Vous êtes sur la page 1sur 26

239

ALTAR, AMULET, ICON

ALTAR, AMULET, ICON


TRANSFORMATIONS IN DHRA AMULET CULTURE, 740-980

Paul Copp

Des feuilles damulettes bouddhiques, inscrites ou imprimes du texte du


Mahpratisar-dhra, ont t dcouvertes Dunhuang et dans des tombes
dans diverses rgions de Chine. Elles datent du VIIIe, du IXe et du Xe sicle et
elles rvlent une histoire complexe et trans-asiatique, de pratiques amultiques.
Les premiers exemples crits la main ou partiellement crits la main
et partiellement imprims vhiculent des pratiques dorigines indienne
et dAsie centrale consistant personnaliser les amulettes dincantation en y
inscrivant les noms et les images de leurs porteurs. Ces exemples dmontrent
galement une relation troite avec les critures du Mahpratisar-dhra
lui-mme, dont le texte semble avoir inspir la production de ces amulettes.
Les exemples xylographis plus tardifs sont issus dune culture technique qui
a perdu ses liens directs avec les premires pratiques une culture qui est
entre en Chine par les routes terrestres de commerce avec les rgions occidentales. Ces dernires feuilles tmoignent galement de nouvelles relations
aux conventions de la xylographie contemporaine chinoise et aux nouvelles
iconographies arrives en Chine par les routes maritimes. Cet article examine galement la nature de lecacit des amulettes la lumire des dessins
dautels trouvs Dunhuang.

Beginning in the 1940s, amulets of the Buddhist Wish-Fulfilling Spell (Da suiqiu
tuoluoni ; Mahpratisar-dhra)1 silk or paper sheets inscribed
with texts and images rolled and enclosed within armlets or other carrying cases
have been found on corpses in tombs in China om Chengdu to Luoyang
. The amulet sheets as well as their containers were objects of deep resonance
within Indic and Sinitic ritual traditions, they as well as objects closely related
to them are described in Buddhist texts of many periods and are instantiated in
various material contexts not only in China but across South and Central Asia. In
this essay I focus on the sheets, which occur in manuscript, xylograph, and part
manuscript/part printed form. To date, approximately eighteen examples, dating to
the eighth, ninth, and tenth centuries, have been discovered in tombs and in the

Foshuo suiqiu jide dazizai tuoluoni shenzhou jing


(Mahpratisar-dhra-stra), Taish shinsh daizky (100 vols., eds. Takakusu
Juir , Watanabe Kaigyoku , et al., Tky: Taish issaiky kankkai
, 1924-1932; hereaer T) no. 115 See also T no. 115
Cahiers dExtrme-Asie 17 (2008): 239-264
cole anaise dExtrme-Orient, Paris, 2010
Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

240

Paul Copp

Dunhuang trove.2 Aside om their intrinsic interest, the sheets are important as
testimonies of the culture of Buddhist amulet usage in this period.
The Visual Structure of the Amulet Sheets
The designs on the sheets tend to be concentrically displayed in three zones
(Fig. 1): an outer zone consisting of images of ritual implements, Sanskrit syllables,
and/or deities; a middle range consisting of the inscribed incantation itself, whether
in transliterated Chinese or in Indic script; and in the center an image either of
a tableau of the empowerment of the donor, featuring a vajra-wielding deity, or
a simpler image of the eight-armed bodhisattva Mahpratisara, sometimes also
including an image of the donor in a worshipful pose to its side. The arrangement
can be either of concentric circles or squares.
Taken as a whole, the tableaux on the sheets appear to be diagrammatic representations of the rites of dhra incantation found in surviving ritual manuals,
specifically those of the Wish-Fulfilling Spell. Their centers render the focal point
of the rite either the action of empowerment or the icon of the incantationdeity; the outer borders figure the ritual setup, which takes the form in Buddhist
incantation texts of both physical implements and their bodily and mental/sonic
counterparts, visually represented on the amulets in the forms of mdras or seedsyllables; linking them we have the dhra itself. Because of this ritual imagery,
scholars such as Ma Shichang who has provided the best overview of extant
amulet sheets have taken to calling these objects dhra madalas or dhra
altars.3 Indeed, as we will see, the visual structures of the sheets do not only reflect
medieval prescriptions for the setting up of ritual spaces: they also closely mirror
sketches for the creation of such physical as well as visual spaces found among the
Dunhuang documents, including one that seems to have been specifically intended
for use in setting up altars for the ritual inscription of the Wish-Fulfilling Spell.
Let us begin our examination of the visual structures and components of the
sheets with their border zones. In order to understand their arrays of objects and
figures, we must first look to scriptural accounts of rites for the inscription of the
Wish-Fulfilling Spell and their descriptions of the designs of its amulets. One immediately notices that this outer ring of imagery is not prescribed in them. Baosiweis
I exclude om this accounting, and this essay, those that were found in stupas and thus
clearly treated as relics for veneration rather than as personal apotropaic devices. I also limit my
discussion to amulets of the Wish-Fulfilling Spell; other incantations are featured on some amulets
(see, for example, the Dunhuang sheet known as Pelliot Sanskrit #1). Three excellent studies
of the amulet (and reliquary) sheets are Ma Shichang , Da suiqiu tuoluoni mantuluo
tuxiang de chubu kaocha , Tang yaiu 10 (2004),
527-581; Jean-Pierre Drge, Les premires impressions des dhra des Mahpratisar, Cahiers
dExtrme-Asie 11, 25-44; and Katherine R. Tsiang, Buddhist Printed Images and Texts of the
Eighth-Tenth Centuries: Typologies of Replication and Representation, in Matthew Kapstein
and Sam Van Schaik, eds, Esoteric Buddhism at Dunhuang: Rites and Teachings for This Life and
Beyond (Leiden: Brill, 2010), 201-25
Ma Shichang, art. cit.,

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

241

Fig. 1: Incantation amulet made for Madame Wei. Ink and pigment on silk.
21.5 cm x 21.5 cm. Mid eighth century? Copyright Yale University Art Gallery,
Hobart and Edward Small Moore Memorial Collection, bequest of Mrs. William
H. Moore. xliii.004).

(d. 721) translation of the Wish-Fulfilling Spells scripture, the Incantation


Scripture Spoken by the Buddha of the Dhra of Great Sovereignty Whereby One
Achieves What One Seeks (Foshuo suiqiu jide dazizai tuoluoni shenzhou jing
),4 mentions only that the spell is to be written down and that
an image reflecting the nature and needs of the donor is to be made in its center.
The text states, for example, that

If a monk is to wear [the amulet], then in the center of the incantation draw a single
Adamantine Spirit adorned with jewels. Below him make a monk kneeling in the hu
style with his palms together. The Adamantine One has his hand upon the top of the
head of the monk.5
T 115
Foshuo suiqiu jide dazizai tuoluoni shenzhou jing T 1154, 20: 642a7-

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

242

Paul Copp

Fig. 2: Incantation amulet made for Jing Sitai. Ink on paper. 35cm x 35 cm. Mid
eighth century? Discovered near Xian in 1967. After Jean-Pierre Drge Les
premires impressions des dhra des Mahpratisar, Cahiers dExtrme-Asie
11 (1999-2000), fig. 9.

