Pramāṇa · Search · About · API

Mediaeval Sinhalese Art c18: Appendix to Chapter VIII. Sāriputra

Mediaeval Sinhalese Art c18

Language
English

Contents

Appendix to Chapter VIII. Sāriputra

c18.2

Grünwedel's work on 'Buddhist Art in India,' while specially lacking in reference to ancient or modern Sinhalese art, nevertheless makes mention of the interest attaching to works containing formulas for the construction or drawing of Buddhist images, giving detailed dimensions. These he speaks of as modern and as specially characteristic of the northern school. It is therefore of special interest to be able to give some account of the work known in Ceylon as Sariputra, which deals with the dimensions of images in general and Buddha in particular.

c18.3

I am indebted to my learned friend Mudaliyar E. R. Gooneratne of Galle, for a translation of the Sariputra, based on a transcript of the copy in the Colombo Museum, compared with three copies which he with difficulty secured in the Southern Province and which were deficient in the first and more general part which appears in the Museum ms. Before proceeding to give some idea of the contents of the work, I quote a part of the learned Mudaliyar's remarks upon it :—

c18.4

“The work takes its name from the author, who says at the end of the work that he heard *Sariputra* it, from whom he does not state, though the inference in such cases, is that he must have heard it from one of the ṛishis. From the style of the verses, though they are presented to us in an incorrect form, it is plain that the work is the production of a master hand. The author must not only have been well versed in Sanskrit, but (it is clear) that he was also a thoroughly efficient artist.

c18.5

“*Sāriputra* is a common name in Buddhist history. One or two distinguished disciples of Buddha owned it, and subsequently at various periods, several distinguished authors and grammarians went by this name. The meaning of the term is, the son of Sāri (Sāri-putra). . . .

c18.6

“The making and painting of images have been in the hands of a particular section of the Sinhalese; not being versed in the Sanskrit, and not understanding the texts, it is plain that they only took care to have the paraphrase jotted down, to enable them to take the measurement, regardless of the accuracy of the Sanskrit verses. Hence the work is full of errors, and I have had great trouble to get the verses corrected and put into proper metre before commencing the translation.

c18.7

“In the first part which treats of images in general, in the paper copy sent me, all the Sanskrit stanzas were omitted and by the aid of the gloss I had to recast them.

c18.8

“The second part which is confined to the making of images of Buddha, is more complete; only a few of the verses are missing, but those that were found were full of errors and omissions, and were deficient in metre. I had them collated with three copies which I procured. It is curious, however, that the missing verses and omissions in one copy were nearly the same in the others, leading me to infer that they must be copies of one original work. . . .

c18.9

“In conclusion I have to acknowledge with thanks the very great assistance rendered me by Talamalagala Somānanda Thēra, of the Simbalī Āvāsa, Galle. But for his willing support . . . I am doubtful whether this translation would have seen the light of day.”

c18.10

Mr. Gooneratne also remarks, “(1) I have confined myself to translating the original literally. (2) In some places the text is so vague, as regards the exact positions of the parts of which measurements are given, and the terms used are so uncommon, I am not sure whether my rendering is quite correct. (3) Several of the verses had to be recast, of course embodying the exact meaning of the gloss, nevertheless there may be deficiencies. . . .

c18.11

Regarding the date of the Sinhalese gloss (*sannaya*) which constitutes the most important part of the mss. regarded from the imager’s point of view, the following statement, an annexure discovered in one of the old copies, is supplied by Mr. Gooneratne:

c18.12

“About A.D. 1165, after the translation of the Sinhalese commentaries into Pāli, the pious and illustrious King Sarvajña Parākrama Bāhu ascended the throne of Laṅkā. Having noted with concern the corruptions that had crept into the Saṅgha since the sway of the great King Vaṭṭagāmini Abhaya he took action to purge it from its corrupt adherents, with the aid of the illustrious Mahā Kāsyapa of Dimbulāgala (Dambulla), and he thoroughly renewed the Cetiya at Purvārāma and other temples (Anurādhapura) and the Uposatha hall, and embellished the parapet around the Bōdhi tree with various beautiful paintings. Mahā Kāsyapa then approached his Majesty and prayed for a translation of the Sāriputra, a portion of the old Vedaṅga, which treats of the dimensions of the images of Buddha for the guidance of lay artists. The King deputed this work to a pupil of the Mahā Thēra Kāsyapa, who was competent to undertake it.”

