [means] 'having been made to drink tejana-fluid by the smith'"). The word tejana refers to an action or process that makes something "sharp, hot, or bright" (all three of these ideas are commingled in words that derive from the Skt. root [surd]tij, such as tejas, a word that, among its other sometimes elusive senses, can refer to the sharp edge of a knife or the tip of a flame). [35]

3. At 6.43.45ab the text reads s[bar{a}]yakena sup[bar{i}]tena t[bar{i}]ksnena ni[acute{s}]itena, and N[bar{i}]lakantha glosses, sup[bar{i}]tena: susthu p[bar{a}]yitena \ t[bar{i}]ksnena s[bar{u}]ksmadh[bar{a}]rena yato ni[acute{s}]itena [acute{s}][bar{a}]noll[bar{i}]dhena ("sup[bar{i}]ta [means] 'immersed [i.e., made to drink] thoroughly'; t[bar{i}]ksna [(sharp) means] 'having a very thin edge,' since it was ni[acute{s}]ita [(sharpened, whetted), that is,] 'licked by a whetstone'").

4. At 6.49.21 N[bar{i}]lakantha glosses p[bar{i}]t[bar{a}]n as p[bar{a}]yitajal[bar{a}]n ("having been forced to drink fluid [or water]," i.e., having been immersed in some kind of bath).

5. Skipping about thirty occurrences, as he understandably passes over most of the war books with little to say, N[bar{i}]lakantha glosses p[bar{i}]t[bar{a}]n at 7.63.4 (B. 7.87.4), simply, as p[bar{a}]yitodak[bar{a}]n "having been forced to drink fluid [or water]").

6. The last passages in the MBh where p[bar{i}]ta characterizes a weapon or blade are 12.99.17-20 and 12.101.7, and N[bar{i}]lakantha's final gloss is at 12.99.17. Here he writes: p[bar{i}]t[bar{a}]h: ks[bar{a}]rap[bar{a}]n[bar{i}]yena sambh[bar{a}]vit[bar{a}]h ("p[bar{i}]t[bar{a}]h [means] produced by means of a caustic immersion"). The gerundive p[bar{a}]n[bar{i}]ya means literally "(the fluid) that is to be drunk," i.e., a bath or immersion into which the iron is submerged. The word ks[bar{a}]ra basically means "caustic, acrid, corrosive" and can refer to acid, saline, or alkaline solutions.

The Word p[bar{i}]ta as a Technical Term of Metallurgy

N[bar{i}]lakantha's first gloss does not agree with the later ones: It seems to describe a smith's putting a hard, slender edge on a blade by causing the rasa, the sap, the juice, the essence, of the metal, to be drawn out of it. [36] Perhaps he here refers to the final tempering of the blade, in which it is heated and then allowed to cool slowly, a process that removes the brittleness of steel that has been hardened by prior quenching. [37] In addition to, or instead of, tempering, N[bar{i}]lakantha's first gloss could refer simply to the general processes of heating and hammering iron at the forge.

By contrast, all five of N[bar{i}]lakantha's later glosses of p[bar{i}]ta refer to iron being treated with a bath of some kind. He usually glosses the word with p[bar{a}]yita ("made to drink, immersed in, dipped in"), once specifying the bath (the p[bar{a}]n[bar{i}]ya, "that which is to be drunk," that in which "something is to be immersed or dipped") as tejanajala, once as ks[bar{a}]rap[bar{a}]n[bar{i}]ya, and twice simply as water. There are at least three metallurgical ways to understand these liquid treatments of iron blades: the quenching of the iron (which imparts hardness to iron that has been effectively carburized [i.e., made into a low-carbon steel] by the smith as part of his working the iron at the forge); the pickling of an iron implement with some kind of caustic solution to clean and recondition it for use; and the etching of a blade with a reagent that, particularly with blades made of Indian wootz steel (see, below, the discussion of [acute{s}]aikya/saikya, may bring out striking moir[acute{ e}] patterns based on the varying crystalline structure of the underlying metal.

The general idea of N[bar{i}]lakantha's explanation of p[bar{i}]ta in terms of a liquid treatment of iron is not only plausible, it accords very well with some other characterizations of weapons often made in the MBh. As noted above, p[bar{i}]ta often occurs with the description [acute{s}][bar{i}]ta, "ground on a whetstone." [38] Additionally, the epic occasionally refers to the weapons having been "cleaned" (dhauta), [39] "cleansed or polished with stones" ([acute{s}]il[bar{a}]dhauta), [40] "soaked (or given some kind of liquid treatment) by smiths" (karm[bar{a}]rap[bar{a}]yita), [41] "polished by smiths" (karm[bar{a}]ram[bar{a}]rjita), [42] "cleaned, or bathed, in oil" (tailadhauta), [43] or "soaked, or dipped, in oil" (tailap[bar{a}]yin, tailap[bar{a}]yita). [44]. Since we can take the word p[bar{i}]ta as one more process among several by which the MBh describes smiths producing or preparing iron weapons, and since N[bar{i}]lakantha's explanations of p[bar{i}]ta take us to these processes directly, and s ince we lack any other plausible way to understand the word, it clearly seems best to follow N[bar{i}]lakantha's lead and take the word as referring to a smith's treating with some kind of liquid the iron he is working.

Is it further possible to infer which of the three processes the epic might have understood by the word p[bar{i}]ta? Does p[bar{i}]ta refer to the quenching of the blade as part of its forging? To a cleansing bath to remove dirt, rust, etc.? Or to an etching of the blade to bring out any "marvelous forms" (citrar[bar{u}]p[bar{a}]ni) that may then become apparent on the blade's surface? I think the word probably refers, in general, to the quenching of the blade, whether as part of its original fabrication, or as part of its reconditioning for use in battle. There is no particular evidence I know of in the MBh to suggest that etching was actually done.

And between pickling and quenching, quenching is the more likely interpretation. First of all, quenching would seem to constitute considerably higher praise of an iron blade than pickling, notwithstanding all of the emphasis upon the weapons' sparkling luminosity. Quenching a carburized piece of iron makes the metal much harder; so describing a blade as p[bar{i}]ta would praise it in an important functional way, similar to saying that it was t[bar{i}]ksna, or that it had been "sharpened, whetted" ([acute{s}]ita). But if p[bar{i}]ta means merely "pickled," "cleansed," then the adjective describes, more mundanely, old weapons. N[bar{i}]lakantha's gloss at 4.38.26, p[bar{i}]t[bar{i}]h: karm[bar{a}]rena tejanajalam p[bar{a}]yit[bar{a}]h, indicates the key idea--a bath that serves to make the iron hard, sharp, fiery, and painful. The word tejana could also refer to luminosity, and thus pickling, but the forge, with its hot charcoal and its working of the hot iron, as well as its tejanajala for quenching the hot ir on, represents a more impressive process of imparting to the metal the desired tejas than does pickling, which, at best, merely reactivates the energy put into the blade by its original forging. In both cases the tejana bath, perhaps blessed with mantras, would probably be understood to involve various unseen forms of Agni (the god of fire), but the working at the forge involves an impressive process with Agni manifest. [45] Secondly, while a pickling bath might be appropriate to some large metal pieces, it seems likely that with regard to blades, rust would likely remove much of the metal of a blade's thin edge in a short time. Thus any reconditioning of old blades probably involved reheating and rehammering at the forge, requenching, retempering, and regrinding to make a somewhat smaller blade from the remnant of an older blade.

So, in conclusion, N[bar{i}]lakantha's explanation of p[bar{i}]ta as a liquid treatment of iron seems the best way to understand this word, and the most plausible interpretation of this treatment is to see it as the quenching of iron blades as part of their fabrication at the forge. This reasoning is supported by a clear allusion to quenching iron at MBh 12.120.l9ef. I think "hardened" is the best translation of most instances of p[bar{i}]ta in the MBh, rather than a more literal "quenched" or "immersed." That translation expresses the desired effect of the quenching process; that such an effect is obtained seems to be the point of praising a blade as p[bar{i}]ta.

An Incidental Point on p[bar{i}]ta, "fat, swollen"

The word p[bar{i}]ta occurs at 3.l86.66 in the seer M[bar{a}]rkandeya's description of marvelous clouds that will arise in one phase of the world's coming to an end.

tato gajakulaprakhy[bar{a}]s tadinm[bar{a}]l[bar{a}]vibh[bar{u}]sit[bar{a}]h

uttisthanti mah[bar{a}]megh[bar{a}] nabhasy adbhutadar[acute{s}]an[bar{a}]h 65

ke cin n[bar{i}]lotpala[acute{s}]y[bar{a}]m[bar{a}]h ke cit kumudasamnibh[bar{a}]h

ke cit ki[bar{n}]jalkasamk[bar{a}][acute{s}][bar{a}]h kecit p[bar{i}]t[bar{a}]h payodhar[bar{a}]h 66

ke cid dh[bar{a}]ridrasamk[bar{a}][acute{s}][bar{a}]h k[bar{a}]k[bar{a}]ndakanibh[bar{a}]s [46] tath[bar{a}]

ke cit kamalapatr[bar{a}]bh[bar{a}]h ke cid dhi[dot{n}]gulakaprabh[bar{a}]h 67

The first [acute{s}]loka describes the occurrence of these massive clouds and the next begins a protracted description of their variety: "Some are dark like blue lotuses and some look like white lotuses. Some clouds are like the filaments of a flower and some are puffy (p[bar{i}]ta)." Van Buitenen took p[bar{i}]ta as "yellow" here, even though the next p[bar{a}]da (67a) refers to some that are "turmeric-colored" (h[bar{a}]ridrasamk[bar{a}][acute{s}][bar{a}]h). The strong contrast between n[bar{i}]lotpala and kumuda in 66ab sets up the expectation that 66cd will contain a similarly clear juxtaposition. One of the most striking elements of the filament of a flower's stamen is its slenderness; if we take p[bar{i}]ta to be the past participle of [surd]p[bar{i}], pyai, "become full, swell, puff out," we have the same kind of clear juxtaposition that 66ab presents. Given that this interpretation also removes the blemish of redundancy from an obviously carefully constructed passage, it seems clear we have one of th e rare survivals of this participle. Three manuscripts from Bengal and one Devanagar[bar{i}] manuscript read p[bar{i}]na, the normal past participle of [surd]p[bar{i}], pyai, instead of p[bar{i}]ta in 66d, an improvement or correction of the text that can be construed as agreeing with the interpretation of p[bar{i}]ta I propose.