The scriptural basis for the images that surround the spell becomes clear, however, when we compare the full versions of these accounts with what seem to be
the earliest known fully surviving sheets (Figs. 1, 2).6 Although the matches are
not perfect, the surrounding images bear a close resemblance to the objects that,
according to the Scripture, are to be either placed physically or depicted in drawings
within the ritual space in which the sheets are to be fashioned. The vases adorning the four corners of both sheets are likely representations of the bottles filled
with agrant water to be physically placed at the four corners of the ritual space.
Similarly, the weapons sporting ribbons arrayed along the sides of the sheets are
no doubt pictures of the knives, swords, and halberds with flowing silk ribbons
the practitioner is told to draw upon the ground of the ritual space. Furthermore,
the Scriptures iunction to make [a] lotus blossom and within its center draw
Ma Shichang takes the sheet found at Fenghao Rd. in Xian to be the oldest
(Ma Shichang, art. cit., 528-530.). Based on the iconography at its center, however, I think this
sheet probably dates om a later period.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

243

a burning pearl was also taken up by the amulet makers, who executed images
representing exactly such lotus-and-pearl combinations.
Although the Scripture and the amulets do not correspond in every detail,
the resemblances are striking enough to warrant an interpretation of the amulet
sheets as conscious renditions of the amulet-inscription rites. This correspondence
becomes even more interesting, and opens further historical insights into the world
of the creators and buyers of the sheets, when we read the parallel account om
the slightly later Bukong (a.k.a. Amoghavajra, 705-774) version of scripture
into the mix. In the Baosiwei translation, it is clear that the silk-adorned weapons
and fiery pearls are to be drawn onto the ground of the ritual site; but the Bukong
version makes it equally clear that the images are to be made instead on the amulet
sheets themselves.7 Like the Baosiwei version, the Bukong text prescribes specific
renderings for particular donors or situations, but the images it describes are more
elaborate. For example, in the case of an amulet to be made for a man who seeks
sons (zhangfu qiu zi zhe ), the text states that one should draw jeweled
mountains at each of its four corners:
To the four sides of the real-word (zhenyan ) [that is, the spell] you should draw
various kinds of seals, and further draw lotus flowers either two or three or four or
even five lotus flowers. These flowers should all be open and spread out, their eight petals
set close together, their stalks bound with silk ribbons. On each flower draw a trident;8
on the tridents further [draw] tied ribbons of silk. Further, draw a yuefu ax, also on
the lotus flowers. Further, [draw] white lotuses upon some you should draw a sword,
on others draw a conch. The lotuses that you draw should all [float] within a treasure
pool. Since a man is to wear it, you should not draw a child [in the center of the sheet,
as is done with the amulet for a woman seeking sons given just previous to this one],
but instead draw a god adorned with various gems.9

For the most part, this prescription provides the template for all the amulets
described in the succeeding parts of the text, though there are variations in detail. In
the case of amulet sheets to be made for monarchs, for instance, their four corners
feature not bejeweled mountains but the four Heavenly Kings. This detail seems
to have been well known, at least at Dunhuang: the Four Heavenly Kings adorn
an amulet sheet dated to 980 found at Dunhuang (Fig. 3), though in other details
this sheet is unlike any described in either version of the Scripture.
Although the images prescribed in the two versions of the Scripture do not match
exactly those on any surviving amulet I am aware of, the shi in medium om ritual
ground to amulet sheet is striking and clearly of great significance for understanding medieval amulet-making traditions. It is dicult, however, to understand this
change precisely. Given that most of the amulets show no sign that their makers
followed the Bukong version of the Scripture indeed, what evidence we have
Pubian guangming qingjing zhisheng ruyi baoyin xin wunengsheng damingwang dasuiqiu
tuoluoni jing T 1153, 20: 623c-624a.
Taking sai as an abbreviated form of sanchaji , trident (Skt. Trila). I am
grateful to a reviewer of this article for this suggestion.
T 20: 624a-b.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

244

Paul Copp

Fig. 3: Xylographed incantation amulet, Stein Painting 249 (Ch. Xlviii. 004). Woodblock
print, ink on paper. 45.7 cm x 31.75 cm. Woodblock print, ink on paper. Dated 980. British
Museum 1919, 0101, 0.249 (Ch. xliii. 004).

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

245

suggests that they favored the Baosiwei version it seems unlikely that the shi
in design was a direct result of the new versions appearance on the scene. Instead,
particularly in view of other deviations om scripture that I will discuss below, it
seems far more likely that the Bukong version reflected the visual structure of the
amulet sheets then in circulation, and not the other way around; in other words,
the Bukong version was an outgrowth of, rather than a blueprint for, contemporary
Tang and Central Asian traditions of amulet usage.10
Altars and Amulets
Preparatory drawings found at Dunhuang, apparently for use in the construction
of altars, help both to clari the nature of the amulet images and to place the amulet
sheets within wider traditions of ritual imagery and practice. Comparison with altar
diagrams such as Stein Painting 172 (Fig. 4), in fact, confirms what our readings
of versions of the Scripture of the Wish-Fulfilling Spell strongly suggested: namely,
that the images on the sheets are not simply vague indexes or evocations of ritual
spaces, but, to a great extent, precise renderings of their layouts and contents. This
is clear not only om the placement of the images the three zones characteristic
of the amulet sheets are present as well on the altar maps, though in necessarily
more generalized form but especially in the small details of their outer sections.
A comparison of the border zone of Stein Painting 172 with those of the amulets
reveals that many of the images plants, fungi, weapons, gems, and wheels are
nearly identical.11 More than this, the dynamic, even activated, nature of the images
rendered on both Stein Painting 172 and the amulets is striking. The swords, for
example, are not simply inert metal objects placed in a certain way: the coronas
of flames surrounding them visually suggest that these swords are represented as
being active in their particular ecacies, in keeping with the ritually active nature
of the amulets and, I will suggest, the altars as well. Similar considerations apply
to the wheels, gems, and other motifs. Were the diagrams simple maps for the
placement of objects and images on the ground of the daochang , such visual
flourishes would hardly have been necessary indeed, on other altar sketches such
as Stein Painting 234 (Fig. 5), a map of an altar for use in the recitation of the
Superlative Spell of the Buddhas Crown (Foding zunsheng tuoluoni ;
That these amulet sheets were in use not only in Tang and later China is made clear by
the discovery of two sheets in a tomb at Astana, Turfan (Xiiang), whose design diers markedly
om those found within China proper. For a discussion of these sheets, see Ma Shichang, Da
suiqiu tuoluoni mantuluo tuxiang de chubu kaocha, 532-53
That the diagrams seem to have been used in constructing actual ritual spaces or images
is suggested, as Sarah Fraser has pointed out, by the corrections written on to Stein Painting 172,
evidently by the diagram makers supervisor (Sarah E. Fraser, Performing the Visual: The practice of Buddhist wall painting in China and Central Asia, 618-960 [Stanford: Stanford University
Press, 2004] 153). These corrections for example, put the deity in the center and move the
mudra to the le, in Frasers translation (153) are probably traces of a dialog between the
artists and the ritual master.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

246

Paul Copp

Fig. 4: Altar diagram, Stein Painting 172 (Ch. 00189). Ink on paper. 58.5 cm x
57.5. Late ninth century? After Roderick Whitfield, Art of Central Asia: The Stein
Collection of the British Museum (Toky: Kdansha , 1982), v. 2: figure 79.