c18.13

I do not give the text and translation of the Sāriputra in full, as they deserve separate publication elsewhere. The parts of the translation omitted are the astrological details, some of the more minute measurements, and some of the verses of which the exact meaning is not clear. The second part of the work (as found in the Colombo Museum ms.) treating in particular of standing, seated, and reclining images of Buddha, is here first dealt with; afterwards the first or general part and a further gloss accompanying it. The work is not now well known either by imagers and painters or by the priesthood in the low-country, but is familiar to the best craftsmen in the Kandyan districts. The diagram (Pl. XXVII. and Fig. 95) of a seated image is drawn by D. S. Muhandiram to illustrate the use of plumb lines, and to show the appearance of a correctly measured image. The chief point to be comprehended is the use of the lines: a framework, of which a plan appears in the top left hand corner of the figure, and another (Fig. 94), taken from an actual example given to me by the same artist and now in my possession, in the text, is fixed horizontally over the place where the image is to be made, and the plummets suspended from it. The plumb lines so hung serve as axes of reference, determining the position of the chief parts of the body. They act as a simple ‘pointing machine.’ The lines shown in the diagram are those generally relied upon by Kandyan workmen, but others are available if still more exact work is required, or if the size of the image is so great as to make their use advisable. The ‘karṇa sūtra’ of the diagram is the ‘pārśva sūtra’ of the text, which latter is probably correct. The height is reckoned at 5 ‘faces’ from top of forehead to throne, i.e. 67½ inches, an approximation used, it is said, in small work instead of the 62 inches of the canon.

c18.14

Fig. 94. Plan of lamba taṭuva, showing points from which plumb lines are suspended.

c18.15

In the case of reclining images the ‘lines’ are not, of course, plumb lines, but threads drawn from point to point.

c18.16

Some explanation of the units of measurement must be given. They may be tabulated thus:—

c18.17

| One *mukha* | 13½ | *aṅgal* (lit. fingers), ‘inches,’ or units. |
| One *tāla* | 13½ | do. |
| One *bhāga* | 4½ | do. |
| One *veda* | 4 | do. |
| One *kala* | 2 | do. |
| One *kolaka* | 2 | do. |
| One *matra* | 1 | do. |
| One *yava* | ⅛ | do., a ‘barleycorn’ |

c18.18

The first three units as thus tabulated apply to images of Buddha only. The ordinary architect’s cubit (vaḍu riyana), the foot (pada), and span (viyata) are scarcely mentioned. The angala or ‘inch’ is presumably the vaḍu angala, (distance from first to second joint of forefinger when bent), the twenty-fourth part of a vaḍu riyana, and measures 1 3/4 English inch (see p. 142). The ‘angala’ of the text, however, is not merely a fixed measure, but also may be translated ‘unit,’ as the instructions are intended to provide for the construction of images of all sizes, both large and small. But there has been much discussion as to the exact dimensions of the images in the case of images of Buddha,* and special books have been written on the measurement of his person, usually considered to be twelve cubits†; his span is usually considered much longer than the span on which calculations are at present based; but such views are purely mythical. The other units employed are technical and not in common use. The most important is the mukha, or ‘face.’ This is the equivalent of 13 1/2 inches when images of Buddha are spoken of. It is an unit corresponding to, but not identical with, the ‘head’ of European artists. It is measured from the base of the hair on the forehead to the chin and is thus less than a ‘head’ by the height of the crown of the head above the forehead. The standard height of an image of Buddha, including the crown of the head, is 124 angal, as stated in verse 22 of the canon, or more exactly 9 ‘faces’ and half a bhaga, making 123 3/4 inches (about E. 13 feet 4 inches). The detailed measurements given actually add up to 121 1/2 angal (9 ‘faces’) + 3 1/2 angal for the crown of the head. It will be seen that this canon of proportion corresponds to about 7 1/2 heads, in the terms of European art.

c18.19

Some words are perhaps needed as to the purpose and use of artistic canons of proportion. Probably they have been in use from the earliest times in the East, as we know that they were in the West. But there is a tendency, not altogether unreasonable, to regard them as purely mechanical devices, reliance upon which is a fetter upon free artistic expression, and characterises rather the decline than the flower of a great tradition. But it is the abuse, not the use, of canons of proportion that does this, and it is therefore worth while to enquire into the rationale of their use by Indian craftsmen and imagers. It is in drawing from the life that a canon is likely to be a hindrance to the artist; but it is not the method of Indian art to work from the model. Almost the whole philosophy of Indian art is summed up in the verse of Sukrācārya’s ‘Sukranītisāra’ which enjoins meditation upon the imager:

c18.20

“In order that the form of an image may be brought fully and clearly before the mind, the imager should meditate; and his success will be proportionate to his meditation. No other way—not indeed seeing the object itself—will achieve his purpose.”

c18.21

The canon, then, is of use as a rule of thumb, relieving him of some part of the technical difficulties, leaving him free to concentrate his thought more singly on the message or burden of his work. It is only in this way that it must have been used in periods of great achievement, or by great artists. It must also be remembered that the use of some such device was always absolutely essential in the case of the many colossal images made, and often executed in a relatively cramped space, such as the interior of a rock vihāra. Further, only a few of the measurements given are used in the case of the smaller images; all the minuter details are necessary only in the case of colossal statues. And the fact that the canon is now used mechanically by mere craftsmen is no contradiction of its value, for even their work is the better, not the worse, for their reliance upon tradition and rule. This at first sight may be doubted, but will be obvious to those familiar with the degeneration that everywhere results when even the old tradition, however rigidly interpreted, is abandoned, and a futile endeavour made to imitate a foreign style or the individualism of Western art.*

c18.22

∗ See J.C.B.R.A.S., Vol. V., No. 19, 1874. † Clough gives Maha-purusha, ‘name of Buddha alluding to his height, which is said to have been four of his own cubits measured from the tip of his middle finger to his elbow, and equal to twelve cubits of common measure.’

c18.23

A translation of the text of the canon now follows, beginning with Part II., which deals with images of Buddha. Part I. deals with the measurements of images of gods and men.