II. [acute{S}]AIKYA

The case of [acute{s}]aikya is more complicated and confused. There are seventeen passages in the MBh where [acute{s}]aikya (or the allomorph saikya [47]) occur. In eight of these the word is compounded with a following [bar{a}]yasa; [48] in one, with [bar{a}]yasamaya; [49] and in one other, with ayasmaya. [50] In seven other instances it occurs as a free-standing word, without any derivative of ayas, and then is used five times as an adjective [51] and twice as a noun. [52] There are also at least two instances where [acute{s}]aikya occurs in the Vulgate text but not the Poona text, both times as a free-standing word. [53] In seventeen of these nineteen instances the word is connected to battle gear (once it refers to body armor, twelve times to a club or mace, gad[bar{a}] [ten of these refer to Bh[bar{i}]ma's club]). The Petersburg dictionary offers two uncertain glosses for the word [acute{s}]aikya: "1) adj. etwa damascirt" [54] and "2) m. etwa eine Art Schleuder." [55] The first of these glosses is an in ference based on the passages where [acute{s}]aikya modifies the word ayas or one of its derivatives and this sense is supported by two of N[bar{i}]lakantha's seven inconsistent glosses on the word. [56] The Petersburg lexicographers made this inference in the light of long European knowledge of India's pre-eminence in steel-making in the ancient world, [57] knowledge that began to be complemented by reports beginning in the eighteenth century that described India's crucible-fused, "wootz" ["steel," or "superior iron"], [58] which formed the basis of the Islamic world's damask, or damascened, steel fabrications. These reports made a strong and lasting impression on European metallurgy in the nineteenth century. [59] The second gloss of the Petersburg dictionary is a conjecture based on another explanation given by N[bar{i}]lakantha at four places where [acute{s}]aikya occurs as a free-standing noun or adjective. [60] From N[bar{i}]lakantha's explanation of [acute{s}]aikya as "(something, e.g., a pot, pouch, o r receptacle) carried in or supported by an apparatus of cords," a [acute{s}]ikya, B[ddot{o}]htlingk and Roth conjectured "Schleuder," hurling-sling, which Monier-Williams broadened out simply to the ambiguous English "sling." Ganguli's translations of [acute{s}]aikya mirrored N[bar{i}]lakantha's interpretations quite faithfully--[acute{s}]aikya with a word derived from ayas is translated as "hard iron" or the like, and [acute{s}]aikya by itself is rendered in terms of "sling" and the like. Van Buitenen translated [acute{s}]aikya/saikya nine times in the first five books of the MBh and, initially, he too followed N[bar{i}]lakantha's pattern of interpretation, though later he utilized these two interpretations eclectically. An examination of this word's record in the MBh reveals a simpler but very interesting situation, one which suggests that [acute{s}]aikya has nothing to do with the word [acute{s}]ikya or with any kind of sling.

Passages Where [acute{s}]aikya/saikya Modifies a Word Derived from ayas

Most of the seventeen instances of the word are clear and straightforward. The pattern becomes evident if we look first at occurrences of [acute{s}]aikya/saikya with a word derived from ayas (iron, or simply, [generic utilitarian] metal). The passage quoted at the outset of this article (12.99.20) contains the final instance of [acute{s}]aikya/saikya in the MBh, but it is representative of the majority of cases.

jvalitair ni[acute{s}]itaih p[bar{i}]taih pr[bar{a}]sa[acute{s}]aktipara[acute{s}]vadhaih

[acute{s}]aiky[bar{a}]yasamayais t[bar{i}]ksnair abhigh[bar{a}]to bhaved vasu

Darts, spears, and battle-axes with hardened (p[bar{i}]ta) blades that have been honed sharp (ni[acute{s}]itaih...t[bar{i}]ksnaih) [61] are made of [aute{s}]aikya iron or metal, which Ganguli renders here with "all made of hard iron." Earlier, at 3.157.63-64, as Bh[bar{i}]ma fights with the r[bar{a}]ksasa Manimat, the MBh says of him: [62]

so 'tividdho mahe[acute{s}]v[bar{a}]sah [acute{s}]akty[bar{a}]mitapar[bar{a}]kramah gad[bar{a}]m jagr[ddot{a}]ha kauravyo gad[bar{a}]yuddhavi[acute{s}][bar{a}]radah t[bar{a}]m pragrhyonnadan bh[bar{i}]mah sarva[acute{s}]aiky[bar{a}]yas[bar{i}]m gad[bar{a}]m taras[bar{a}] so 'bhidudr[bar{a}]va manimantam mah[bar{a}]balam

Bh[bar{i}]ma's club here is described with almost the same expression as the various weapons at 12.99.20--ayas that is sarva[acute{s}]aikya. Van Buitenen translates according to the Petersburg dictionary: "entirely made of damasked steel." N[bar{i}]lakantha's commentary here registers the second [63] of his three explanations of the word [acute{s}]aikya: "what pulverizes one's enemies" ('[acute{s}]aiky[bar{a}]m': [acute{s}][bar{i}]kayati, [acute{s}]atr[bar{u}]n par[bar{a}]bhavatiiti [acute{s}]aiky[bar{a}]). Ganguli, who always tracks N[bar{i}]lakantha quite faithfully in his renderings of this word, has: "And having taken that iron mace, inlaid with golden plates, which caused the fear of foes and brought on their defeat, he darted it with speed towards the mighty Maniman, menacing (him) and uttering shouts." [64]

The same words and ideas occur again at 3.255.4, where Bh[bar{i}]ma attacks Jayadratha, who is attempting to carry Draupad[bar{i}] off, and we read:

hemacitrasamutsedh[bar{a}]m sarva[acute{s}]aiky[bar{a}]yas[bar{i}]m gad[bar{a}]m pragrhy[bar{a}]bhyadravad bh[bar{i}]mah saindhavam k[bar{a}]lacoditam [65]

N[bar{i}]lakantha is silent on this occasion and Ganguli translates "with a mace entirely of Saikya [sic] iron." Van Buitenen, however, unaccountably departs from his last rendering and sees a kind of sling: "Bh[bar{i}]ma seized his club, its bulge sparkling with gold, the loop wrought of solid iron, and he stormed at the Saindhava who now was summoned by time." This interpretation might be plausible if the text read sarv[bar{a}]yasa[acute{s}]aiky[bar{a}]m gad[bar{a}]m, but as it is, the meaning here must be exactly the same as it was at 3.157.64b. In another description of Bhima's club at 5.50.8a we read saiky[bar{a}]yasamay[bar{i}]m ghor[bar{a}]m gad[bar{a}]m k[bar{a}][tilde{n}]canabh[bar{u}]sit[bar{a}]m, which van Buitenen translates "his steel club adorned with gold." The Vulgate does not read [acute{s}]aikya/saikya here. [66] The same pattern of usage-[acute{s}]aikya/saikya modifying a word derived from ayas in the context of weapons--also occurs at 6.50.21, 104; 6.59.11; 9.10.44; 9.31.37. There is one p assage (7.95.35) that departs slightly from this pattern: in an episode describing S[bar{a}]tyaki fighting with Greek warriors, the word describes metal used for body armor rather than for a weapon. S[bar{a}]tyaki's arrows are so powerful they pass right through the Greeks' brass (or bronze, k[bar{a}]msya) and [acute{s}]aiky[bar{a}]yasa breastplates and clean through their bodies! The metals are praised obliquely in order to praise S[bar{a}]tyaki, for there would be nothing especially remarkable about his arrows penetrating leather or mail body coverings.

[acute{s}]aiky[bar{a}]yas[bar{a}]ni varm[bar{a}]n[bar{i}] k[bar{a}]msy[bar{a}]ni ca samantatah bhittv[bar{a}] deh[bar{a}]ms tath[bar{a}] tes[bar{a}]m [acute{s}]ar[bar{a}] jagmur mah[bar{i}]talam

That [acute{s}]aikya-ayas must generally be steel follows from the fact that it must be a special kind of iron (the occurrence of ayas and its derivatives, unmarked by any other special term for a particular kind of metal such as t[bar{a}]mra or k[bar{a}]msya, [67] creates that presumption) and steel is far and away the most important special form of iron known in the history of metallurgy. As noted above, we have sources outside the text informing us that in ancient India, about the same time the MBh was in the process of being formulated, a highly regarded steel was produced. [68] Herodotos, Ktesias, and other ancient historians tell stories about Indian steel; in one, the fourth-century B.C. Indian king Porus (P[bar{u}]ru) is said to have given Alexander one hundred talents of Indian steel. More telling than these sometimes ambiguous or dubious historical anecdotes is the fact that the Romans imported considerable quantities of this steel at a somewhat later time, [69] and that this steel was the basis of the famous steel of Damascus. [70]

Finally, N[bar{i}]lakantha provides rudimentary support for this interpretation at two locations: for the compound [acute{s}]aiky[bar{a}]yas[bar{a}]ni (varm[bar{a}]ni) at 7.95.35, N[bar{i}]lakantha offers the gloss [acute{s}]onit[bar{a}]yomay[bar{a}]ni, "made of red metal," that is, iron. N[bar{i}]lakantha offers essentially the same gloss to [acute{s}]aiky[bar{a}]yasamayaih ("sarvalohamayaih") at 12.99.20. For reasons that I will make clear shortly, this steel was probably closely related to the wootz steel of modem south India, the steel formerly shipped west to Damascus and on which damask or damascene steel was based. It is not necessarily the case, however, that this steel was "damasked," that is, etched with acid so as to bring out the striking moir[acute{e}] patterns of the crystal structure of the metal.

Passages in Which [acute{s}]aikya/saikya Occurs Without Any Word for ayas

When [acute{s}}aikya/saikya [71] occurs without any derivative of ayas, it seems usually to be a metonymic abbreviation of the fuller noun phrases based on ay[dot{a}]s, and in most of these cases (five of seven occurrences) it seems to be an adjective meaning simply "made of [acute{s}]aikya iron," that is, "steel." Twice, it is clearly an adjective that is used as a noun. N[bar{i}]lakantha, however, explains this usage otherwise: he derives this word [acute{a}]aikya from [acute{s}]ikya (an arrangement of ropes or cords used to carry things, a sling) and takes [acute{s}]aikya to be an adjective that characterizes something as "in a sling," or "part of a sling," at four

different locations. Ganguli consistently follows N[bar{i}]lakantha explanation of the word in his translation. Van Buitenen follows N[bar{i}]lakantha's lead at first, but at 5.50.28 he ignores the [acute{s}ikya-based explanation; also, as noted above, he interprets one of the instances of [acute{s}]aiky[bar{a}]yasa (at 3.255.4) in terms of thi s explanation, something which Nilakantha or Ganguli never do.