Uavaya dhra), the images of lamps and water bottles were drawn much more
simply. They are not aflame with the potencies of ritual transformation; they are
merely lamps and bottles depicted in the flatter utilitarian ame of everyday action.
The incipient state of our understanding of all these images imposes the need for
caution in our claims about them; yet we seem to have two dierent types of altars
here, each with its distinctive type of map. The first type, including Stein Painting
234, apparently depicts a space to be constructed out of actual objects placed in
certain ways on the ground or the floor of a temple. The second type, including
Stein Painting 172, is composed of far more intricate images whose components are
anything but simple. Stein Painting 172 seems, in fact, to be a drawing of a drawing: a sacred image of the sort indicated by the term maala in its later usage in
the Esoteric Buddhist tradition (and in modern English usage). The altar space
does not provide structures for the physical performance of a rite, but is rather an

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

247

Fig. 5: Altar diagram, Stein Painting 234 (Ch. 00186). Ink on paper. 44 cm x
30.5 cm. After Roderick Whitfield, Art of Central Asia: The Stein Collection of
the British Museum, v. 2: figure 81.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

248

Paul Copp

image whose very presence oers blessings and divine presence. Altars of this type
are potent objects rather than fields of human ritual action.12
The close relationship between the Dunhuang altar diagrams and the amulets
is clear in the instructions found in the two Chinese versions of the Scripture. It
is clearest in the case of the Bukong version, which instructs its reader to draw an
altar on the surface of his amulet sheet rather than on the ground. But the iconic
nature of the altar is suggested as well in the more influential Baosiwei version
by the absence of instructions in the text to enter the altar space physically. If we
consider Stein Painting 172 as a representation of the kind of altar alluded to in the
Baosiwei text, the iconic nature of the sutras altar becomes even more apparent:
for the image on the diagram provides no space at all for the practitioner to situate
himself within. Thus it could not be the image of a space for actual ritual activity,
such as those described in many other ritual manuals. Some dhra manuals, in
fact, eoin the practitioners to copy out the spells to the sides of altars.13 Perhaps
we may make sense of this otherwise confusing prescription by suggesting that
renderings like Stein Painting 172 were intended to be placed to the side of the site
where the ritual enactment of the dhra was to occur.
Sarah E. Fraser has suggested that the employment of the verb to use (yong
) rather than to draw/paint (hua ) in the corrections made to the details of
Stein Painting 172 indicates that the images were not to be painted or drawn, but to
be rendered using sand in the manner of the sand maalas common in Tibetan
Buddhism today.14 This is indeed very likely, though the material used was probably
not sand but the substance known in a range of ritual texts simply as powder (fen
).15 The diagram would seem, in fact, to be the earliest material evidence for the
powder altars that had been in use in Buddhist ritual in China since at least the
middle of the fih century, and which seem to have come especially into vogue during
the tenth century the age, most likely, when Stein Painting 172 was executed.16
Most importantly for our purposes, the iconic nature of the altar maala to
be rendered near the dhra practitioner suggests that altars and amulets were in
key ways functionally as well as visually identical. Once a tan comes to
Fraser also draws close connections between diagrams such as Stein Painting 172 and
the amulets particularly Stein Painting 249 and notes the talismanic function of the latter
set. She sees a high/low distinction in the cultures of the two types, with the handwritten
altar diagrams part of highly individualized and sophisticated monastic practices of contemplation, and the amulets particularly their printed variety as popular expressions of prayers
(drawing a link between the printed amulets and other printed icons, on which see below) that
were mass-produced and important in the economic life of the monasteries (Fraser, Performing
the Visual, 157-8).
See, for example, Yizi foding lunwang jing ; T 951, 19: 226b, and
Wu foding sanmei tuoluoni jing ; T 952, 19: 264b.
Fraser, Performing the Visual, 15
See, for example, Da fangguang pusazang Wenshushili genben yigui jing
, T 1191, 20: 876c-877b; Foshuo shengbaozang shen yigui jing ,
T 1284, 21: 350b and 351c; and Foshuo yuqie dajiao wang jing , T 890, 18: 560a-b.
See Da Song sengshi le , T 2126, 54: 240b-c.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

249

be conceived as an iconic representation of a ritual space rather than as an actual


space in which a practitioner enacted a ritual, the ecacies of its presence on the
ground beside oneself are clearly very much like those of its presence on ones arm,
or tucked within ones sleeve. Distinctions between amulets and altars collapse in
that particular ritual ame wherein sign and signified are one: an image of a sacred
space is the sacred space, with all its attendant potencies; an image tucked in a ring
of a vajra coruscating with blessings is none other than the very thing it depicts.
Transitions in Amulets and Amulet Craft
Returning now to the amulets, we come next to the Indic syllables of the spell
itself, whether transliterated into Chinese characters or rendered (with varying skill)
in Indic script in the intermediary zones of the sheets. The texts of the incantations
fall into two groups: an earlier and a later. In the earlier examples, manuscripts and
half-xylographs, the spells and central images oen take a personalized character,
featuring the names of the donors who wore them and the imagery of personal
empowerment; in the later examples, which are all xylographic, the incantations are
never personalized, no images of personal empowerment occur, and (interestingly,
since they more oen quote the Scripture) the central images pertain to iconographical traditions quite dierent om those evidenced in both extant Chinese versions
of the Scripture of the Wish-Fulfilling Spell.
This shi in the format of the amulets seems in part the result of a shi in
the techniques of their production. Whereas the older amulets, manuscripts and
half-manuscripts, carried forward what we will see to have been a nearly panAsian amulet tradition, the later examples, fully xylographic, were shaped within
the maturing Chinese cra tradition of woodblock printing, which seems to have
absorbed certain elements of Chinese Buddhist painting practice. This transition
saw the loss of the vibrant and far-traveled custom of personalizing the potency
of the spells themselves, and of designing the images and texts entirely as blessings for the specific individuals who wore them. With the transition to full block
printing, the names of the donors were, in keeping with the conventions of most of
the other Chinese merit-making images of the age, relegated to colophons located
at the lateral or bottom margins of the sheets, away om the main action of the
incantation. Coincident with this marginalization of the donor was a shi in the
iconography of the central images away om those inspired by the Scripture to
images of the multi-armed forms of the Bodhisattva of Wish Fulfillment (Suiqiu
pusa ) that were then spreading across Buddhist Asia, both as sculptures
and as descriptions in ritual and iconographic manuals for visualization.
Manuscripts and Half-Manuscripts
A striking dierence between the earlier and later amulet sheets is the presence
on the earlier versions of donors names within the texts of the incantations. On