Sāriputra. Part II., translation

c18.s1.1

Hail to the Omniscient!

c18.s1.2

1. “Let that Lord Buddha, having cheeks of a golden hue, lips of light red, captivating eyes, brows that are lovely as a bow, shining like the moon, who eclipsed the lotus, who is beauteous as the faces of the daughters of Māra, guard thee.

c18.s1.3

2. “Hearken, we shall now describe the dimensions of images of Buddha in the three postures of sitting, standing and reclining.

c18.s1.4

3. “Images of Buddha are made from gold, copper, clay, stone, wood, burnt clay, and lime.*

c18.s1.5

4. “. . . The size of the (image-) house should be in proportion to the size of the image, and vice-versa. . . .

c18.s1.6

8. “We shall now describe the various heights of images. Those that are less than two cubits (E. 62 inches) are called very small (*alpa*) images:

c18.s1.7

9. “Images above two cubits and up to five cubits are small (*adhama*): images above five cubits and up to eight cubits are called middling (*madhyama*):

c18.s1.8

10. “Images above eight cubits, and up to twelve are called large (*uttama*): images above twelve cubits (E. 31 feet) are great (*maha*).

c18.s1.9

11. “. . . No images of gold or other metal should be cast hollow within.

c18.s1.10

12. “The making of hollow images will ere long result in the loss of wife and wealth, and lead to quarrels and famine,

c18.s1.11

13. “and to delay in the performance of work, and dread of enemies to royalty. The joints of images of wood and stone, must be placed exactly where the joints are, and not otherwise. If images are made thus it will result in length of days.

c18.s1.12

14. “If this rule be deviated from it will result in the loss of wealth and life. If the measurements be out by half an *angala*, or (even) a barleycorn, it will result in loss of wealth, and death. Therefore the measurements must be neither more nor less.

c18.s1.13

15. “The height of an image must be calculated by a division of it into a scale† of inches . . .”

c18.s1.14

† *Patti lella if of wood, ‘kadavasa kole’ (for kadadasi kolaya) if of paper.*

c18.s1.15

16—21. “These verses treat of the relation between height and breadth of the image, whether lucky or not, in accordance with Shaḍvargaya.

c18.s1.16

22. “One hundred and twenty-four *angal* scale of an image is called the *uttamadasatale*. [cf. Grünwedel, ‘Buddhist Art in India,’ 1901, p. 167, note by A. G.]

c18.s1.17

23. “Two *kala* and half an *angala* is called a *bhāga*; thirteen and a half *angal* is called a mukha. From the top of the head* to the sole of the foot is nine mukha and half of a bhāga.

c18.s1.18

∗ *Cf. Sukracarya: “There are the following kinds of images, arranged in order of durability: (1) Images made of sand; (2) Images made of flour or meal; (3) Drawings; (4) Plasterings; (5) Earthen images; (6) Wooden images; (7) Stone images; (8) Metal images.”*

c18.s1.19

24. “According to this scale the length of the face is three bhāga. The uṣṇīṣa† is equal to eight yava. The length of the part just above the brow, covered by hair is three aṅgala.

c18.s1.20

† Protuberance upon the skull, one of the thirty-two chief physical perfections or ‘lakṣaṇas.’

c18.s1.21

25. “From the centre of the forehead to the centre of the space between the eyes, and from the end of the nostril to the jaw is one bhāga.‡ The piece of flesh projecting beyond the chin is half an aṅgala.

c18.s1.22

‡ The face is divided into three equal parts.

c18.s1.23

26. “From the base of the throat to the top of the neck is 3½ inches, and from thence to [the end of the breast and thence to] the end of the navel, and from thence to the end of the penis is one tāla in each case.

c18.s1.24

27. “The length of the thigh is two tāla; the foot and the knee are two bhāga; these are the principal measurements.

c18.s1.25

28. “The length of the [upper] arm is 27 aṅgala; the length from the elbow to the wrist is 21 aṅgala. The length of the hand [from the wrist to the end of the fingers] is 13½ aṅgala . . . , the same as that of the face.

c18.s1.26

30. “The back of the head, from the crown to the nape of the neck, is one face and three aṅgala [42½ aṅgala]; the space from one ear to the other [measured over the crown] is 10 kolaka [20 aṅgala] and twice this sum is the circumference [viz., 40 aṅgala, the same as stated below, but without including the thickness of the hair].

c18.s1.27

32. “The diameter [of each tuft of hair] is eight barleycorns, and the height six; the area within which the hair is found is three faces and one veda [i.e., 44½ aṅgala round]; there are 360 tufts of hair.

c18.s1.28

36. “The breadth of the chest is 40½ aṅgala . . . [?]

c18.s1.29

37. “The space between the breasts is 19 aṅgala . . . [?]

c18.s1.30

40. “The edge of the hair in front by the forehead is curved, in the shape of a bow or like an arch. The height of the eyebrow is an aṅgala, and the breadth two barleycorns . . .