According to N[bar{i}]lakantha, in this usage, the word [acute{s}]aikya typically means "resting in a [acute{s}]ikya" ([acute{s}]ikyastha); anything that is typically carried in or suspended from a [acute{s}]ikya can be described as [acute{s}]aikya. In his comment to 2.45.27 N[bar{i}]lakantha explains that "a [acute{s}]ikya is a sling, a k[bar{a}]vad[bar{i}] [Prakrit for a sling, or a pair of slings that hang balanced from the shoulders [72]], made of a leather thong (or strap) and serving as the support for [some kind of] receptacle." [73] At 4.38.34 he describes a [acute{s}]ikya as a similar apparatus carried by two men, probably a pole that supports one or more slings or reticules for carrying a load or loads. [74] N[bar{i}]lakantha is not the first thus to explain [acute{s}]aikya: Franklin Edgerton, in the apparatus to MBh 2.45.27, quotes Devabodha's gloss of [acute{s}]aikya, "bh[bar{a}]nd[bar{a}]dh[bar{a}]rah, a support (or rigging) for a tool, instrument, weapon, etc."--a bandoleer, a "tool-belt," the strap supporting a sword's scabbard or the loop for carrying a club.

This explanation is plausible on its face, and if the derivation of [acute{s}]aikya from [acute{s}]ikya were the only way to understand [acute{s}]aikya, some might consider the matter to be closed. But this explanation does not really fit the usage of the word very well: it separates its free-standing uses from those occurring with ayas, virtually positing the existence of a second word [acute{s}]aikya/saikya. Also, [acute{s}]aikya/saikya never occurs apart from certain kinds of metal objects, and the word seems always to refer to and describe these objects. Even if we grant that N[bar{i}]lakantha's [acute{s}]aikya could represent a specialized, basically military, usage, there is no mention of such bandoleers, belts, or slings anywhere else in the epic, not even in lists of warriors' accouterments. So it is hard to avoid the impression that the usage of [acute{s}]aikya/saikya with ayas is the primary usage, that its occurrence as a free-standing word is merely an abbreviation of that primary usage, and that the notion that free-standing [acute{s}]aikya/saikya describes an object as carried by means of a belt or sling is artificial. As we will see, there is nothing that corroborates the claim that [acute{s}]aikya/saikya should be understood to mean "carried in a belt," "transported by a sling," or "hurled with a sling"--and no descriptions of such practice or use. But when we come to consider the two instances of the word that are not plain, we will find corroboration for a different etymology of the word, one that fits its observed usage like a casting fits a mold.

First the two instances where the word is used syntactically as a noun.

1. At 5.47.97ab Arjuna is reciting indications of the Dh[bar{a}]rtar[bar{a}]stras' impending doom. One of these omens is that saikyah ko[acute{s}][bar{a}]n nihsarati prasanno hitveva j[bar{i}]rn[bar{a}]m uragas tvacam sv[bar{a}]m ("My glad saikya slithers out of its sheath like a snake quitting its worn out skin"). Saikya here must refer to a sword or knife and it must be a noun meaning simply "(my) steel." Van Buitenen renders this with "dagger." [75] The Vulgate text reads here khadga, "sword," instead, so N[bar{i}]lakantha and Ganguli have nothing to say.

2. At 5.50.28, Dhrtar[bar{a}]stra expresses awe at Bh[bar{i}]ma's prowess in battle. He asks, [acute{s}]aiky[bar{a}]m t[bar{a}]ta catuskiskum sadasrim amitaujasam \ prahit[bar{a}]m duhkhasamspar[acute{s}[bar{a}]m katham [acute{s}]aksyanti me sut[bar{a}]h ("How, my boy, will my sons stand the painful blows when Bh[bar{i}]ma swings his infinitely powerful, four-cubit long, six-edged [acute{s}]aiky[bar{a}]?"). [acute{S}]aiky{bar{a}] here is clearly used as a noun meaning club, gad[bar{a}] (a word that occurs explicitly in the very next p[bar{a}]da), and once again it makes good sense to construe [acute{s}]aiky{bar{a}] as a shortened form of [acute{s}]aiky[bar{a}]yasamay[bar{i}], "made of steel" (cf. 5.50.8), which is how van Buitenen renders it ("[that] shattering steel bludgeon"). N[bar{i}]lakantha says otherwise, explaining [acute{s}]aiky[bar{a}] as [acute{s}]ikyasth{bar{a}]("carried in a bandoleer"); it is "fixed in a [acute{s}]kya for fear of gouging the earth if the club falls down," he adds; [76] but this explanation is based on etymology alone. Ganguli sticks with the gist of N[bar{i}]lakantha's etymology here, but rejects his particular explanation, when he translates "he will hurl his mace from the sling."

I move on to passages where [acute{s}]aikaya/saikya does not seem to be used as a substantive. [77]

3. At 2.49.9cd [acute{s}]aikya seems simply to be an adjective: asim rukmatsarum [acute{s}]alyah [acute{s}]aikyam k[bar{a}][bar{n}]canabh{bar{u}]sanam ("[acute{s}]alya [gave him, Yudhisthira,] a gold-ornamented sword that had a hilt of gold," a sword that was [acute{s}]aikya, i.e., made of steel. Following N[bar{i}]lakantha's understanding, Ganguli translates "and [acute{s}]alya, [stood ready to present Yudhisthira] with a sword whose hilt and straps were adorned with gold." Van Buitenen follows the two of them here: "[acute{s}]alya the gold-hilted sword and gold-threaded bandoleer."

4. At 4.38.34cd the adjective appears as saikya in describing Sahadeva's sword: nistrinm[acute{s}]o 'yam guruh p[bar{i}]tah saikyah paramanirvranah ("This heavy, hardened [p[bar{i}]ta], saikya sword, superbly flawless...."). The Vulgate reads s[bar{a}]yakah here, though N[bar{i}]lakantha notes that some manuscripts do read [acute{s}]ikya instead. Van Buitenen appears to translate it "well-tempered," as he renders the above phrase, "this heavy, well-tempered copper sword."

5. At 7.109.10ab Bh[bar{i}]ma's club is again described: sarva[acute{s}]aiky[bar{a}]m catuskiskum gurv[bar{i}]m rukm[bar{a}][dot{n}]gad[bar{a}]m gad[bar{a}]m ("the wholly [acute{s}]aikya club, four-cubits long and heavy, with bands of gold affixed"), where the qualifier sarva[acute{s}]aikya is reminiscent of the sarva[acute{s}]aiky[bar{a}]yasa we saw above and reinforces the sense that [acute{s}]aikya refers to the substance of the gad[bar{a}]. The word [bar{s}]aikya is not read in the Vulgate of this passage, so neither N[bar{i}]lakanth nor Ganguli have anything to offer us here.

6. The Vulgate agrees with the Poona text at 7.138.18a, a passage describing the light reflected from the metal clubs during the Mah[bar{a}]bh[bar{a}]rata's eerie night battle: gad[bar{a}][acute{s}] ca [acute{s}]aiky[bar{a}]h parigh[bar{a}][acute{s}] ca [acute{s}]ubhr[bar{a}]h ("and [acute{s}]aikya maces and gleaming clubs"). Though Nilakantha makes no comment on this occurrence, he has established the [acute{s}]ikya-derivation for instances of [acute{s}]aikya without an ayas word, and Ganguli translates: "And maces twined with strings."

Intermediate Result: [acute{s}]aikya/saikya Refers to Steel and Has Nothing to Do with Slings

Among the seventeen occurrences of the word so far presented, [78] [acute{s}]aikya/saikya refers to body armor worn by Greek soldiers once; [79] once to darts, spears, and battleaxes in aggregation; [80] three times to swords; [81] and twelve times to maces. Among these latter occurrences, ten refers to Bh[bar{i}]ma's gad[bar{a}], [82] one to Duryodhana's, [83] and one refers to clubs generally. [84] In all these occurrences, a single meaning of the word, "special [acute{s}]aikya iron," that is, steel, seems adequate and completely satisfying.

So a check of seventeen of eighteen occurrences of [acute{s}]aikya/saikya in the MBh confirms the basic correctness of the first of B[ddots{o}]htlingk's and Roth's glosses of [acute{s}]aikya, but gives no support to distinguishing a second major meaning for the word, based on N[bar{i}]lakantha's sling-based explanations. In a series of etymological notes published in 1965, Sukumar Sen proposed an etymology for [acute{s}]aikya as a single lexeme, an etymology that is based on the occurrences with ayas- words and which agrees broadly with B[ddot{o}]htlingk's and Roth's first gloss. [85] Sen wrote that the word [acute{s}]aikya "is a apparently a derivative of [acute{s}]ekya- or [acute{s}]ikya- with vrddhi, and probably meant 'stony' or 'hard as stone'. Indo-Iranian *[acute{s}]ik[bar{a}] occurs in Old Persian as [theta]ik[bar{a}] 'stone, rubble,' and probably survives in Old Indo-Aryan as sikat[bar{a}] (also [acute{s}]ikat[bar{a}]) 'sand'." [86] This might be a plausible history of the word, though if peo ple actually thought of [acute{s}]aikya literally as "stony" or "hard as stone," it would be strange for them to praise steel, or even iron, with that adjective. Manfred Mayrhofer rejected Sen's derivation of [acute{s}]aikya in his earlier etymological dictionary of Old Indian, [87] perhaps because he accepted N[bar{i}]lakantha's derivation of [acute{s}]aikya. [88] There is, however, a better explanation of the word.