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

250

Paul Copp

extant sheets of this kind we encounter such names as Madame Wei (Wei Daniang
), Iron-head Jiao (Jiao Tietou ), and Jing Sitai (Fig. 6) written
into blanks le in the texts of the incantations.17 This feature reflects Indic dhra
amulet practices and points to a nearly pan-Asian set of practices extending well
into China, a connection that is relevant quite aside om the obvious and trivial
fact that their dhras are all in Sanskrit, or pseudo-Sanskrit, and usually written
in Indic script. To be sure, the inclusion of the names of ritual beneficiaries within
religious texts had by the late Tang long been common in various Chinese traditions; yet, the claim advanced by some scholars that the insertion of the donors
name within the text of the dhra he or she wore was a Chinese innovation based
on such native traditions is incorrect.18 As a case in point, dhra charms om
Gilgit, now northern Pakistan small slips of paper handwritten with dhras
that Oskar von Hinber has called protective charms (Schutzzauber) display
precisely the same forms of name-insertion within their incantation texts. They
are particularly relevant for this study because at least one of them bore the text of
a dhra labeled Mahpratisar.19
In other ways, however, the personal dhra amulets om Gilgit were quite
dierent om the Chinese specimens. Their physical and visual characteristics
dier: the Gilgit charms are simple slips of paper devoid of imagery, bearing only
the text of a spell and the name of the intended target of its protective influence;
whereas, as we have seen, the Chinese examples are multi-layered hybrids of text
and image. Their social loci also dier: surviving examples om Gilgit were all made
for royalty, while all Chinese amulets found so far belonged to men and women
of humbler station. There is at least one textual record, however, that suggests
that dhra amulets were worn as well by members of the imperial house: in 758
Interestingly, two of the amulets whose spells contain the names of donors seem to
date om transitional moments in the history of the production of the amulets. The first, made
for a man named Jiao Tietou, was handwritten and painted, yet its central image is an iconic
representation of the bodhisattva Mahpratisar (see below for the significance of these images).
The second was made for a certain A-luo . Like late amulets, it is a woodblock print, yet
like early amulets, its central image is of the donor receiving divine blessings, and the name of the
donor has been written in to the text of the spell. Both amulets are contained in the collection
of the Xian Forest of Stone Tablets Museum. See Xian Beilin Bowuguan , Xian
Beilin Bowuguan (Xian: Shaanxi renmin , 2000), 152-15
See Ma, Da suiqiu tuoluoni mantuluo tuxiang de chubu kaocha, 530. For related
claims for the Chinese origins of the amulet practices prescribed in the stra, see Xiao Dengfu
, Daojia Daojiao yingxiangxia de Fojiao jingji (Taipei: Xinwenfeng
, 2005), 773; and Xiao Dengfu, Daojiao shuyi yu miao diai (Taipei:
Xinwenfeng, 1994), 194-
See Oskar von Hinber, Die Palola his, Ihre Steininschriften, Inschriften auf Bronzen,
Handschriftenkolophone und Schutzzauber: Materialien zur Geschichte von Gilgit und Chilas
(Antiquities of Northern Pakistan, Reports and Studies, vol. 5, Harald Hauptmann, ed.; Mainz:
Verlag Philipp von Zabern, 2004), 16-17, and Namen in Schutzzaubern aus Gilgit, Studien zur
Indologie und Iranistik 7 (1981): 163-17 In my forthcoming monograph on dhra practice I
discuss connections between the Chinese incantation amulets and similar textual amulets found
not only in Gilgit, but in Gandhra and Khotan, as well.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

251

(Qianyuan 1), in the midst of the chaos of An Lushan (d. 757) and
Shi Simings (d. 761) rebellion against Tang rule, Bukong, translator of the
second Chinese version of the Scripture of the Wish-Fulfilling Spell, gave to the Tang
Emperor Suzong (711-762; r. 756-762) one text of the Great Wish-Fulfilling
Dhra, written in Sanskrit (Fanshu Dasuiqiu tuoluoni yiben
). Though it might initially seem that this gi consisted of the Sanskrit text of
the Scripture itself, perhaps the one whose re-translation Bukong later oversaw, we
next read that Bukong urged the emperor to meditate on this seal and wear it on
your belt (nian jian er dai zhi ), promising that benefits to his turbulent
reign would result.20 This detail strongly suggests that the text was an amulet
sheet of the Wish-Fulfilling Spell.
Like the images that adorn the outer rims of the early sheets, those at their
centers only rarely match scriptural accounts for how the amulets should be rendered. The early manuscripts and half-manuscripts, in their depictions of vajrawielding gods oering blessings to the bearers of the sheets, are evidence instead
of local customs that, while they took place in apparent knowledge of scriptural
guidelines, adapted them in the process of creating individualized amulet sheets.
The Yale amulet, dated by Ma Shichang to the mid-eighth century, wholly hand
drawn and inscribed without any printed elements, bears at its center a tableau of
empowerment similar to one prescribed in the Scripture (Fig. 1). Dominating the
scene, slightly o-center and filling its portion of the square om top to bottom,
is a majestic figure seated with legs crossed on an upturned flower that, convention indicates, must be a lotus. The figures upright half-naked torso and rather
feminine head are each wreathed in a fiery aureole; the latters dark oval of fire
contrasts strikingly with the lighter flames of the larger and rounder mandorla set
slightly behind and below it. A faint mark on the divinitys forehead might be a
third eye. A crown of jewels and leaves (?) adorns its head; it wears a necklace, an
armlet high on its le arm, and bracelets on its wrists. Both according to scriptural
accounts and in terms of the action of the scene, its hands figure the clear meaning
of the tableau. In its le hand it holds a large vajra-club, marking the divinity as a
Vajra-Wielder (Vajradhara or Vajrapi) clearly a representation of the Vajra
(or Adamantine) Spirit (jingang shen ) mentioned in the Scripture. Its right
arm is extended with hand resting just above the black hair of a kneeling woman in
robes, her palms together in an attitude of supplication. This woman is, of course,
Madame Wei, the bearer of the amulet. An inscription flanking the scene reads,
Received and borne in her 63rd year; the bearer (shouchizhe ) Madame Wei
wholeheartedly makes oerings.
The presence of the Adamantine Spirit recalls the multiple configurations
for amulet sheets listed in the Scripture. However, when the details of the image
are checked against those prescriptions it becomes clear that the amulet does not
really match any of them. The two prescriptions for female bearers contained in
extant editions of the scripture both of which are for women seeking to become
Daizong chao zeng sikong dabian zheng guangzhi sanzang heshang biaozhi ji
T 2120, 52: 829b.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