c18.s1.31

42. “The length of the eye from the unguent-spot§ to the other end is three aṅgala; and its breadth six barleycorns. The two lids are two barleycorns.

c18.s1.32

§ Karavīra, the small aperture at the beginning of the eye where unguents are placed, hence called aṇḍun-vala in Sinhalese.

c18.s1.33

43. “The eye is divided into three parts, one black and two white. In the black part the disc is of the size of a louse. [Part I. gives greater detail.]

c18.s1.34

44. “The space between the eyes is one kala; the height of the nose at the top is one aṅgala, and the skin of the eyelid half a mātra.

c18.s1.35

45. “The length of the nostril is one aṅgala and a kala, the height of the nostril is one aṅgala, and the breadth of the hole is six barleycorns.

c18.s1.36

47. “The mouth is between the two cheeks [lit. parts which display or smile]. . . . The upper lip has three curves.

c18.s1.37

48. “The width is 1½ kala [three aṅgala]; the breadth of the [lower] lip is one aṅgala. . . .

c18.s1.38

50. “The face will be beautiful if the aforesaid parts are [thus] completed. The breadth of the ear is two aṅgala, the [total] length five kolaka; the length of the part above the pendant is four aṅgala; the pendant is three kolaka [six aṅgala].

c18.s1.39

51. “The width of the pendant part is half an aṅgala, the pendants on either side the same.

c18.s1.40

∗ The translation reads ‘forehead,’ but evidently ‘head’ is intended. Nine mukha and half a bhāga are 123¾ aṅgala, practically the 124 aṅgala of v. 22.

c18.s1.41

Standing images of Buddha

c18.s1.42

The space between the pendant of the ear and the neck is 1½ *angal*. The height of the outside of the ear is 1½ *angal*.
54. “The length from the base of the neck to the nipple, thence to the other nipple, and thence to the base of the neck is 13½ *angal*.*
55. “From the base of the neck to the armpit is one *tala*, and the part over is one *kolaka*. The breadth of the palm is six *angal*, and its length seven *angal*, the thickness of the palm one *kolaka*, the breadth of the thumb 1½ *angal*.
56. “The length of the middle finger is six *angal* and four barleycorns; the fourth finger is 5½ *angal*; the forefinger is less by one barleycorn. [Verses 57-63 give further minute details of the measurements of the fingers and finger nails.]
64. “The length of the foot is 13½ *angal*, and from the ankle bone to the heel is one *bhaga* [two *angal*]. [Verses 65-71 give detailed measurements of the toes.]
72. “The breadth of the back between the two armpits is two *mukha*. . . .
73. “. . . the space between the two thighs at their beginning is one *angala*, between the two knees one *kolaka* [two *angal*] and the space between the two [great] toes one *kala* [two *angal*] and a barleycorn.
74. “The *maddhya sutra* [medial plumb line] drawn from the crown of the head must touch the tip of the nose, the centre of the chest, the navel, the private part, the inner side of the knees, and the inner sides of the feet.
75. “The *parsva sutra* [side plumb line] drawn from the crown of the head must touch the side of the forehead, the centre of the cheek, the side of the arm, the centre of the thigh, the centre of the knee, as well as the centre of the ankle joint.
76. “The girth must be three times the breadth of the several parts. . .
77. “The end of the hand [raised in] blessing must be in a line with the end of the chin, the cheek and the nose, at a distance of eight barleycorns from the end of the mouth and the nipple of the breast. [This is incomprehensible.]
78. “The *kata sutra* [loin plumb line] must be drawn from the side of the little finger [of the raised hand] to fall over the middle of the hip, the middle of the thigh, and outside the knee.
86. “The *kaksha sutra* [armpit plumb line] must be drawn from the side of the armpit, by the side of the hip, the calf and its termination, and by the fifth toe of the leg.
87. “The *bahu sutra* [shoulder plumb line] must be drawn by the centre of the side of the [left] arm by the end of it and outside the palm.
88. “A *maddhya sutra* [medial plumb line] must be drawn [behind] from the centre of the crown of the head, the centre of the rising part of the back, the centre of the [small of the] back, and between the heels.†
90. “If the plumb line is drawn amiss, the image maker will be afflicted with illness . . . if correctly it will lead to the success of royalty and the protection of the maker. The section specially referring to standing images of Buddha ends here.

c18.s1.43

Seated images of Buddha

c18.s1.44

91. “I shall now describe *yogasana* [meditative] images. The tip of the nose, the thumb and the heel must be visible to the eye [of the image].
92. “The image must exemplify a person with an unfettered, quiescent, and absolutely pure mind, pre-eminent from head to foot.
93. “From the place where the feet touch the throne, to the top of the hair on the head, [Figure: DIAGRAM OF LAMBA TATUWA OR FRAME FROM WHICH PLUMB LINES ARE SUSPENDED. BAHU SUTRA KAKSHA SUTRA KARNA SUTRA MUKHA SUTRA MADDHYA SUTRA MADDHYA SUTRA CHAKSU OR MUKHA S. KARNA SUTRA KAKSHA SUTRA BAHU SUTRA Level of frame (lamba tatuwa) from which plumb lines are suspended. SCALE OF 'FACES' AND INCHES.]