A New Etymological Explanation of [acute{s}]aikya/saikya

The very first occurrence of [acute{s}]/saikya in the MBh, at 2.45.27, is a special use that illustrates the fundamental meaning and use of the word. Duryodhana is whining about the opulent gifts Yudhisthira received on the occasion of his Royal Consecration:

yan naiva madhu [acute{s}]akr[bar{a}]ya dh[bar{a}]rayanty amarastriyah tad asmai k[bar{a}]msyam [bar{a}]h[bar{a}]rs[bar{i}]d v[bar{a}]runam kala[acute{s}]odadhih 26 [acute{s}]aikyam rukmasahasrasya bahuratnavibh[bar{u}]sitam drstv[bar{a}] ca mama tat sarvam jvarar[bar{u}]pam iv[bar{a}]bhavat 27

I will offer a literalistic translation of this passage after some discussion. Here the word [acute{s}]aikya refers not to a weapon but to a goblet, a k[bar{a}]msya, though this k[bar{a}]msya is not made of brass; [89] nor is it made of steel, even though it is said to be [acute{s}]aikya. The full and correct understanding of [acute{s}]aikya in the passage (and in the MBh generally), however, is assisted greatly by a nearby passage, where the Vulgate text and a few northern manuscripts also read [acute{s}]aikyam. The Poona editor, Franklin Edgerton, chose however the strange-seeming but correct siktam. This [acute{s}]loka, 2.49.15, forms part of a second and longer account of Duryodhana's complaint concerning his cousin Yudhisthira. As he describes Yudhisthira's abhiseka (the "sprinkling" with water that consecrated Yudhisthira as king), Duryodhana says:

up[bar{a}]grhn[bar{a}]d yam indr[bar{a}]ya pur[bar{a}]kalpe praj[bar{a}]patih tam asmai [acute{s}]ankham [bar{a}]h[bar{a}]rs[bar{i}]d v[bar{a}]runam kala[acute{s}]odadhih 14 siktam niskasahasrena sukrtam vi[acute{s}]vakarman[bar{a}] ten[bar{a}]bhisiktah krsnena tatra me ka[acute{s}]malo 'bhavat 15

The Tub (that is) the Ocean presented him (Yudhisthira) the V[bar{a}]runa shell, which Praj[bar{a}]pati had presented to Indra in a previous Era. Siktam with a thousand gold coins, it was well made by Vi[acute{s}]vakarman. He (Yudhisthira) was sprinkled ("consecrated") with that (shell) by Krsna. I despaired at that.

Ganguli translates the Vulgate with the "sling"-idea: "And the Ocean himself brought in a sling that big conch of Varuna which the celestial artificer Viswakarman had constructed with a thousand Nishkas of gold, and which Prajapati had in a former Kalpa, presented unto Indra!" Van Buitenen, silently reading the Vulgate text here, also renders l5ab with N[bar{i}]lakantha's etymology of [acute{s}]aikya in mind: "The sling had been well made by Vi[acute{s}]vakarman out of a thousand gold pieces." But, as I believe I have shown above, the [acute{s}]ikya etymology is wrong, and Edgerton's siktam points to the true meaning of the word [acute{s}]aikya. The root [surd]sic, si[bar{n}]cati, "sprinkle, pour," was used in Vedic texts to describe the casting of molten copper, [90] and it seems clear that this sense of that verb is operating here. The epic text is saying that the conch used to consecrate Yudhisthira was a casting (sikta, "poured [into a mold], cast") made from the melting down of a thousand pieces of gold , "well executed by the divine craftsman." [91]

I propose that the same root lies behind our word [acute{s}]aikya/saikya, And I suggest that [acute{s}]aikya be understood as an historical mutation of saikya, which should in turn be understood as a derivative adjective based on the noun s[acute{e}]ka, "a pouring," from the root [surd]sic, si[bar{n}]cati. Saikya is thus an adjective meaning "(metal) suitable or ready for pouring, casting; or (metal) made by pouring, casting, melting, fusion," and [acute{s}]aikya is simply an allomorph of saikya. That is, [acute{s}]aikya/saikya signifies metal that has been produced by melting, and, sometimes, casting. This word describes wootz steel accurately since it was made by fusion. [92]

The wootz of modern times is made by fusing small pieces of wrought iron in a crucible with pieces of wood that supply carbon to the metal as they burn and are absorbed into the iron. [93] It seems the metal did not always fuse completely, [94] but fusion of the metal is the basic idea and fusion is the basic method by which the iron absorbs the relatively high levels of carbon (over 1.5%, but less than 2%) that make wootz a very fine steel. There is nothing in the modem wootz process that could not have been carried out twenty-five hundred years ago, [95] and, given the ancient reports of Indian steel fabrication and the great likelihood that the saikya/[acute{s}]aikya metal of the MBh is steel, it seems likely that the MBh was referring to this or a similar steel fused in a crucible process. If we see saikya/[acute{s}]aikya as deriving from [surd]sic, si[bar{n}]cati, the word then describes well the metal produced by the wootz process, so I suggest that saikya/[acute{s}]aikya in the MBh means "metal ready for casting," or "molten metal," or "a casting," that is, "metal once molten."

Understood in this way, [acute{s}]aikya was a meaningful alternative to sikta at 2.49.15a. We now return to 2.45.27, the very first occurrence of [acute{s}]aikya in the MBh. We are in the midst of the first, and shorter, account of Duryodhana's jealous recital of the various marvelous presentations made to Yudhisthira in connection with the P[bar{a}]ndava's royal consecration. After recounting how some wealthy brahmins from the west gained admittance to the ceremony by offering beautiful water jars made of gold, Duryodhana speaks the lines we have earlier cited. I submit that an accurate, though purposely stilted, translation of these two [acute{s}]lokas (2.45.26-27) is:

The Tub (that is) the Ocean had brought him Varuna's goblet--a casting of a thousand pieces of gold, studded with many gemstones--such as not even the women of the Gods used to carry mead to Indra. Having seen (all these things), everything of mine now seems like an embarrassment! [96]

This is the only instance in the MBh where [acute{s}]aikya/saikya occurs not in a battle- or weapons context and where it does not clearly refer to an iron, that is steel, implement. The most natural way to construe it here is as a nominal referring to k[bar{a}]msyam, a cup typically made of brass and thus a casting, as siktam refers to the [acute{s}]a[dot{n}]kham at 2.49.14-15. And if [acute{s}]aikya is understood as metal that was "molten, poured, cast," it is natural to construe the following genitive (rukmasahasrasya, "of a thousand of gold") as a genitive of material: The goblet was made of gold pieces that had been fused and cast. [97] The following characterization of the k[bar{a}]msya's appearance, bahuratnavibh[bar{u}]sitam ("studded with many gemstones"), does not give us much interpretive help, but the analogous characterization of

The [acute{s}]a[dot{n}]kha at 2.4.14-15, sukrtam vis[acute{S}]vakarman[bar{a}] ("well made by Vis[acute{s}]vakarman"), strongly reinforces the metal-working sense I am claiming for sikta/saikya/[acute{s}]aikya.

The etymological idea which siktam at 2.49.15 suggests, and which is illustrated with the k[bar{a}msyam...[acute{s}]aikyam at 2.45.26-27, could not better suit all the other instances of [acute{s}]aikya/saikya examined earlier. The adjective refers to metal that has been prepared by the metallurgical technique of fusion, which produces metal suitable for pouring and casting (seka). Any metal subjected to fusion could well be described as saikya, whether copper, gold, brass, iron, [98] or steel. Evidence cited by D.C. Sircar and Sukumar Sen gives some support to this interpretation. [99]

I think we should not lightly disagree with commentators from India's own traditions, but the evidence in the case of saikya/[acute{s}]aikya clearly indicates that the metallurgical understanding of this word altogether disappeared from Sanskrit sometime after the MBh's composition. Why that happened will have to be the subject of further research, as will other interesting issues, such as whether steel castings were made in ancient India, or whether the implements forged from Indian crucible steel were ever etched by ancient Indian sources may contain evidence bearing upon these questions and further careful reading of them with these questions in mind is called for.

As Karnataka has an important part in the history told in the second part of this article, I would like to dedicate this work to several scholars in and around the Central Institute of Indian Languages, Mysore, who offered me a great deal of help and friendship during my stay there in 1988: D. P. Pattanayak (Director of the Institute), the late H. L. N. Bharati, N. Radhakrishna Bhat, C. Pattanayak, Robert Zydenbos, and especially K. Narayan.

(1.) V.S. Sukthankar et al., general editors, The Mah[bar{a}]bh[bar{a}]rata for the First Time Critically Edited, 19 vols. (Poona: Bhandarkar Oriental Research Institute, 1933-66); R. S. Kinjawadekar, ed., [acute{S}]r[bar{i}]manmah[bar{a}]bh[bar{a}]ratam with the Bh[bar{a}]ratabh[bar{a}]vad[bar{i}]pa of N[bar{i}]lakantha, 8 vols. (Poona: Citrashala Press, 1929--36). This study was greatly facilitated by the digitized form of the MBh created by Muneo Tokunaga, Machine-readable Text of the Mahaabhaarata: Based on the Poona Critical Edition, First revised version (V1); Upgrade Version (1.1), October, 1996, produced by Mrs. Mizue Sugita (Kyoto: ftp://ccftp.kyoto-su.ac.jp/pub/doc/ sanskrit/mahabharata, 1996). This version has now been superseded, however, by John Smith's Mah[bar{a}]bh[bar{a}]rata (http://bombay. oriental.cam.ac.uk). (Professor Tokunaga has performed a similar signal service with his R[bar{a}]m[bar{a}]yana: Machine-readable Text of the R[bar{a}]m[bar{a}]yana [Kyoto: ftp://ccftp.kyoto-su.ac/jp/pub/d oc/sanskrit/ramayana,1993].)

(2.) Represented chiefly in the form of King Yudhisthira's debilitating [acute{s}]oka.

(3.) Dilip K. Chakrabarti, The Early Use of Iron in India (Delhi: Oxford Univ. Press, 1992), 124.

(4.) Ellipses like this, that presume a main clause required to complete a relative clause, are not rare in the MBh.

(5.) A check of B[ddot{o}]htlingk's Nachtr[ddot{a}]ge to the first issue of the dictionary failed to turn up any indication that he was aware of N[acute{i}]lakantha's glosses or understood p[bar{i}]ta any differently. Monier-Williams' dictionary simply mirrors the glosses of the Petersburg dictionary in its second edition of 1899.

(6.) K. M. Ganguli, tr., The Mah[bar{a}]bh[bar{a}]rata of Krishna Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text, 11 vols., P. C. Roy, sponsor and publisher (Calcutta: Bharata Press, 1884-96).

(7.) Edward W. Hopkins, "The Position of the Ruling Caste in Ancient India," JAOS 13 (1888-89): 276. S. D. Singh, who, in his Ancient Indian Warfare with Special Reference to the Vedic Period (Leiden: E. J. Brill, 1965), extended and corrected what Hopkins wrote about weapons and warfare, does not mention this common enough attribute of metal weapons either, nor does J. N. Sarkar in his wider-ranging survey, The Art of War in Medieval India (Delhi: Munshiram Manoharlal, 1984).

(8.) J. A. B. van Buitenen, tr., The Mah[bar{a}]bh[bar{a}]rata, 3 vols. (Chicago: Univ. of Chicago Press, 1973-78). See 1.17.16 (vol. 1, pp. 75 and 443).