252

Paul Copp

pregnant call for the depiction of either


a child bedecked with gems or a blackfaced Mahkla, the Great Black God
(daheitian ), like Vajrapi a major
figure in the then-developing dhra
and Esoteric traditions. The only amulet
described in extant editions of the Scripture that calls for a vajra-bearing figure
is intended for monks. The relevant lines
read: If a monk is to wear it, then in the
center of the incantation draw a single
Adamantine Spirit adorned with jewels.
Below him make a monk kneeling in the
hu [or Central Asian] style with his
palms together. The Adamantine One
has his hand upon the top of the head of
the monk.21 This description, clearly, is
very close to the image on the Yale sheet
change monk to female donor and
we have a precise description of its central
image. The Adamantine One is indeed
adorned with jewels, and its right hand
reaches out above the head of the one it
Fig. 6: Jing Sitai. Detail of Figure 2.
blesses.
This iconographical near miss
becomes more interesting, and is revealed
as further suggestive of Tang practical traditions, when we notice that at least one
and perhaps three other Tang-era amulets bear strikingly similar images at their
centers, all of which come slightly closer to the details of the textual description
than does the image on the Yale sheet. In one of them, the one inscribed on behalf
of Jing Sitai (Fig. 2), the resemblance is especially clear.22
Hand-drawn in the center of the spell stamped upon the Jing Sitai amulet is a
scene that, like the Yale sheet, includes the image of a donor kneeling in the hustyle, with one knee on the ground. But, like Madame Wei, Jing was no monk: he
is shown wearing a headscarf as well as a beard. An Jiayao and Feng Xiaotang
, in their analysis of the sheet, notice the discrepancy between the Scripture
and this particular tool of its enactment. They suggest that Jing, or the makers of
his amulet, understood seng , the word for monk used in the Scripture, to refer
not only to monks but to laymen as well.23 But since they do not oer any textual
2 Foshuo suiqiu jide dazizai tuoluoni shenzhou jing, T 1154, 20: 642a7-
2 For these other sheets see Ma, Da suiqiu tuoluoni mantuluo tuxiang de chubu kaocha,
536-7, 56
2 An Jiayao and Feng Xiaotang, Xian Fengxi chutu de Tang yinben Fanwen tuoluoni
jingzhou , Kaogu , 1995: 8

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

253

Fig. 7: Incantation amulet model? Ch. 00187. Ink on paper. Size unknown.
After Aurel Stein, Serindia: detailed report of explorations in Central Asia and
westernmost China carried out and described under the orders of H. M. Indian
government, v.2, (Oxford: Clarendon Press, 1921), Ch. 00187.

parallels to such an understanding of seng, it seems unlikely that their proposal is


correct. How to understand the relationship between the image and the scriptural
account thus remains an open question.
The appropriateness of the image of an Adamantine Spirit for monks amulets
must have been known, at least at one time, in the region near Dunhuang, even
though no amulets bearing this image have been discovered there; for Aurel Stein
discovered an amulet model (or perhaps a map for a ritual space as we have seen
this distinction is not always clear) that contains instructions for the making of an
amulet to be worn by monks (seng dai ). At its center, within the heart of
an incantation (yu zhouxin ), it features the image of an Adamantine Spirit
(Fig. 7).24 Though this document om the far northwest should not be taken as
sure evidence of what artisans around the capitals were aware of, it is highly suggestive of contemporary artisanal lore. Yet we are le to wonder about the use of the
2 A closely related, and even more strikingly suggestive, amulet diagram is contained on
Stein no. 4690. Unfortunately, I discovered this manuscript too late to incorporate it into this
essay, but I will take account of it in my forthcoming monograph on material dhra practice.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

254

Paul Copp

Adamantine Spirit image in amulets for non-monastics. For most non child-seeking
people of Tang China, in fact, no amulet prescription given in the Scripture would
likely have seemed an obvious choice. As the translation of an Indic text (or at least
as a text couched in those terms), the accounts of possible adult donors are described
in terms of the four basic classes of traditional Indian society brhmaa, katriya,
vaiya, and udra; the only exceptions are kings, merchants, and pleasure seekers.25
This may have helped make Vajrapi the deity of choice for non-monastics in the
region of the imperial capital. Clearly, at any rate, the rich accounts of the Scripture,
though they were at least on some level in play within the culture of the amulet
makers, did not fully govern artisanal practice. This will become even more obvious below when I consider the wholly extra-scriptural images of the eight-armed
Bodhisattva of Wish-Fulfillment.
Like the iconic altar figured on Stein Painting 172, though in a dierent
way, the Adamantine Spirit Amulet diagram thus seems to blur the distinction
between amulet and altar. The original purpose of this sheet is even more dicult to
understand than that of the fuller altar diagrams discussed above. My analysis here
remains provisional, pending deeper research into the ritual culture of Dunhuang
Buddhism of the ninth and tenth centuries. The image on the sheet (Fig. 7) depicts
a ritual space whose inner half is labeled pure (qingjing ) suggesting that
the document is a map of an actual ritual arena and whose ground throughout
is marked o by rectangular spaces, apparently providing areas for the placement,
or representation, of specific implements similar to those on Stein Painting 172
and the amulets discussed above. But the center of the sheet, a circle set within an
outline image of an eight-petalled flower (no doubt a lotus), itself enclosed within
a larger circle, casts doubt on such an easy identification of the nature of the sheet.
As already noted, the text contained within the innermost circle (the site, of course,
on the amulets of the central image) consists of instructions for rendering a Vajra
Spirit within the heart of an inscribed incantation though whether that image
is to be drawn on a sheet (or on the ground of an iconic altar), or whether one is to
do the drawing in a ritual space like the one indicated by the image, is impossible to
determine for certain. The brief text reads: Adamantine Spirit: in the center draw
one26 within the heart of an incantation; for monks to wear (jingang shen: zhong
huazuo yizhe yu zhouxin; seng dai : ; ). It seems unlikely
that both the spell and the deity were to be rendered within the small space provided
within the flower (unless the spell were very brief indeed); thus, the sketch would
seem to be a map for the creation of a space in which amulets should be created.
But this takes us quickly into speculation; we need to know more about the cultural
position of altar sketches such as this one and Stein Painting 17

2 There is, as well, the category basic, which, since it does not call for any image at all,
does not seem to have been in play in the making of the amulets that remain to us.
2 The inscription is dicult to read in parts. What I have taken as huazuo yizhe
(paint one) might instead read jinzuo (make only one).

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

255

Fig. 8: Xylographed incantation amulet, discovered in Luoyang. Woodblock


print, ink on paper. 38 cm x 29.5 cm. Dated 926-927. After Jean-Pierre Drge
Les premires impressions des dhra des Mahpratisar, Cahiers dExtrmeAsie 11, fig. 9.