c18.s1.45

∗ This appears to contradict v. 37, but is evidently correct. † Verses 84, 85, and 89 speak of two or three other plumb lines, but less clearly defined and not named.

c18.s1.46

Fig. 95. Seated figure of Buddha, drawn to a scale of five ‘faces,’ and showing the principal plumb lines used in making images [D. S. M.].

c18.s1.47

*Seated* *images of* *Buddha*

c18.s1.48

the length is 62 *angal*.* The expert imager must make the figure in the due proportion.
94. “The left foot is beneath, the right foot must be placed over it, the hands must be placed likewise. The image will then be pleasing.
95. “The length from the feet to the hands is one face, and from thence to the private part and navel one *bhāga*.
96. “From the base of the throat to the navel is 27 *angal* [i.e., two faces]. From the base to the top of the neck is one *bhāga* less one inch.†
97. “From the collar bone to the throat joint is half an *angala*. The face is 13½ *angal*. The height of the crown of the head is three *angal*, the top 1½ *angal* [cf. v. 24].
98. “The three‡ sedent positions are as aforesaid, and the skilful craftsmen must make them accordingly. [Verses 99-104 give details of position and dimensions of hands, knees and feet].
105. “The space between the two knees, the height from the ankle to the end of the hair of the head, as also from [each] knee to the [opposite] shoulder are the same.§ [Verses 106-111 give measurements of the body, and arms, and indicate their relation.]
112. “A line must be drawn from the middle of the forehead, the middle of the tip of the nose, over the middle of the chest, and over the middle of the navel, the middle of the private part, between the two joints of the thumbs, and the middle of the ankles of the feet;
113. “This is the *maddhya sūtra* [medial plumb line] which must be drawn at a distance of twelve inches from the crown of the head.‖ 114-115. “From the plumb line to the centre of the forehead is three inches, to the tip of the nose one inch, to the bridge of the nose three inches, to the chin three inches, and to the end of the neck seven inches: to the base of the neck seven inches and three barley-corns, to the chest three inches, to the middle of the belly one inch, and the line must touch the private part.
116. “This plumb line should be drawn as described above. [Another] line must be drawn by the end of the mouth, the end of the nipple of the breast, the end of the wrist joint, and the side of the instep [?]. This is called the *mukha sūtra* [mouth plumb line].
117. “ [Another] line must be drawn by the armpit, the side of the hip, the side of the calf and by the elbow. This is called the *kaksha sūtra* [armpit plumb line].
118. “The expert imager should mark what is said above, and draw the lines accordingly. Sideways from the centre of the crown, the side of the ear, the side of the arm and the elbow, this line must be drawn over the three seated images.**
119. “The height of the seat should be either one third or one quarter or one fifth of the height of the seated image. The description of seated images ends here.

c18.s1.49

∗ *More roughly in practice, for small images, and as in the figure, five faces, not including the hair.* † *i.e., half an angala less than the measure given in v.

c18.s1.50

²⁶ The discrepancy appears to be explained by v. 97.* ‡ *1. Yogāsana, meditative, with hands laid flat on one another on the lap.

c18.s1.51

² The right falling to the ground over the thigh.

c18.s1.52

³ The right hand raised and held palm forward, the left hanging flat by the side or holding garment.* § *This is a very important check, useful even in the case of quite small images or drawings.* ‖ *So that the line may clear by one inch the tip of the nose, as indicated in the next verse; it will be seen that the plumb line serves as an axis of reference from which by measurement the position of various parts of the body is known.* ** *Cf. v. 98, also referring incidentally to the three varieties of seated image.*

c18.s1.53

120. "I give below a description of recumbent images. The recumbent image should be *Sariputra* made exactly in the attitude of a devotee.
121. "The image must be made without contracting the vital joints, [allowing them to be pressed on the throne], then only will it lead to all prosperity.
122. "Otherwise it will end in the destruction of the place as well as of wealth. The length of the [image-] house should be divided into sixteen parts.
123. "Two of these should be vacant at either end, and the remainder occupied by the reclining image. The breadth of the house should be divided into seven parts, three and a half of which should be occupied by the couch of the reclining image.
124. "The breadth of the recumbent image must be one *Brahma pada*, and half of a *preta pada*. . . . The couch must be within the *Brahma sutra* precincts of the house.*
125. "When the reclining image is two *angal* within the limits of the *Brahma sutra* it will lead to all increase of prosperity; if it is made to extend beyond the *Brahma sutra* it will destroy wealth and life.
126. "The *maddhya sutra* must be drawn from behind the ear, the side of the pendent of the ear, the navel and the centre of the private part, and the knees and between the feet.†
127. "The *parsva sutra* [must be drawn] from the [upper] side of the crown, by the side of the neck, below the elbow and the wrist, and over the middle finger and the ankle.
128. "The *mukha sutra* must be drawn from the middle of the crown, the middle of the forehead, the middle of the nose, the side of the breast and the side of the lower hip.
129. "The *vamsa sutra* [abdominal line] . . . [not clear].
131. "The space between the two nipples of the breasts is 11 *matra* [11 *angal*, cf. with the 13½ *angal* of v. 54, where the body is upright and not contracted]. . . .
133. ". . . The *netra sutra* [eye line] must be drawn from the side of the arm on which the head rests.
134. "The line must be drawn from the thumb by the side of the cheek, the neck and the pendent of the ear. The [under] armpit must be in a line with [i.e., immediately beneath] the beginning of the shoulder of the other arm.
136. "The height of the pillow is the same as that of the face. The full rounded face when complete with all its features is beautiful.
137. "The measurements of the standing, seated and reclining images of Buddha should be as above, moreover, the images should be illustrative of the perfections of Buddha.
138. "If the image is perfect it will lead to the increase of merit and prosperity, otherwise it will lead to the destruction of all wealth and length of days.
139. "The image of the Asura Rāhu‡ should be drawn in accordance with the position he occupied at the estimable feet [of the recumbent Buddha], covered with every ornament, holding flowers and garlands in his hands, bending in three places, and in every respect perfect. “The end of the measurement of images as heard by Sāriputra.” [Figure:] Part II. of Sāriputra ends here.