(9.) A description of a r[bar{a}]ksasa attack on Hanum[bar{a}]n: tasya pa[tilde{n}]c[bar{a}]yas[bar{a}]s tiksn[bar{a}]h sit[bar{a}]h p[bar{i}]tamukh[bar{a}]h [acute{s}]ar[bar{a}]h \ [acute{s}]ar[bar{a}]sy utpalapatr[bar{a}]bh[bar{a}] durdharena nip[bar{a}]tit[bar{a}]h. (5.44.20).

(10.) MBh 1.17.16a. It is worth noting that the first three times the Poona text of the MBh reads the word p[bar{i}]ta in connection with weapons, the Vulgate does not read the word, substituting first tiksna ("sharp") and then d[bar{i}]pta ("shining," or "blazing," though the text does not seem to mean actually flaming arrows). In the first two instances the reading p[bar{i}]ta is either in the process of being replaced in the tradition of these episodes, or is being read into them by some manuscripts. In the case of the Vulgate's reading of tiksna at 1.17.16, where the word modifies the club called the parigha, it is worth noting that in more than one hundred occurrences in the Poona MBh parigha never seems to be modified by t[bar{i}]ksna.

(11.) Van Buitenen, Mah[bar{a}]bh[bar{a}]rata, 1: 75 and 443. The phrase occurs in a description of the battle between the Gods and the Asuras for the amrta at the churning of the ocean: h[bar{a}]h[bar{a}]k[bar{a}]rah samabhavat tatra tatra sahasra[bar{s}]ah \ anyonyam chindat[bar{a}]m [aucte{s}]astrair [bar{a}]ditye lohit[bar{a}]yati [parallel to] parighai[acute{s}] c[bar{a}]yasaih p[bar{i}]taih samnikarse ca mustibhih \ nighnat[bar{a}]m samare 'nyonyam [acute{s}]abdo divam iv[bar{a}]spr[acute{s}]at (1.17.15-16).

(12.) See below for a brief outline of the prior archaeological and linguistic record. For the latter I rely upon the meticulous scholarship of Wilhelm Rau's Metalle und Metallger[ddot{a}]te im vedischen Indien (Wieshaden: Franz Steiner Verlag, 1974). It is worth noting that though we speak of "iron" implements, such implements, especially sharp-edged blades, have actually become a type of low-carbon steel in the process of their being worked at the forge. The outer surfaces of the iron are "carburized," that is, they pick up carbon, in the process of being heated and reheated on the charcoal of the forge. For a description of this process and its importance, see Tamara S. Wheeler and Robert Maddin, "Metallurgy and Ancient Man," in The Coming of the Age of Iron, ed. T. A. Wertime and J. D. Muhly (New Haven: Yale Univ. Press, 1980), 116f. See, too, Theodore Wertime, The Coming of the Age of Steel (Chicago: Univ. of Chicago Press, 1962), 202.

(13.) Bronze seems not to have been widely used in ancient India, because tin (Skt. Tr[acute{a}]pu) was scarce; see Rats, 20-21.

(14.) Encyclopaedia Britannica (Chicago: William Benton, 1961), s.v., "Bronze and Brass Ornamental Works."

(15.) As Chakraharti points out, the word loha is used as the generic term for metals in the Artha[acute{s}][bar{a}]stra; see The Early Use of Iron in India, 107. We see the same general situation in the epics. In the MBh, when referring to metal, loha, or lohita, is sometimes clearly iron (when it refers to implements certainly made of iron [as at 7.14.28; 9.11.20; and, if my arguments about the meaning of the word p[bar{i}]ta are correct, 12.l01.7]), sometimes probably iron (as at 4.57.4; 5.149.82, 157.18, 158.11; 7.154.27, 156.8; and R[bar{a}]m 7.32.42), sometimes ambiguously copper or iron (as at 1.125.23; 5.15.32, 38.13; 12.56.24, 79.22, 283.4), sometimes simply generic metal (as at 12.87.13 and R[bar{a}]m 3.45.41), sometimes ambiguously generic metal, iron, or copper (as at 2.54.28; 4.57.7; 18.2.24, 3.5), or sometimes clearly copper (as at 13.127.18). At 3.22l.63b, in lohitasragvibh[bar{u}]sanah, describing the red god Skanda, lohita could refer to a necklace, or front breastplate worn around the neck, possibly of copper; or it could be an adjective meaning "red," as van Buitenen takes it ("sporting blood-red garlands and jewelry"; he says "blood-red" because lohita in the MBh is also used as a noun meaning "blood"). This list is given to register the range of meaning of loha/lohita, and while I think it captures most of the metal-related instances of the words, it is not intended to be exhaustive.

(16.) It can be inferred for N[bar{i}]lakantha, as for the MBh fifteen hundred or more years earlier, that when the word loha is used in connection with weapons, iron is being referred to, because weapons would normally be made of iron, once iron smelting and iron working became widespread. Iron is much more easily (read cheaply) available than copper, and iron that has been worked at the forge is more effective and durable than is copper. And though iron weapons are not inherently better than bronze weapons, they are much more readily available. Once the Iron Age was fully underway in a particular region the use of copper or bronze for utilitarian purposes would have been unusual.

(17.) As Radomir Pleiner ("The Problem of the Beginning of the Iron Age in India," Acta praehistorica et archaeologica 2 [1971]: 14) and D. K. Chakrabarti (The Early Use of Iron in India, 108) note, the word t[bar{i}]ksna, usually an adjective meaning "sharp," is sometimes used with the meaning steel in the Artha[acute{s}][bar{a}]stra (see, for example, 2.12.15 and 24 and 2.13.48). There may be some odd instances of this usage in the MBh. Perhaps the characterization t[bar{i}]ksnadh[bar{a}]ra, "having a sharp edge," applied to an (ayasmaya) [acute{s}]astra from time to time (e.g., MBh 1.29.2; 6.78.31; 7.31.2; etc.), does show an awareness of the smith's practical knowledge that higher carburization of a weapon's cutting edge (i.e., its transformation from iron to steel) makes a critical difference in the weapon's value. But it is impossible, given the texts I have seen, to distinguish a "steel edge" from a "sharp edge." See N[bar{i}]lakantha's gloss ad 6.43.45, for what it is worth.

(18.) This occurs in a proto-classical up[bar{a}]jati passage describing, for a second time, the P[bar{a}]ndavas departure for the forest. As they set out: presy[bar{a}]h puro vim[acute{s}]atir [bar{a}]tta[acute{s}]astr[bar{a}] dhan[bar{u}]msi varm[bar{a}]ni [acute{s}]ar[bar{a}]m[acute{s}] ca p[bar{i}]t[bar{a}]n maurv[bar{i}][acute{s}] ca yantr[bar{a}]ni ca s[bar{a}]yak[bar{a}]m[acute{s}] ca sarve sam[bar{a}]d[bar{a}]ya jaghanyam [bar{i}]yuh (3.24.3).

(19.) 4.5.24; p[bar{i}]ta is much better attested this time than in the previous two instances, but again the Vulgate reads d[bar{i}]pt[bar{a}]n.

(20.) Van Buitenen, Mah[bar{a}]bh[bar{a}]rata, 3: 533.

(21.) See van Buitenen, 3: 18ff.

(22.) 4.38.26: vip[bar{a}]th[bar{a}]h prthavah kasya g[bar{a}]rdharpatr[bar{a}]h [acute{s}]il[bar{a}][acute{s}]it[bar{a}]h h[bar{a}]ridravarn[bar{a}]h sunas[bar{a}]h p[bar{i}]t[bar{a}]h sarv[bar{a}]yas[bar{a}]h [acute{s}]ar[bar{a}]h.

(23.) The "turmeric hue" mentioned in this sloka is a complication I will discuss below.

(24.) MBh 4.32.6, 38.29, 33, 34, 53.

(25.) Bridge and Raymond Alichin, The Rise of Civilization in India and Pakistan (Cambridge: Cambridge Univ. Press, 1982), 311, 318, and a summary on p. 345. See too Chakrabarti, The Early Use of Iron in India, 60-61, and (with regard to the upper Gangetic valley) 173-74. To the south, the site of Ahar in R[bar{a}]jasth[bar{a}]n provides iron artifacts from strata with carbon-14 dates of 1500 B.C. (Allchin and Allchin, 326) or even earlier (Chakrabarti, 68), and Hallur in Karnataka provides samples from 1150-1030 B.C. (Allchin and Allchin, 329) or even two hundred years earlier (Chakrabarti, 80). For the most recent general account of north India, see F. R. Allchin et al., The Archaeology of Early Historic South Asia: The Emergence of Cities and States (Cambridge: Cambridge Univ. Press, 1995), 65-66; 71-72; 79ff., esp. pp. 83-84. Chakrabarti and other archaeologists are convinced that siderurgical technologies arose in India without diffusion from the west, probably in areas south of the purported Aryan trac k in north India (The Early Use of Iron in India, 21-22). Obviously, to whatever extent these arguments may be true, they only strengthen the case against interpreting the utilitarian metal of the MBh as copper.

(26.) Rau, Metalle, 18-25:

Sehe ich recht, so lassen sich die Aussagen unserer Quellen nur bei Annahme folgender Hypothese miteinander vereinigen und in einem sinnvollen Zusammenhang bringen.

Solange in der [ddot{a}]lteren Zeit dem Edelmetall nur ein Nutzmetall, das Kupfer, gegen[ddot{u}]berstand, war der Ausdruck ayas eindeutig, denn f[ddot{u}]ur Kupfer-Zinn-Legierungen gab es noch kein eigenes Wort. Als jedoch in einer zweiten Phase der Entwicklung das Eisen aufkam, muBte man unterscheiden und tat es in der Weise, daB man einerseits von "rotem" anderseits von "grauem" oder "dunklem" Nutzmetall sprach. Hatte scblieBlich in der dritten Periode Eisen als Nutzmetall das Kupfer endg[ddot{u}]ltig verdr[ddot{a}]ngt, so gen[ddot{u}]gte das Wort ayas nunmehr, um auch ohne den Zusatz [acute{s}]y[bar{a}]ma oder krsna das Eisen zu bezeichnen. (p. 24)

Pleiner's slightly earlier review of early Indian literature (exclusive of the epics) comes to the same general conclusion ("The Problem of the Beginning of the Iron Age in India," 11). Chakrabarti, who does not refer to Rau's work, broadly concurs with Rau's hypothesis as it pertains to post-Vedic literature. He says: "From the [acute{S}]atapatha Br[bar{a}]hmana onwards [ayas] has meant only iron in almost all the texts" (Early Use, 123), after stipulating that "the meaning of ayas in the RV is uncertain, but it is a generic term meaning possibly both copper-bronze and iron. It is unlikely that it meant only iron" (p. 122).