Xylographs, Episodes, and Icons


The absence of names inserted within the spells of nearly all the full xylographs
(Figs. 3 and 8) marks them as quite a dierent sort of objects om the earlier group.27
The printed sheets also oen contain personalizing features, but those features
reflect practices then popular in China of appending colophons or inserting
cartouches with names and dates and adding images of worshipful donors rather
than those of the Indic amulet traditions sketched above. Further, the new central
images and the place or, more usually, the absence of the donors within them
helped to transform the character of the sheets. Adapting the work of Wu Hung
and others (see below footnotes 35-37), I see this shi in the nature of the amulet

2 See footnote 17 for two exceptions.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

256

Paul Copp

as one om episodic to iconic images. This shi was in part the byproduct of the
rise of a new iconography and its place in contemplative rather than empowerment
practices; it seems also to have been abetted by the shi to full block printing and
the subsumption of the amulets under image conventions familiar to the printers.
The sheets now included colophons containing not only the names of the donors
but at times those of the printers as well, and sometimes not only the dates of the
initial carvings but also the dates of individual printings. New, too, were the quotations (or seeming quotations) om the Scripture making explicit the powers of the
amulet. Such quotations reflect the cartouches on hanging silk icons objects of
longstanding popularity that were doubtless well known to the xylographers. These
changes, as noted above, stripped the amulets of their formerly special character,
including their continuities with broader Asian amulet traditions, and transformed
them into rather standard Chinese Buddhist images. On such images which
were, of course, still considered potent ritual presences the names and images of
mortal Buddhists were merely attached; they were not figured as the direct objects
of the icons blessings.
Though identical in basic structure with the earlier manuscripts, the xylographs
were thus very dierent kinds of objects. The images on both kinds of amulets
quali as iconic, a term indicating the ritually active nature of certain images,
but here a further distinction must be made. No longer in themselves enactments
of ritual action (in the case of the amulets that of empowerment) the prints were
shaped according to a logic of contemplation and prayer that had become normal
for the Buddhist imagery of the age. I will return to this point later in the essay.
Key features of this new mode of depiction were the new images placed at the
centers of the sheets. All xylographed amulet sheets feature at their centers full
ontal images of the dhra bodhisattva Mahpratisar bearing his characteristic
implements lariat, ax, sword, trident, wheel, book, banner, and vajra in his
eight arms.28 This follows a template that was found in images in India, Indonesia,
and China, and which predates the four-headed form of the deity described in
scriptural texts beginning only in the tenth century. As noted above, the central
images of these later amulet sheets bear no relation at all with the prescriptions
of the Mahpratisar Stra. Neither Chinese translation of that text describes the
eight-armed figures that are found on the amulets and in sculptures across Asia
in this period.29 The earliest textual accounts of such forms do, however, seem to
2 As does one singular manuscript, which once belonged to a man named Iron-Head
Jiao, discovered in a tomb near Xian. See Ma Shichang, Da suiqiu tuoluoni mantuluo tuxiang
de chubu kaocha, 528-9, and 56
2 Although as a referent to iconographic practice it is highly ambiguous, there is a possible trace of the eight-armed figure in an episode in one of Bukongs biographies (one that,
oddly enough, echoes in strikingly precise terms a tale preserved in his own translation of the
Scripture; T 1153, 20: 621b-622a). In the episode, which is contained in the biography of Bukong
composed by Zhao Qian (fl. 766-774), while on an ocean voyage Bukong encounters a
nearly fatal storm. To calm the storm he chants the Wish-Fulfilling Spell in the following way:
taking up in his right hand a five-pronged bodhi-mind mallet [that is, a vajra] and in his
le a volume (jia [= qie ]) of the Scripture of Praj[pramit], Mother of the Buddhas, he

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

257

date om early in the same period as the amulets. They are found not in Chinese
translations of scriptures but in manuals om Sino-Japanese ritual traditions,
including those reported by Kkai (774-835), once a disciple within Bukongs
own ritual lineage. The earliest unambiguous trace of the presence in China of
the eight-armed Bodhisattva of Wish Fulfillment is contained in Kkais Notes
on the Secret Treasury (Hizki ), a text that, as Ryichi Ab states, consists
of one hundred agmentary sections of [Kkais] handwritten record of the oral
instructions he received om Huiguo , one of Bukongs disciples.30 Among
those instructions we find the following notes for the depiction of the Great
Bodhisattva of Wish-Fulfillment: Deep yellow in color, with eight arms. The upper
le hand [holds] a golden wheel of fiery brilliance atop a lotus; in the next hand
is a Brahmanic book;31 in the next a jeweled banner; and in the next a lariat. His
upper right hand holds a five-pronged vajra; in the next is a halberd; in the next a
jeweled sword; and in the next a hook-ax.32 Kkais notes seem to date to 805, the
year of Huiguos death, which strongly suggests that this form of the bodhisattva was known in China by at least the late eighth century. The description was
maintained in Japanese ritual practice for at least another century and a half, and
in at least one text it became part of the guidelines for visualization practice. In the
Shingon , monk Shunnys (890-953) Contemplations of the Ritual Spheres
of the Essential Deities (Yson djkan ), the practitioner is told, in the
section concerning Contemplation of the Ritual Sphere of the Great Bodhisattva
of Wish Fulfillment, that in this meditation the syllable pra transforms into an
Indic book, which then transforms into the bodhisattva complete with its eight
arms, each holding the items as described in the Hizki.33 The eight-armed figure
does not appear in a Chinese translation of a stra until the tenth century the
Scripture of the King of the Great Teachings of Yoga (Foshuo yuqie dajiao wang jing
), translated by Fatian (a.k.a. Faxian , d. 1001) and then
extended [his arms] and performed the rite of empowerment (zuofa jiachi ), chanting
the Wish-Fulfilling [Spell] (Da Tang gu dade zeng sikong dabianzheng guangzhi Bukong sanzang
xingzhuan , T 2056, 50: 292c. A slightly dierent version is collected in Song gaoseng zhuan , T 2061, 50: 712b). Though Bukong had only
two hands to employ, he seems here to be mimicking the image of the spells deity, which as
we have seen is depicted employing these very objects. In fact, as I will discuss below, Kkais
Hizki , supposedly the notes of instructions given by Bukongs student Huiguo ,
describes the deity as holding a vajra in one of his right hands and a book in one of his le a
fact that may not simply be coincidental.
30. Ryichi Ab, The Weaving of Mantra: Kkai and the Construction of Esoteric Buddhist
Discourse (New York: Columbia University Press, 1999), 124- See 487-8 n. 60 for Abs discussion of the nature of this work.
3 That is, literally, a Brahmanic case (fanqie ) in which a collection of inscribed
leaves was kept.
3 Hizki, 2: See also Ishida Hisatoyo , Mandara no kenky (Toky:
Tky butsu , 1975), 44-45, for a discussion of this form of the Zuigu bosatsu
within the context of the Womb Maala. I am grateful to the reader of this essay for the
journal who pointed me to Ishidas discussion.
3 Yson djkan, T 2468, 78: 50c.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