c18.s1.54

∗ *This refers to a division of the site of a house or temple into four areas by means of lines called Brahma sutra, Deva sutra, etc., dividing the whole into 64 squares. The four central squares compose the Brahma pada; the 12 surrounding squares the Deva pada; the 20 next surrounding squares the Manushya pada; the 28 outer squares the Preta pada. Diagonals are drawn from corner to corner. A house should not be built upon the Brahma or Preta areas. We are here told that the image in a temple should be within the Brahma area. Similar particulars are given in the Brihad Samhita, II., 6, vv. 55-58. The terms Brahma pada, etc., are also used in this verse as units of measurement: padaya=a foot; also a quarter, part, etc.* † *The line—not here a plumb line, of course—is asymmetrical as regards the head, because the head is raised on a pillow out of the straight line of the body, and so with the other lines.* ‡ *The figure of Rahu is not generally found in association with reclining images in Ceylon.*

Sāriputra, Part I

c18.s2.1

The first part of Sāriputra treats of images in general, not as above, exclusively of images of Buddha. Here the *mukha* or ‘face’ consists of 12 *aṅgal* only, instead of 13½, and the *bhaga* of 4 *aṅgal* instead of 4½. The other units are unchanged.

c18.s2.2

Verses 1—30 give very detailed measurements of the head and features, body and limbs, hands and feet. The dimensions are slightly less than those above detailed: thus the breadth of the back between the armpits is two faces (of twelve *aṅgal* each only), the middle finger six *aṅgal* long instead of six *aṅgal* and four barleycorns, and so on. Verse 30 concludes: “all the rest not described may be suitably supplied.” Verses 30—43 treat of the plumb lines and measurements to be made thereby. Reference is made to the part ‘slightly slanting,’ this seems to refer to a slight bend in the waist, which, like the ‘sway’ in thirteenth-century Gothic work, gives grace and delicacy. I do not think that reference is here intended to the markedly curved figures known as *trivaṅka** (triple-curved).

c18.s2.3

31. “Of images of which a part is erect, and a part slightly slanting, we give dimensions of the part slightly slanting, between the *Brahma sūtra* [=medial plumb line, above] and the *lalāṭa sūtra* not previously mentioned.

c18.s2.4

32. “The dimensions of the images of gods and men are nine *tala* [=nine *mukha*, or faces] less by two *aṅgal* from the crown of the head. From the termination of the hair down to the foot the *Brahma sūtra* should be hung.

c18.s2.5

33. “The space between the two toes is one *tala*. The image should be erect and firmly fixed. The part slanting should be measured by a curved line.

c18.s2.6

34. “In the part slightly slanting, from the *Brahma sūtra* to the *lalāṭa sūtra* is two *kala* and one *aṅgala* [five *aṅgal*].

c18.s2.7

36. “. . . From the *Brahma sūtra* to the face is three *kala* and half an *aṅgala* [6½ *aṅgal*; can this be a mistake?].

c18.s2.8

37. “From the [*Brahma*] *sūtra* to the chin is one *aṅgala*; thus the measurement regarding the face.

c18.s2.9

43. “From the first toe of the foot over the base of the throat and the forehead, also from the eyebrow eye end of the nostrils and base of the throat and also over the breast and side and the centre and beginning of the thigh, three lines must be drawn. . . [Figure:] Next follow measurements given in Sinhalese in the original copy, without any mention of its being, like the foregoing, a gloss (*sannaya*) and without any *śloka* as its basis. A part of this is quoted].

c18.s2.10

“The length of an image is 124 *aṅgal*, when it is divided by four, 31 *aṅgal* will be the resultant. . . When a line is drawn horizontally through the navel, the portion of the elbow below it will be two *aṅgal*, and the loins below, four *aṅgal*; below that is the hip, and below that the private part.” [The prose next refers to many other lines, such as the *purva* and *pṛṣṭha*; *hikka* (=*Brahma*?); and lines drawn horizontally 1½ inches above the navel, across the navel, the centre of the private part (six inches below the last), and across the calf of the leg (eight inches below the last). It would appear that these notes are with regard to a seated image of Buddha. Much that appears very vague in the prose would no doubt be quite clear to the expert imager. I have quoted only so much as *Sāriputra* appears to be important or quite clear.]