(27.) See note 15 above.

(28.) MBh 4.57.7. At 11.25.14 various accouterments of the five Kekayas are described as t[bar{a}]mra (though this may well be merely a reference to their color; see their lohitadhvaja at 7.22.11) and they wear body armor of gold. See too 9.34.31 and 36.25, which are not military contexts, but Which juxtapose copper vessels to iron ones and other luxuries.

(29.) See MBh 7.31.17. 95.35, 150.10, 24; 8.17.106. 59.26. Most of the time the word refers to goblets, bowls, and milk-pails: See 2.45.26, 49.3; 4. 14.17, 63.47; 12.220.133, 221.59; 13.57.28, 30, etc. The more complicated term virak[bar{a}]msya, rendered as "champion's goblet" by van Buitenen at 1.176.30, seems to indicate (by comparison with 7.2.29 together with S. K. De's critical note at page 1143 of his edition of the seventh book) more a bowl or platter used to deliver a hero's or champion's regalia, often a wreath, garland, or precious necklace (m[bar{a}]l[bar{a}]). Besides the two passages just mentioned, this expression occurs also at 7.87.62. Other terms used for copper-based metals in the Artha[acute{s}][bar{a}]stra--kamsa, [bar{a}]rak[bar{u}]ta and t[bar{a}]la--do not seem to occur in the MBh; see R. P. Kangle, ed., tr., The Kautil[bar{i}]ya Artha[acute{s}][bar{a}]stra, 2 vols. (Bombay: Univ. of Bombay, 1969, 1972), 2.12.23, 17.14; 4.1.35; et passim.

(30.) "These golden-nocked, stone-whetted arrows yellow like turmeric and this quiver marked with five tigers are Nakula's... "(van Buitenen, ad loc.)

(31.) As Hopkins concluded, none of this seems to refer to actually flaming arrows intended to set their targets afire: "The Position of the Ruling Caste," 277.

(32.) It is worth pointing out that as my basic method of locating these passages is by electronically searching the Tokunaga text of the Poona MBh, I may be missing some passages of the Vulgate that were eliminated from the Poona text. I have checked all passages which read t[bar{i}]ksna or d[bar{i}]pta in the Poona text (as these are two of the main variants of p[bar{i}]ta in the Poona text) to see if p[bar{i}]ta is read as a variant for them, but no instances turned up this way.

(33.) N[bar{i}]lakantha ad MBh 4.32.6.

(34.) As the Petersburg Dictionary points out, p[bar{i}]ta occurs, at least in later literature, at the end of some compounds describing persons as "having drunk" something, typically specified by the first element of the compound (this usage is parallel to the occurrence of other forms from the same root being used in similar ways, e.g., somapa, somap[bar{a}]van, somap[bar{i}]tha). The form ap[bar{i}]ta, "one who has not drunk" occurs in the Vulgate of MBh 2.48.38a, though the critical edition reveals it to be a reading that did not occur widely in the tradition. This ambivalence of grammatical voice is not unusual in Sanskrit past participles of this kind, see Altindische Grammatik, 11.2, [ss]432.

(35.) The Artha[acute{s}][bar{a}]stra of Kautilya describes the concoction of tejana powders and tejana oils as means to create the illusion that one's body or some object is burning or glowing (see Artha[acute{s}][bar{a}]stra 14,2.18-26).

(36.) For an historical account of the role of philosophical and alchemical ideas involved in siderurgical metallurgy, see Theodore Wertime, The Coming of the Age of Steel. In note 3 on p. 200 of this work, Wertime notes that Arab writers, discussing the moir[acute{e}] patterns of damascene swords, agreed that these patterns "derive[d] their configuration from the 'water' in the metal."

(37.) It should be observed here that the process of quenching does not enhance the hardness of pure iron, but only of carburized iron, that is, iron that has become steel in some measure. Wertime (The Coming of the Age of Steel, 8) says that rapid cooling, quenching, works well only with iron containing between 0.2 and 1.5% carbon; if this upper limit is correct, it would suggest that [acute{s}]aikya steel, a high-carbon steel produced by fusion of the metal with 1.5 to 1.8, or even 2.0% carbon content (see part two of this article) would not benefit by quenching. As noted above (see note 12), the iron of weapons, at least their cutting edge, was normally, at some point, carburized. See R. F. Tylecote, A History of Metallurgy (London: The Metals Society, 1976), 55-56, for a description of this with regard to Roman smithing practice. Only such carburized iron provides utility superior to that of pure copper and is equivalent to bronze, because pure iron is about as soft as copper (see N. R. Banerjee, The Iro n Age in India [Delhi: Munshiram Manoharlal, 1965], 1-2).

(38.) It is worth noting that this descriptive occurs in the MBh many times more often than does p[bar{i}]ta. Grinding a blade on a whetstone is part of the shaping of a thin-edged blade in the course of its fabrication. With regard to this topic, it may be that N[bar{i}]lakantha and Ganguli point the way to understanding an occasional but puzzling epic word for arrow, namely [acute{s}]il[bar{i}]mukha. In his comment to 4.38.30 N[bar{i}]lakantha says that [acute{s}]il[bar{i}]mukha refers to arrows and swords because they have the mukha (a "face," a "front") that is like the mukha of a [acute{s}]il[bar{i}], a (female) frog. This explanation is highly uncertain, since the word [acute{s}]il[bar{i}] seems unknown apart from the word [acute{s}]il[bar{i}]mukha and N[bar{i}]lakantha's explanation of it in terms of [acute{s}]il[bar{i}], which he glosses as bhek[bar{i}], "frog." (The element [acute{s}]il[bar{i}]- occurs in compound forms with the root [surd]bh[bar{u}], bhavati, which forms mean "turn to stone," as [a cute{s}]il[bar{a}] means "stone, rock." It could well be that [acute{s}]il[bar{i}]mukha is an ancient expression for arrow and simply means a shaft, a weapon, that has a face or mouth of stone, or one that is as hard as stone.) But if we accept for the moment N[bar{i}]lakantha's gloss of [acute{s}]il[bar{i}] as "frog," which sense might this make? The answer is contained in one of Ganguli's N[bar{i}]lakantha-inspired translations at 4.38.30 (B. 4.42.11): "pointed like a toad's head." A frog viewed from above, that is looking down onto its "top" (one sense of mukha) has a triangular shaped body pointed at one end and wider at the other. Many double-flanged "arrow-heads" (the mukha of the arrow) have the same general shape ([Delta]) It is also the case that the edge of a sword blade (the "face" or "mouth" of a sword, another mukha) is given this shape as well through the smith's use of the grindstone. If these speculations were correct, a better understanding of [acute{s}]il[bar{i}]prstha and [acute{s}]il[bar{ i}]mukha in the description of Arjuna's sword at MBh 4.38.30 and 54 might be possible. [acute{S}]il[bar{i}]mukha would then refer to the edge on the sword's "face," the edge mainly used for cutting; [acute{s}]il[bar{i}]prstha refers to the edge on the sword's opposite side. The sword is a double-edged sword with a "frog(-shaped edge)," a [acute{s}]il[bar{i}], on both the front edge (mukha) and back edge (prstha) of the blade.

(39.) 4.59.32, 7.162.41.

(40.) 4.53.26 (which N[bar{i}]lakantha glosses with drsadi tejitaih).

(41.) 6.90.32.

(42.) 9.19.13, etc.

(43.) 5.19.4 (where N[bar{i}]lakantha glosses it with tailam[bar{a}]rjita, "polished with oil"), 6.68.21, 81.6, 83.28, 92.48, 7.154.26: 8.10.18, etc.

(44.) 7.130.27, 9.27.28; oil may be used instead of water for the quenching of steel.

(45.) Wertime (The Coming of the Age of Steel, 192-93) observes that "steel-users over the centuries were likewise in remarkable agreement about the importance of quenching agents, in which they saw some special alchemical or chemical quality." He quotes Roberts-Austen's review of the history of tempering and quenching: "The belief ... in the efficacy of curious nostrums and solutions for hardening steel could hardly have been firmer in the third century B.C. than the sixteenth of our era. Pure cold water is now usually employed for hardening, but it was far too simple a material for many a sixteenth century artificer to employ...." These generalizations are likely to be valid for Indian metallurgists, as well.

(46.) According to Monier-Williams' dictionary there are three varieties of tree, shrub, or creeper to which the name k[bar{a}]k[bar{a}]ndaka might refer: 1) Diospyros tomentosa, Roxb.; 2) Melia Azedarach, Linn. (= Melia Bukayun, Royle); and 3) Carpopogon pruriens, Roxb. (= Mucuna pruriens, DC). The first is one of the main species of ebony (N. L. Bor, Manual of Indian Forest Botany [Bombay: Geoffrey Cumberlege, Oxford Univ. Press, 1953], 45-46), noted for its very dark wood and bark. The second, sometimes called the Persian Lilac or Chinaberry, has panicles of mauve-petaled flowers with dark-purple staminal tubes (Bor, Manual, 253). The third is a twining vine that is most notable for the irritating bristles on its seed pods (see Bor, Manual, 95). But possibly relevant to the current context are this creeper's purplish flowers (J. D. Hooker, The Flora of British India, 7 vols. [Ashford, Kent: L. Reeve & Co., 1827-97], 2: 187). All three species are widely distributed in India and any one of the three would m ake sense in the context, though the mauve flowers of the Melia seems to me to the clearest opponent of the turmeric color registered in p[bar{a}]da a. Ganguli translates 666cd-67ab (B. 3.188.75cd-76ab) with "and some resemble in tint the filaments of the lotus and some are purple, and some are yellow as turmeric and some of the hue of the crow's egg." Van Buitenen renders 66ab-67cd with "some darkling like blue lotuses, others like white lotuses or fibers, others yellow or turmeric-ochre, the color of spiders and red lotus petals, and vermilion" (van Buitenen gets "spiders" from the fact that the word k[bar{a}]k[bar{a}]ndaka signifies "spider" in the [acute{S}]u[acute{s}]ruta samhit[bar{a}]: according to B[ddot{o}]htlingk and Roth and Monier-Williams).

(47.) The Poona text reads saikya rather than [acute{s}]aikya at 4.38.34, 5.47.97 and 5.508; [acute{s}]aikya is a weakly attested variant of saikya in these three cases, one that turns up basically in the commentators. The Vulgate reads neither [acute{s}]aikya nor saikya in these three passages. In a number of passages in the critical edition where the editor reads [acute{s}]aikya, the word is not widely attested, though it is sufficiently attested and is the correct reading. In a number of the passages where [acute{s}]aikya is read by the Poona edition, saikya is an attested variant.