258

Paul Copp

only in a four-headed form reflecting an apparently later South Asian iconographic


transmission that seems to have had no impact on East Asian developments.34
By the ninth century, icons of the single-headed eight-limbed Mahpratisar,
the form present on the xylographs and in East Asia more generally, had been
spreading across Buddhist Asia for decades. They appear in sculptural form in
eastern India and central Java, as well as, of course, in painted and printed images
in Tang, Five Dynasties, and early Song China. Gerd Mevissen has suggested that
surviving metal and stone statues and stone temple reliefs om India and Java
may date om as early as the eighth century.35 The most securely datable of these
images seems to be a large stone relief carved in the centre of the north-eastern
wall of the Candi Mendut, a temple in Central Java of the late eighth or early ninth
century, about 3 km to the east of Borobudur.36 Assuming that the iconography
originated in India, an early or mid eighth century date for its inception seems
easily possible. For the purposes of this essay, however, it is enough to note that
we have here a new, apparently southward, orientation of Chinese amulet culture,
which, combined with native Chinese printing conventions, fundamentally altered
the older, westward-oriented culture of the amulets.
The basic categorical distinction between episode and icon, which has been
delineated (in dierent ways and using dierent terms) by Wu Hung,37 Stephen F.
Teiser, 38 and Victor H. Mair,39 proves very useful in understanding the dierence
between the two kinds of amulet sheets. Wu Hung, whose terminology I follow
here, divides the images partly in terms of whether their designs are self-contained
(episodes) or whether their structures assume a direct relationship between the
icon and the viewer. Of course, one might object that in the case of amulets sealed
within armlets or other carrying cases, all of the sheets were in a very literal sense
self contained, and that none could reasonably be thought to have had any kind
3 Foshuo yuqie dajiao wang jing, T 890, 18: 568a-b. Though he does not mention this
text, Mevissen has tracked these images to the eleventh-twelh century iconographic manuals
known as the Sdhanaml and the Nipannayogval, sources that, if Mevissen is correct about
their dates, significantly post-date the Chinese translations. See Gerd Mevissen, Studies in
Pacarak Manuscript Painting, Berliner Indologische Studien 4/5 (1989): 362-3; and Images
of Mahapratisara in Bengal: Their Iconographic Links with Javanese, Central Asian and East
Asian Images, Journal of Bengal Art 4 (1999): 99, where he notes that the earliest actual statues
exhibiting this form date only to the eleventh century at the earliest.
3 Mevissen, Images of Mahapratisara, 99-10
3 Ibid., 10
3 Wu Hung 1989, The Wu Liang Shrine: The Ideology of Early Chinese Pictorial Art.
Stanford: Stanford University Press, 13
3 Stephen F. Teiser, The Scripture on the Ten Kings and the Makings of Purgatory in Medieval
Chinese Buddhism (Honolulu: University of Hawaii Press 1994), 42-43, uses the terms narrative
and icon. Teiser focuses on silk hanging scrolls. Although I have tried to apply the terminology
with care while adjusting it to the material under consideration, Teisers warning that we must
remember that the distinction is modern and arbitrary, and at times impedes an understanding
of medieval culture is well taken (ibid., 43).
3 Victor H. Mair, Records of Transformation Tableaux (pien-hsiang), Toung Pao 72
(1986): 3-4, makes an analogous distinction between bianxiang and maala.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

259

of relationship with viewers, direct or otherwise. In terms of the reception of the


amulets, this was certainly the case; however, my purpose here is to explore their
creation, sources, and rationale. From such a point of view, the concept of icon
helps to clari how the new designs of the fully xylographic amulets were the
products of an essential shi over time in their character a change in how the
dhras inscribed upon them were imagined.
In episodic images, as Wu Hung describes them, the significance of the
representation is realized in its own pictorial context. In contrast to an iconic representation, the viewer is a witness, not a participant.40 In the case of the amulet
sheets that figure episodes of the empowerment of their bearers both in text
and image we can take the first of these statements in the strongest possible
sense: not only is the significance of the depiction realized, but the very action
it depicts is carried out. The empowerment of Jing Sitai through his amulet, for
example, is not merely a scene in a depicted narrative of empowerment; according
to the ritual logic of inscribed dhras, wherein the material inscription is active
as an incantation, we are witness to the actual event itself. By contrast, in the later
iconic images of the multi-armed deity, the image is there for the viewer to venerate in the manner exemplified, in some cases, by the accompanying image of the
donor. The viewer, there by happenstance, and the pictured donor, are both in
this sense participants in the activity figured by these sheets. But here lies the
dierence between the earlier and later amulet sheets: in the later examples this
participation is indirect, contingent, inessential to the images main ritual function,
which consists in revealing the divinity. Though in texts accompanying paintings,
depictions of worship are sometimes described as being in themselves eternal
meritorious enactments of ritual (see below for an example), clear distinctions must
be drawn between the intensity and kind of the ritual actions figured in the two
sets of images. Veneration is, it must be emphasized, a central and powerful ritual
action in Buddhism, but it situates its actors at a distance om the divinity in
another order of reality. Empowerment, by contrast, positions its human object
directly within the stream of blessings and divine power. Thus though it is not
how the analytical distinction of episode and icon was originally drawn, or probably
intended the episodic dhra sheets are self-contained, of a piece; the iconic
dhra sheets, in contrast, imply a divide that those on the human side (our side)
cannot truly bridge.
Seen in this light, the spells inscribed on the two kinds of sheets are very dierent sorts of presences. The spells of the empowerment episodes envelop the human
traces (in the form of names) within them, nourishing them like fish within a
stream; mortal life partakes of their power. The incantations surrounding the icons,
in contrast, are themselves iconic extensions of the majesty of the central divinity.
When they are recited within the narratives of stras, dhras are sometimes said
to become part of the light that circles the buddhas who utter them light that
emanates out om them and fills the cosmos.41 Such images of spiraling spells
40. Wu Hung, The Wu Liang Shrine, 13
4 For an example we may take another dhra stra translated by Baosiwei, which reports

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

260

Paul Copp

may have been part of what the xylographers intended to evoke by arranging the
syllables in wheels that surround the central figure in many of the most visually
and technically sophisticated amulet prints. The incantation discs which are in
the minority, perhaps due to the relative technical diculty of producing them
resonate with a family of iconic visual structures central to contemplative forms of
dhra practice, derived om the circular objects of contemplation and visualization
known as moon discs (or simply moons, yuelun ), that includes syllable
wheels (zilun ), incantation wheels (zhoulun ), and dhra wheels
(tuoluoni lun ). These various wheels (or discs or circles, depending
on the context) were employed in a range of contemplative practices in dhra and
early Esoteric rites, most of which in their most basic forms involve the imagination
of a disc, likened to a full moon, upon which syllables, either alone or in series,
are made to appear. Though in the more elaborate practices of later Esoteric and
Tantric traditions these visualizations can be very elaborate indeed, in their earlier
forms, dating to the age of the Mahpratisar Stra, they were much simpler.
A passage om Bodhirucis translation of the Amoghapa-dhra Stra, a text
exhibiting a close relation with the Mahpratisar, makes clear the connection of
these objects of active imagination to the middle structuring zones of the amulets.
In the midst of imagining the visual form of the Buddha Mahvairocana, one is
to perceive (guan ) above the deitys heart (a standard site for the visualization
of the discs) a great moon disc, brilliant and utterly pure. At its round edge are
arrayed a hundred shimmering syllables (zi ) of burning gold, each moving along
[the edge of the disc].42 This spinning wheel of syllables then becomes the site
of more elaborate perceptions/imaginings, but the basic form of the wheel and its
characters which in many cases are explicitly said to form incantations makes
clear, I think, the connection with the amulets. Of course the amulets, though
cousins of the incantation wheels figured upon them, were not themselves used
in contemplative or other ritual practices of veneration (at least such were not the
uses for which they seem to have been made).43 Nevertheless, the iconic character
of their visual forms, which included their inscribed spells as much as their central
deities and their larger mandalic structures, was the product of a religious culture
that emphasized the worship of the visual forms of deities.
Earlier I discussed the close relationships between even in some respects
the functional identity of drawn iconic altars and the painted or printed amulet
sheets they so closely resemble. I showed how these relationships, conceptual as well
as figural, clari important features of the ritual characters of the amulet sheets.
Another related genre of images consists of painted icons on silk (oen bearing
the images and names of their donors or intended beneficiaries) and block-printed
versions thereof (mass-produced without individualizing features and containing
that as the Buddhas worlds-illuminating radiance was returning into his body, it proclaimed itself
to be the Wheel-Turning King One Syllable Mind-Spell of the Wisdom of all Tathgatas. See
Da tuoluoni mofazhong yizi xinzhou jing , T 956, 19: 315c.
4 Bukong juansuo shenbian zhenyan jing , T 1092, 20: 299c.
4 Cf. Fraser, Performing the Visual, 157-