c18.s2.11

∗ Not ‘three sided’ as suggested by the translation in the Mahāvamsa, footnote on p. 256, ed. 1889. The images in low relief within the porch at Laṅkatilaka have been pointed out to me as ‘trivaṅka.’

c18.s2.12

Comparison with other canons.

c18.s2.13

Of the great series of works composing collectively the *Silpa Śastra*, or Science of the Arts, very little is at present known. The measurements given in the second part of Sāriputra are those usually met with in the parts of these works which deal with images. To illustrate this point, and to present more clearly the chief peculiarities of the canon, the following table of some of the principal measurements is prepared, the numbers being given in ‘inches’ (*aṅgal*):

c18.s2.14

| | Sāriputra [figures of Buddha.] | Sāriputra [other figures.] | Sukrācarya.* | Bṛihad Samhita.† |
| Whole figure to top of forehead ... | 9 faces | 9 faces | 9 faces | 9 faces. |
| Face ... ... ... | 13½ | 12 | 12 | 12 |
| Crown of head ... ... | 3½ | 3 ? | 3 | 3 ? |
| Three parts of face ... | 4½ each | 4 each | 4 each | 4 each |
| Neck ... ... ... | 3½ | 4 | 4 | 4 |
| Neck to breast ... ... | 13½ | 12 | 12 | 12 |
| Breast to navel ... ... | 13½ | 12 | 12 | 12 |
| Navel to penis ... ... | 13½ | 12 | 12 | 12 |
| Thigh ... ... ... | 27 | 24 | 24 | 24 |
| Knee ... ... ... | 4½ | 4 | 4 | 4 |
| Shank ... ... ... | 27 | 24 | 24 | 24 |
| Foot, height ... ... | 4½ | 4 | 4 | 4 |
| ” length... ... | 13½ | 12 | 14 | 12 |
| Hand ... ... ... | 13½ | 12 | 12 | 12 |
| Palm ... ... ... | 7 | 7 | 7 | 7 |
| Middle finger ... ... | 6½ | 6 | 5 | 5 |

c18.s2.15

† *Brihat Samhita of Varaha Mihira, English Trans. by N. Chidambaram Iyer, Madura, 1884, Pt. II., Ch. XI.*

c18.s2.16

It is evident that the universal Indian standard is one of nine ‘faces,’ which in the terminology of European canons of proportion would be just over seven ‘heads.’ The figure of Buddha is made only relatively ‘heroic’ by increasing the unit, without altering the proportions to any appreciable extent. But there appears to have been some variation in the standards; the Rūpāvaliya (pp. 112, 113, above), for instance, gives the height of a man or a woman at eight, and an Asura at nine, Devas and Buddhas at ten ‘faces’; but verse 129 gives proportions agreeing with those prescribed in Sāriputra for figures of the Buddha. Slightly conflicting statements are also made by craftsmen; as ten for Buddha, nine for gods, eight for men, and seven for women. One craftsman reversed the last, explaining this on the ground that tall women were considered beautiful.

c18.s2.17

Age of the canon, and the Silpa Śastra generally.

c18.s2.18

Owing to the systematic neglect of the Silpa Śastra by English and Indian Sanskritists, it is difficult to give any adequate account of the technical literature of Art and Craft in India. This literature has evidently remained for centuries in the hands of craftsmen, *Age of the* *Silpa* *Sastras*

c18.s2.19

∗ *Sukranitisāra, or Elements of Polity, by Sukrācarya, ed. Calcutta, 1890.*

c18.s2.20

whose knowledge of Sanskrit as a language was usually slight, though they might know the text by heart. They relied on verbal glosses in the various vernaculars, Tamil and Telugu in Southern India, and Sinhalese in Ceylon. It is quite clear that the Silpa Sastras are the remains of a very carefully systematised and complete literature, and it is certain that it must have taken many centuries for them to have become so imperfect and corrupt as they now are.*

c18.s2.21

The golden age of Indian art and literature, when India was most herself, I conceive to have been the period dating from the time of the Upanishads and of Buddha, to the close of the fifth and sixth centuries A.D. The civilisation of this time found its ultimate literary expression in the Rāmāyaṇa (about 300 A.D.) and the Mahābhārata (about 400 A.D.) and is in popular tradition centred round the court of Vikramāditya (about 375 A.D.).

c18.s2.22

I have above (p. 63) referred to the now isolated occurrence of special crafts widely scattered in different parts of India, and suggested that this indicates a once more universal distribution. The study of Indian Art and Craft, in its profound unity in variety, suggest that its present remains are but the wreck of its former grandeur. If so, this must have been the case long ago, before the Hindu culture of the North was overrun by the Muhammadans. It may well have been in the brilliant era of the Gupta rulers, culminating in the rule of Vikramāditya. The great iron pillar at Delhi, erected in his reign, witnesses still to a knowledge of metallurgy enabling the craftsmen of that day to forge a bar of iron larger than any that could have been forged in Europe before the end of the nineteenth century. It is probable that the Silpa Sastras were written down towards the close of this period. They had previously been handed down orally by the hereditary craftsmen, amongst whom their secret was jealously guarded. The brevity of the memory verses and their technical language further preserved their meaning from common comprehension. The craftsmen are reluctant to allow their technical books to be made public.