(48.) 3.157.64, 255.4; 5.50.8 (saikya); 6.50.21, 104, 59.11; 7.95.35; 9.31.37.

(49.) 12.99.20.

(50.) 9.10.44.

(51.) 2.45.27, 49.9; 4.38.34 (saikya): 7.109.10, 138.18.

(52.) 5.47.97 (saikya), 50.28.

(53.) 2.49.15, where Edgerton's correct reading of siktam proves extremely important, and 5.47.17.

(54.) "1. perhaps '[steel] that has been damascened'."

(55.) "2. perhaps 'a kind of shot-hurling sling'."

(56.) His glosses to 7.95.35 and 12.99.20, both of which are passages where [acute{s}]aikya modifies a word derived from ayas.

(57.) See below.

(58.) The derivation of the word wootz is unclear. The OED appears to follow Hobson-Jobson (Henry Yule and A. C. Burnell, A Glossary of Colloquial Anglo-Indian Words and Phrases, and of Kindred Terms, Etymological, Historical, Geographical and Discursive [London: J. Murray, 1886; new edition by William Crooke, London: J. Murray, 1902; reprinted, Calcutta: Rupa and Co., 1986], s.v. wootz), which source cites an article published in 1795 as the first attestation of the word in English and then observes: "The word has never since been recognised as the name of steel in any language, and it would seem to have originated in some clerical error, or misreading, very possible for wook, representing the Canarese ukku (pron. wukku) 'steel.'" (Chakrabarti, The Early Use of Iron in India, p. 1, n. 2, accepts this explanation, citing Hobson-Jobson.) Another suggestion noted by Hobson-Jobson would derive the word from the Canarese opposition of uchcha and nicha (from Sanskrit ucca and n[bar{i}]ca, "high and low," "superio r and inferior"), postulating a Kannada phrase such as "uchcha kabbina," "superior iron." J. S. Jeans (Steel: Its History, Manufacture, Properties, and Uses [London: E. & F. N. Spon, 1880], 294-95), adopts this explanation without citing any source. Hobson-Jobson indicates a preference for the first of these two explanations, but it seems plausible that some European traveler might have conflated ukku and uchcha and thus invented "wootz," which gained currency in discourse where there were no native experts to correct it.

(59.) See Cyril Stanley Smith, A History of Metallography: The Development of Ideas on the Structure of Metals Before 1890 (Chicago: Univ. of Chicago Press, 1960), ch. 4: "European Attempts to Duplicate Damascus Steel," pp. 25-29. See, too, Wertime, The Coming of the Age of Steel, 199-211. Other good notices of this history are found in Chakrabarti, The Early Use of Iron in India, 1-22 and 156-58, and Joseph Needham, The Development of Iron and Steel Technology in China (Cambridge: The Newcomen Society, 1964), 45.

(60.) At MBh 2.45.27, 49.9, the Vulgate's reading of Poona 5.47.21, and Poona 5.50.28. At 3.157.64 N[bar{i}]lakantha offers yet a third explanation of [acute{s}]aikya/saikya.

(61.) In their note to R[bar{a}]m. 5.44.20 Goldman and Sutherland Goldman quote an observation of G. C. Jhala, the editor of the Sundarak[bar{a}]nda of the R[bar{a}]m[bar{a}]yana, to the effect that the word [acute{s}]ita ("whetted, honed") is redundant in a line (in a context involving weapons) that contains the word t[bar{i}]ksna ("sharp"). But there really is no redundancy, as [acute{s}]ita does not primarily mean "sharp," but rather specifies one of the smith's operations in making a blade sharp, t[bar{i}]ksna. As I tried to show above, the smith's quenching a properly carburized blade allows an even sharper blade.

(62.) Bh[bar{i}]ma, having gone to pick flowers for Draupad[bar{i}], had been set upon by r[bar{a}]ksasas and yaksas, whom he then slaughtered. In the course of this battle, the r[bar{a}]ksasa Manimat, Kubera's dear friend, had pierced Bh[bar{i}]ma with a spear.

(63.) Second in order of appearance, that is.

(64.) B. 3.160 [159 in the Roy edition].71. The editor of the Poona text, Sukthankar, notes in the apparatus for 64b that a few (unspecified) manuscripts read saikya for [acute{s}]aikya.

(65.) The apparatus for this passage reveals that the southern tradition is joined by several Kashmiri and Bengali manuscripts in reading saikya for [acute{s}]aikya here.

(66.) Here [acute{s}]aikya is very slightly attested, and several northern manuscripts read neither [acute{s}]aikya nor saikya. Ganguli's text reads only [bar{a}]yas[bar{i}]m ghor[bar{a}]m gad[bar{a}]m, but he translates "steel" anyway, perhaps in light of the mention of Bh[bar{i}]ma's club (gad[bar{a}]m) described as [acute{s}]aiky[bar{a}]m t[bar{a}]ta catuskiskum sadasrim amitaujasam below in the same adhy[bar{a}]ya at 5.50.28. Van Buitenen translates this as "...that boundlessly powerful, shattering steel bludgeon that measures four cubits and has six edges," and Ganguli again gives "steel" for [acute{s}]aikya.

(67.) See t[bar{a}]mr[bar{a}]yasa at 9.36.25 and k[bar{a}]msy[bar{a}]yasa at 7.31.17 and 8.59.26. The word k[bar{a}]msya by itself does not always signal the metal of which it is made.

(68.) These accounts are retailed in various secondary works. The most comprehensive, detailed, and critical one I have seen is R. Pleiner's review in "The Problem of the Beginning of the Iron Age in India." Good too is D. K. Chakrabarti, The Early Use of Iron in India, 116-19. Still useful is R. J. Forbes' Metallurgy in Antiquity: A Notebook for Archaeologists and Technologists (Leiden: E. J. Brill, 1950); see pages 438-39. But see Pleiner's criticisms of Forbes, "The Problem," 17 and n. 89.

(69.) "Reports and remarks dating from the first century A.D. and onwards expressly celebrate Indian iron and steel": Pleiner, "The Problem of the Beginning of the Iron Age in India," 17. See E. H. Warmington, The Commerce between the Roman Empire and India, 2nd ed. (London: Curzon Press, 1974 [1st ed. Cambridge: Cambridge Univ. Press, 1928]), 257-59, and 290-91. Relevant too are pp. 265-66, concerning Rome's exports of lead, tin, copper, and brass to India. The most important sources, Pliny's Historia Naturalis and the anonymous Periplus Mans Erythraei are from the first century A.D., and there is some uncertainty and controversy over whether Pliny's reference to the "Seres" refers to a province of China or to the Cheras of South India (see Warmington, 257 and Pleiner, ibid.).

(70.) According to Forbes, 439, Damascus became an important center for the working of Indian iron and steel when Diocletian [emperor from AD. 284-305] founded his armament factories there, and it remained important until Tamerlane carried those factories off to Khorasan and Samarkand in 1399. See Smith, A History of Metallography, ch. 3: "The Damascus Blade" pp. 14-25. For notices on Arab sources discussing iron and steel, and their observations on the superior metals of India and China, see Wertime, The Coming of the Age of Steel, 200, n. 3. Concerning wootz, Hobson-Jobson includes this romantic flourish: "The article [i.e., wootz] was no doubt the famous 'Indian Steel,' the [sigma][acute{i}][delta][eta][rho]o[varsigma]' I[nu][delta]t[kappa][grave{o}][varsigma] [kappa][alpha][acute{i}] [sigma][tau][acute{o}][mu][omega][mu][alpha] of the Periplus, the material of the Indian swords celebrated in many an Arabic poem, the alhinde of old Spanish, the hundw[bar{a}]n[bar{i}] of the Persian traders, ondanique of M arco Polo, the iron exported by the Portuguese in the 16th century from Batical[acute{a}] in Canara and other parts."

(71.) It is worth noting here explicitly what was left implicit above, that the Vulgate never reads saikya in lieu of [acute{s}]aikya. At 4.38.34b it reads s[bar{a}]yakah paranirvranah for saikyah paramanivranah; at 5.47.97a it reads khadgah for saikyah; and at 5.50.8 it reads ast[bar{a}]srim[bar{a}]yasamay[bar{a}]m for [acute{s}]aiky[bar{a}]yasamay[bar{i}]m.

(72.) Pandit Hargovind Das T. Sheth, P[acute{a}]ia-sadda-mahannavo: A Comprehensive Prakrit-Hindi Dictionary (Varanasi: Prakrit Text Society, 1963). D.C. Sircar's Indian Epigraphical Glossary (Delhi: Motilal Banarsidass, 1966) cites the word [acute{s}]eka, which he says is the same as Sanskrit [acute{s}]ikya. Sanskrit also knows the synonymous word [acute{s}]ic.

(73.) N[bar{i}]lakantha ad Vulgate 2.49.27: [acute{s}]aikyam varatr[bar{a}]mayam p[bar{a}]tr[bar{a}]dh[bar{a}]rabh[bar{u}]tam [acute{s}]ikyam 'k[bar{a}]vadi" iti prasiddham \ tatra sthitam p[bar{a}]tram [acute{s}]aikyam.

(74.) N[bar{i}]lakantha ad Vulgate 4.42.16 [acute{s}]ikya iti p[bar{a}]the purusadvayenohyam[bar{a}]n[bar{a}] s[bar{a}]mg[bar{i}]ti prasiddh[bar{a}] tadyogyah.

(75.) "My dagger serenely pops out of its sheath..." I prefer "sword" because it fits the snake imagery better, though the text probably does not imagine a long sword.

(76.) [acute{s}]aiky[bar{a}]m [acute{s}]ikyasth[bar{a}]m adhahpatanena bh[bar{u}]mivid[bar{a}]rana[acute{s}]ankay[bar{a}] [acute{s}]ikye sth[bar{a}]pit[bar{a}]m (Nilakantha ad 5.50.28). Does he here imagine the kind of [acute{s}]ikya carried by two men that he describes at 4.38.34?