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

261

instructions for their proper worship). These images further clari the nature and,
no doubt, contemporary sources for the mode of depiction of our iconic amulets.
Icons seem in fact to have been the standard form of portable Buddhist imagery
in the ninth and tenth centuries, at least as evidenced by the paintings and xylographs found at Dunhuang. In common with amulet sheets that center images
of the eight-armed deity, such images feature iconic representations of divinities
typically either buddhas or bodhisattvas depicted in full ontal posture whose
presence and radiating divinity are the focus of the image as a whole. In the case of
silk paintings, they oen include an accompanying image of their donor as well
as texts ancillary to the image that explain its nature or the facts surrounding its
creation (or both). For an example we may take a typical icon of Guanyin om
Dunhuang, the painting on silk known as Stein Painting 14, dated to 9 It shows
the bodhisattva in standing posture flanked by two much smaller human figures,
a deceased nun and a young man identified as the younger brother of the donor
(Fig. 9). Like the xylographed dhra amulets, this painting prominently features
texts common to its genre, among which is one in praise of the bodhisattva that
states the intention that the image will perpetually enact fulsome oerings to him
(yongchong gongyang ). Other wishes include those for the peace and stability of the realm and for the rebirth of the donors deceased relatives and Buddhist
teacher (the nun pictured) in a pure land.44 Most important for a comparison with
the amulets are statements of the potency of the deity, and by extension its image,
which parallel statements of the ecacy of the dhra xylographs.45
Portable images such as these, which had long been popular by the time of
the transition to the woodblock printing of amulets, surely provided key elements
of the basic iconic template according to which the carvers of woodblocks made
their images a template that had apparently been adapted in turn om still
older iconic images such as those of buddhas at the centers of Pure Land tableaux
on Dunhuang cave shrine walls. Some notable narrative xylograph images remain
om the period, the most famous being sutra illustrations, a genre that for Teiser
typifies the narrative mode.46 Most early printed Buddhist images, however, are
iconic in character. Some of these block-printed ritual icons bear a basic visual
resemblance to xylographic amulets, even though of course they were parts of very
dierent practices and were thus very dierent sorts of objects. Similarities range
om the deity represented to the accompanying texts, which sometimes included
4 The specific target of the good wishes for the realm was the Tang, which (if the author
meant the great three-hundred year empire by that name, and not a later oshoot) would have
been ironic, given that the Tang empire had ended some three years earlier.
4 For a fuller description and analysis of this painting, see Roderick Whitfield, The Art
of Central Asia: The Stein Collection in the British Museum (Toky: Kdansha , 1983),
v. 2, pl. The image and Whitfields analysis are also included in the website of the International
Dunhuang Project (idp.bl.uk), under the search value 1919,0101,0.14. See also Arthur Waley,
A Catalogue of Paintings Recovered from Tun-Huang by Sir Aurel Stein, K. C. I. E., preserved in
the Sub-department of Oriental prints and drawings in the British Museum, and in the Museum of
Central Asian antiquities, Delhi. (London: Kegan Paul, 1931), 2
4 Teiser, The Scripture on the Ten Kings, 4

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

262

Paul Copp

Fig. 9: Painting of Guanyin and devotees. Stein Painting 14 (Ch. Liv.006). Ink
and pigments on silk. 77cm x 48.8 cm. Dated 910. After Roderick Whitfield,
Art of Central Asia: The Stein Collection of the British Museum, v. 2: figure 7.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

ALTAR, AMULET, ICON

263

Fig. 10: Votive print of Majur and associated ritual. Stein Painting 237 (Ch
00151b). 27.9 cm x 16.8 cm. After Roderick Whitfield, Art of Central Asia: The
Stein Collection of the British Museum, v. 2: figure 142.

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

264

Paul Copp

incantations. One typical and clear example is Stein Painting 237 (Fig. 10), a ritual
icon with an inscription in two rows of text to either side of the main figure, which
proclaims: The Great Saint Bodhisattva Majur universally urges all to resolve
to make oerings to and receive and hold [this icon].47 The text below the icon
proper proclaims that its image is the true countenance (zhenyi ) of the great
deity on its adopted seat in the Wutai mountains , proper devotion to which
according to the spiritual and economic capacities of each devotee will fulfill
all prayers, ultimately bestowing the return to eternal bliss of all sentient beings
(huishi youqing tonggui changle ). To this end, the text provides
two incantations for its worship: The Five Syllable Incantation of Majur, The
Bodhisattva of Child-Like Truth (Wenshushili tongzhen pusa wuzi zhenyan
), a-ra-pa-ca-na, the famous and ancient basic mantra of
the Indic deity found in a great many Buddhist texts; and an otherwise unattested
incantation called the Dhra of Majurs Awesomely Potent Dharma Treasure
Store Mind (Wenshushili da weide fabaozang xin tuoluoni
). Xylographs such as this one though, it must be emphasized, of very
dierent nature than our printed amulets were very likely produced within the
same printing shops as those amulets, where they presumably provided a resource
for the execution of the new dhra iconography, thus helping to bring about the
transformation in the visual character of the amulets.
Acknowledgments
I am deeply grateful to Professor Lothar Ledderose for the year I spent in Heidelberg in 20062007 as a member of his research team at the Heidelberger Akademie der Wissenschaen. I began
my work on the Mahpratisar amulets there, where I benefitted greatly by conversations with
Professor Ledderose and the members of his team. I would also like to thank Nobumi Iyanaga,
Luca Gabbiani, and especially Lothar von Falkenhausen for their critiques of this article, which
have greatly improved it. Needless to say, any mistakes or infelicities that remain are mine alone.

4 It is crucial to emphasize the very dierent natures of icons such as Stein Painting 237
and the dhra amulets treated in this essay. Cf. Fraser, Performing the Visual, 155-15

cole anaise dExtrme-Orient, Paris, 2010


Do not circulate without permission of the editor / Ne pas diuser sans autorisation de lditeur

Vous aimerez peut-être aussi