c18.s2.23

A certain parallel may be drawn between the writing down of the materials of the Silpa Sastras, and those composing the epics of the Rāmāyaṇa, the Mahābhārata and the Mahāvamsa. There seems to have been a tendency at this time to put into permanent form, traditions and accumulated learning previously transmitted orally in memory verses.† There is a close correspondence between the literary and the artistic phenomena. Compare, for example, a memory verse from the Rūpāvaliya (e.g. vv. 3 and 4, p. 112, above) with this verse from the Dipavamsa, in which the history of the satsatiya is summarised: “The throne, the Animisa (sanctuary), the cloisters, the Gem-court, the Ajapala (tree), and the Mucalinda (tree) together with the Khirapala grove as the seventh.” In both cases we have mnemonic verses of great brevity, evidently adapted to oral transmission; the artist in each case had to fill in the meagre details, and make the story or the image live.

c18.s2.24

† See Geiger, ‘Dipavamsa and Mahavamsa,’ Colombo, 1905, for a study of the evolution of the ‘epic of Duttha Gamani,’ forming the main part of the M. as composed in the fifth century.

c18.s2.25

I suggest a parallel between the literary impulse which led to the writing down of these mnemonic verses, and their expansion into epic history, and the impulse which led to the writing down of the orally transmitted verses of the Silpa Sastra, and suggest also the probability of these two processes having taken place about the same time.

c18.s2.26

∗ “*Mutilated as they invariably are in many important parts, almost every line of them is not only disfigured by gross errors, perpetuated by a succession of ignorant transcribers, but the technical terms and memorial verses with which the whole abounds, are so little understood either by the artists or the pundits of the present day, that it requires no ordinary exertion to comprehend and explain the exact import of even a single section.*”—Ram Raz, ‘Essay on the Architecture of the Hindus,’ London, 1834, p. 3.

c18.s2.27

I select, then, the fifth century as a probable date for the writing down of the Silpa Śastras, *Sāriputra*, but in doing so I do not attach, of course, importance to the exact figures, and do not forget that not all the works in question are necessarily contemporaneous.

c18.s2.28

It is clear that the artistic canons and memory verses cannot have been formulated at a very early period, for they imply the development of Hinduism from and through Buddhism, such as could hardly have taken place before the close of the second century A.D. They appear to be in the main independent of Buddhism; we cannot suppose that any vital tradition, or one concerned mainly with Hindu representations could have been formulated in connection with the weak and effeminate sculpture of the Gandhāra school, perhaps the least truly Indian and artistically least important of all sculpture in India. There is some indication that the Silpa Śastras date from a time subsequent to the decline of Buddhism as the complete expression of national religion. Thus in the Rūpāvaliya, Buddha appears only as one of the ten avatārs of Viṣṇu; and as Rām Rāz suggests,* there is evidence in other works to show that the worship of the Jainas and Buddhists, though still tolerated and provided for in these works, occupied a position of very minor importance in them. But a purely Buddhist canon like the Sāriputra would not have been preserved in India at all after the complete extinction of Buddhism as a separate cult; and it is not likely to have been composed in Ceylon; so that by this line of reasoning also we are thrown back to the fifth century. Another piece of evidence is afforded by the Sukranītisāra,† or ‘Elements of Polity,’ by Sukrācārya, which was translated into Tibetan in the seventh century A.D., and must therefore have been well known in India for a considerable time previous to this. This work contains important chapters on the measurements of images, summarised in part in the table given above, and also a clear presentation of the artistic ideal of the imager himself.

c18.s2.29

† *In this case, and in the Brihad Saṃhita of Varaha Mitra, the artistic canon does not stand by itself as the body of the work, but forms a single chapter or a few chapters in a work of general information, an encyclopædia.*

c18.s2.30

The truth is that we have to consider in all probability two stages in the history of the works in question, the first that of their formulation at the close of the fifth century in Northern India, the second that of their subsequent use and final decay in the North, finally completed as the result of Muhammadan conquest and influence. During this second period, the traditions flourished in the South, where most of the remaining works are still to be found; new recensions of them were made, with additions and interpolations relating to representations of local deities and saints. Artistic expression in the South culminated in the sculpture and architecture of the sixteenth and seventeenth centuries, after which there followed a gradual decline, rapidly accelerated in the nineteenth century, until now there are but few craftsmen who understand their own books, and but few copies of these books procurable, and those imperfect and exceedingly corrupt.

c18.s2.31

∗ ‘*Essay on the Architecture of the Hindus,*’ London, 1834. *This valuable book contains the fullest account in English of the Silpa Śastras (so far as they relate to architecture) and some discussion of their age.*

Linked texts (7)