(77.) The Vulgate includes a third instance of [acute{s}]aikya being used as a substantive. At 5.47-17ab (B. 5.48.21ab) the Vulgate text reads (in a description of Bh[bar{i}]ma] [acute{s}]aikyena n[bar{a}]g[bar{a}]ms taras[bar{a}] vigrhnan yad[bar{a}] chett[bar{a}] dh[bar{a}]rtar[bar{a}]strasya sainyam, which is another instance where [acute{s}]aikya is used as a noun signifying a weapon (though, here it would have to refer to a sword). N[bar{i}]lakantha explains [acute{s}]aikyena here as [acute{s}]ikyasadr[acute{s}]ena p[bar{a}][acute{s}]ena (quoted in B[ddot{o}]htlingk and Roth as p[bar{a}]trena), though he notes others read the verse in such a way that [acute{s}]aikya must mean "sword." Ganguli translates "by his nooses strong as iron."

(78.) Counting too the Vulgate-only passage at 5.47.17, see above, note 77.

(79.) At 7.95.35.

(80.) 12.99.20.

(81.) To one given to Yudhisthira at 2.49.9, to Sahadeva's at 4.38.34, and to Arjuna's at 5.47.97 (the word is saikya in both these latter instances).

(82.) At 3.157.64, 255.4; 5.47.17 (variants only; B. 5.48.21), 5.50.8, 28; 6.50.21, 104, 59.11; 7.109.10; and 9.10.44.

(83.) When he emerges from his hiding place in the lake at 9.31.37.

(84.) Where these and other weapons reflected the light of torches during the night battle at 7.133.18.

(85.) "Indo-Iranica," Indian Linguistics 7 (1965): 259-61.

(86.) Ibid., 260.

(87.) Manfred Mayrhofer, Kurzegefa[beta]tes etymologisches W[ddot{o}]rterbuch des Altindischen, 4 vols. (Heidelberg: Carl Winter, 1953-80), s.v.[acute{s}]iky[grave{a}]m, "rope-sling for carrying things." This dictionary did not have a separate article for [acute{s}]aikya, and one wonders what Mayrhofer thought of [acute{s}]aikya in [acute{s}]aiky[bar{a}]yasa.

(88.) Kurzegefa[beta]tes ..., s.v. s[acute{i}]kat[bar{a}]. Like Sen, Mayrhofer believes there is a connection between Skt. s[acute{i}]kat[bar{a}] and Old Persian [theta]ik[bar{a}]-, *sik [degrees](see Mayrhofer's second etymological dictionary, Etymologisches W[ddot{o}]rterbuch des Altindoarischen, 2 vols. [Heidelberg: Carl Winter, 1986-96], s.v. s[acute{i}]kat[bar{a}]).

(89.) K[bar{a}]msya derives from kamsa, which originally meant a cup, goblet, or other vessel. The word kamsa developed into the name for the special copper-based metals that were often used for such vessels, namely bronze and especially brass, and k[bar{a}]msya arose as the word referring both to such metal and to things ordinarily made of that metal. Occasionally in the MBh, as here, the word simply means (metal) vessel.

(90.) See Wilhelm Rau, Metalle und Metallger[ddot{a}]te im vedischen Indien, 37, n. 44. Unaware of this, Edgerton suggested that "Perhaps it [siktam] means "sprinkled," in the sense of "spangled, beset," in his edition of The sabh[bar{a}] Parvan (Poona: Bhandarkar Oriental Research Institute, 1944), 508.

(91.) The casting of precious metals was not the common way of fabricating items from them, but it was not unknown; see R. J. Forbes, Metallurgy in Antiquity, 134 and 157, fig. 40. Another way (distinct from Edgerton's proposal; see above, note 90) to understand siktam apart from my suggestion, would be to take it to mean "sprinkled, poured," in the sense that the shell had had a thousand gold coins "poured" into it. Would there then be a suggestion that Yydhisthira was subsequently sikta with the gold coins?

(92.) For now I leave unaddressed the question whether [acute{s}]aikya/saikya steel might actually have involved the casting of some or all the implements described as [acute{s}]aikya/saikya. One wonders particularly about maces (gad[bar{a}]), which would seem too massive to have been forged from a single cake of steel (as swords ordinarily were). Four-cubit long clubs, such as Bh[bar{i}]ma wielded, would have required welding at the forge if they were not cast. In this connection it is interesting to note that at least in Malaysia Bh[bar{i}]masena and Arjuna have been regarded as armourers as well as warriors. Both are said to be depicted in a fourteenth-century relief known as the "Candi Sukub forge scene," with Bh[bar{i}]ma working the metal and Arjuna pumping the bellows; see Edward Frey, The Kris: Mystic Weapon of the Malay World (Singapore: Oxford Univ. Press, 1988), 5-6.

(93.) There are many descriptions of this process, beginning with travelers' accounts in the eighteenth century and continuing in metallurgical discussions today. Forbes' Metallurgy in Antiquity is the most useful source, as its accounts are woven into a general account of the history of siderurgy; see pp. 409-11 for a description of the crucible or fusion process of steel making in general (with special attention to wootz) and pp. 436-40 for a discussion (now quite outdated but still of general value) of the overall history of iron and steel in India. The latter passage contains this description of wootz making:

The [newly smelted] bloom [the bloom is the chunk of wrought iron that results from smelting and which was then commonly cut and worked at the forge] is then cut up into small pieces to pack the crucible better. These crucibles are made of refractory clay. Mixed with the charred husks of rice, the leaves of Asclepios gigantea or Convulvulus laurifolia and the wood of Cassia auriculata they are packed tightly. That charge is seldom over one pound mixed with the proper amount of wood chopped in small pieces. The mouth of the crucible is stopped with a handful of tempered clay, which is rammed closely to exclude air. When the plugs are dry, 24 crucibles are built up in a furnace arch and kept covered with charcoal and now subjected to a fire aided with blast air for about 2 1/2 hours. Then the furnace is allowed to cool and broken crucibles yield the wootz in the form of a cake. (p. 438)

The resulting cakes of high-carbon steel (about 1.5-2.0% carbon), which were often shipped to Rome, Damascus, Persia, and other centers of Middle Eastern metalworking, had superior qualities of hardness, and complex crystalline patterns which varied from one melt and one crucible to another. These crystalline variations were the basis of the striking "watering," or moir[acute{e}] patterns that Middle Eastern metal-workers brought out by etching, and which caused the resulting blades to be known as "damask," or "damascene." Chapter three of Smith's History of Metallography, "The Damascus Blade," provides a wealth of information on the connection between wootz and the damascene products. Finally, chapter five of Chakrabarti's Early Use of iron in India, "The Pre-Industrial Iron-smelting Tradition," contains a comprehensive and detailed account of the entire subject of wootz making which is well complemented by his chapter one ("Research on Early Indian Iron").

(94.) Smith, The History of Metallography, 21-22.

(95.) "Until recent years tribal peoples in many parts of the subcontinent continued to smelt their own ores by methods which are ... relatively simple, and we may expect that the picture of the modern Agaria smiths of Central India ... differs little from that which archaeology has recently reconstructed" (Allchin and Allchin, The Rise of Civilization in India and Pakistan, 309). A little further on they say of travelers' descriptions of wootz steel making, "in default of any other evidence, we may conclude that this was the method employed anciently also" (ibid., 310).

(96.) MBh 2.45.27 is, of course, the first instance where N[bar{i}]lakantha presents his [acute{s}]ikya, "sling," explanation. He writes: [acute{s}]aikyam varatr[bar{a}]mayam p[bar{a}]tr[bar{a}]dh[nar{a}]rabh[bar{u}]tam [acute{s}]ikyam k[bar{a}]vad[bar{i}ti prasiddham \ tatra sthitam p[bar{a}]tram [acute{s}]aikyam ("'[acute{s}]aikya': A [acute{s}]aikya is a thing made of leather that carries a receptacle [a pot, a bowl, a jug, a bottle; or perhaps merely a pad of leather, or a loop or ring]; (it is) what is generally known as a k[bar{a}]vad[bar{i}] [Prakrit for sling; see n. 72]. A receptacle that rests in one of them [i.e., in a [acute{s}]ikya] is called [acute{s}]aikya"). In essence N[bar{i}]lakantha says "[acute{s}]aikya' means a receptacle in a [acute{s}]ikya," and he must have the goblet in mind as the p[bar{a}]tra that it is being carried in while suspended from a pole borne by two men (presumably because it is very heavy). On the basis of N[bar{i}]lakantha's commentary here van Buitenen translates [a cute{s}]aikya with "reticule," rendering 27ab: "it [the brass goblet] was carried in a reticule of a thousand gold pieces and adorned with many precious stones." As to Ganguli, the Vulgate text inserts a line be[grave{t}]ween 2.45.26cd and 27ab: that entirely alters the context of 27ab and Ganguli, basing himself on N[bar{i}]lakantha's argument that [acute{s}]aikya meant a kind of "pot, or vessel," renders it with "jar": "a thousand jars [made] of gold, all well adorned with numerous gems." As I noted earlier, N[bar{i}]lakantha seems merely to be following an old tradition which has been preserved at least in Devabodha's gloss of bh[bar{a}]nd[bar{a}]dh[bar{a}]ra at 2.45.27.

(97.) In discussing the earliest evidence for brass objects Forbes mentions a report, attributed to Pseudo-Aristotle (De mirab. ausc., ch. 49), that the Achaemenian Persian emperor "Darius is said to have possessed an 'Indian' cup which looked like gold but had a disagreeable smell, which points to brass" (Metallurgy in Antiquity, 279).

(98.) In general, smelting furnaces in the ancient world could not pump enough oxygen onto the charcoal to raise the temperature to the 1540[degrees] C required to melt iron. The fusion of iron could only be accomplished when sufficiently refractory clay crucibles were available. See Forbes, Metallurgy in Antiquity, 407-8, for a discussion of the question whether and where cast-iron was known in the ancient world. Cast iron that is produced by fusing iron at the time of smelting, a process that produces a metal that has so much carbon, 2% to 4% [see T. A. Wertime 'The Pyrotechnic Background," in the Coming of the Age of Iron, 13], that it is impracticably brittle.

(99.) Sen (see n. 85 above) and D.C. Sircar quote two words from Bengal that suggest the epic word saikya/[acute{s}]aikya had a wider Indian history than is apparent from the epic alone. There is a seventh-century inscription from central Bengal containing the word sekyak[bar{a}]ara which Sen, consistently with the etymology he is proposing in his article, glosses as "stone-engraver [grater than] metal-engraver." Sen says this word is the source of the Bengali word sekar[acute{a}] "gold-or silversmith." Sircar cites sekyak[acute{a}]ra in his Indian Epigraphical Dictionary with the meaning "a brazier; secondarily, an engraver," and he refers to "Bengali se[dot{n}]kr[bar{a}], a goldsmith."